proxmox-mirrors/ceph
1
0
Fork 0
mirror of https://git.proxmox.com/git/ceph.git synced 2025-08-06 14:14:42 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

update sources to v12.1.0

Browse Source
This commit is contained in:
Fabian Grünbichler 2017-06-27 09:46:02 +02:00
parent 40152f1e46
commit 31f18b776d
2997 changed files with 534452 additions and 48061 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Makefile
View File

@ -1,8 +1,8 @@
RELEASE=5.0
PACKAGE=ceph
VER=12.0.3
DEBREL=pve3
VER=12.1.0
DEBREL=pve1
SRCDIR=ceph
BUILDSRC=${SRCDIR}-${VER}

3
ceph/.gitmodules vendored
View File

@ -55,3 +55,6 @@
[submodule "src/blkin"]
path = src/blkin
url = https://github.com/ceph/blkin
[submodule "src/rapidjson"]
path = src/rapidjson
url = https://github.com/ceph/rapidjson

16
ceph/.mailmap
View File

@ -59,6 +59,8 @@ Chen Dihao <tobeg3oogle@gmail.com>
Chendi Xue <chendi.xue@intel.com>
Chendi Xue <chendi.xue@intel.com> <xuechendi@gmail.com>
Cheng Cheng <ccheng.leo@gmail.com>
Chu Hua-Rong <hrchu@cht.com.tw>
Chu Hua-Rong <hrchu@cht.com.tw> <petertc@gmail.com>
Chris Holcombe <chris.holcombe@nebula.com> <xfactor973@gmail.com>
Christian Brunner <christian@brunner-muc.de> <chb@muc.de>
Christian Marie <pingu@anchor.net.au> <christian@ponies.io>
@ -140,9 +142,11 @@ Haomai Wang <haomai@xsky.com> <haomai@xsky.io>
Haomai Wang <haomai@xsky.com> <yuyuyu101@163.com>
Harry Harrington <git-harry@live.co.uk>
Hazem Amara <hazem.amara@telecom-bretagne.eu> <hazem@hazem-Inspiron-3537.(none)>
He Chuang <hechuang@xsky.com>
Henry C Chang <henry_c_chang@tcloudcomputing.com> <henry.cy.chang@gmail.com>
Hervé Rousseau <hroussea@cern.ch>
Holger Macht <hmacht@suse.de> <holger@homac.de>
Hongtong Liu <hongtong.liu@istuary.com>
Hong Zhangoole <hong.zhangoole@gmail.com>
Huamin Chen <hchen@redhat.com>
Huang Jun <hjwsm1989@gmail.com>
@ -174,6 +178,8 @@ Jenkins <jenkins@ceph.com> <jenkins-build@trusty-small-unique--68a2c286-dc75-466
Jenkins <jenkins@ceph.com> <jenkins-build@trusty-small-unique--e64d6d03-305d-46bd-9a2c-9b546e06937e.localdomain>
Jenkins <jenkins@ceph.com> Jenkins Build Slave User <jenkins-build@trusty-small-unique--a7f82f5f-8832-433e-a632-928924f47e04.localdomain>
Jenkins <jenkins@ceph.com> Jenkins Build Slave User <ceph-release-team@redhat.com>
Jesse Williamson <jesse.williamson@suse.com>
Jesse Williamson <jesse.williamson@suse.com> <jwilliamson@suse.de>
Jiang Heng <jiangheng0511@gmail.com>
Jiantao He <hejiantao5@gmail.com>
Jian Wen <wenjian@letv.com> <wenjianhn@gmail.com>
@ -206,6 +212,7 @@ Jonas Keidel <jonas@jonas-keidel.de>
Jordan Dorne <jordan.dorne@gmail.com>
Josh Durgin <jdurgin@redhat.com>
Josh Durgin <jdurgin@redhat.com> <joshd@redhat.com>
Josh Durgin <jdurgin@redhat.com> <jduring@redhat.com>
Josh Durgin <josh.durgin@inktank.com> <joshd@hq.newdream.net>
Josh Durgin <josh.durgin@inktank.com> <josh.durgin@dreamhost.com>
Kacper Kowalik <xarthisius@gentoo.org>
@ -218,11 +225,14 @@ Kiseleva Alyona <akiselyova@mirantis.com>
Kongming Wu <wu.kongming@h3c.com>
Lan De <lan.de3@zte.com.cn>
Laszlo Boszormenyi <gcs@debian.hu>
Leo Zhang <nguzcf@gmail.com>
Luo Kexue <luo.kexue@zte.com.cn>
Luo Runbing <runsisi@hust.edu.cn>
Luo Runbing <runsisi@zte.com.cn>
Li Hongjie <lihongjie@cmss.chinamobile.com>
Li Wang <li.wang@kylin-cloud.com> <liwang@ubuntukylin.com>
Linfei Hou <hou.linfei@h3c.com>
Liu Hong <liuhong@cmss.chinamobile.com>
Liu Peiyan <liu.peiyang@h3c.com>
Li Tianqing <tianqing@unitedstack.com>
Liu Yang <yippeetry@gmail.com>
@ -353,7 +363,8 @@ Shehbaz Jaffer <shehbaz.jaffer@mail.utoronto.ca> <shehbazjaffer007@gmail.com>
Shinobu Kinjo <shinobu@redhat.com> <shinobu@linux.com>
Shinobu Kinjo <shinobu@redhat.com> <ceph@ceph-stack.redhat.com>
Shishir Gowda <shishir.gowda@sandisk.com>
shiqi <m13913886148@gmail.com> <1454927420@qq.com>
Shiqi <m13913886148@gmail.com>
Shiqi <m13913886148@gmail.com> <1454927420@qq.com>
Shu, Xinxin <xinxin.shu@intel.com>
Shun Song <song.shun3@zte.com.cn>
Shun Song <song.shun3@zte.com.cn> <songshun134@126.com>
@ -387,6 +398,7 @@ Tobias Suckow <tobias@suckow.biz>
Tommi Virtanen <tv@inktank.com> <tommi.virtanen@dreamhost.com>
Tommi Virtanen <tv@inktank.com> <tv@eagain.net>
Tommi Virtanen <tv@inktank.com> <tv@hq.newdream.net>
Tone Zhang <tone.zhang@linaro.org>
Travis Rhoden <trhoden@redhat.com> <trhoden@gmail.com>
Tyler Brekke <tyler.brekke@inktank.com>
Uday Mullangi <udaymjl@gmail.com>
@ -434,6 +446,7 @@ Yannick Atchy Dalama <yannick.atchy.dalama@gmail.com>
Yankun Li <liyankun@unitedstack.com> <YankunLi@users.noreply.github.com>
Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng <zheng.z.yan@intel.com> <ukernel@gmail.com>
Yanchun Bai <yanchun.bai@istuary.com>
Yehua Chen <chen.yehua@h3c.com>
Yehua Chen <chen.yehua@h3c.com> <root@ubuntu1.com>
Yehuda Sadeh <yehuda@inktank.com> <yehdua@inktank.com>
@ -452,6 +465,7 @@ Yehuda Sadeh <ysadehwe@redhat.com> <yehuda@redhat.com.>
Yongqiang He <he.yongqiang@h3c.com>
Yongyue Sun <abioy.sun@gmail.com>
You Ji <youji@ebay.com>
Yunfei Guan <yunfei.guan@xtaotech.com>
Yuan Zhou <yuan.zhou@intel.com> <dunk007@gmail.com>
Yunchuan Wen <yunchuan.wen@kylin-cloud.com> <yunchuanwen@ubuntukylin.com>
Yuri Weinstein <yuriw@redhat.com> <yweinste@redhat.com>

21
ceph/.organizationmap
View File

@ -23,6 +23,7 @@
# 9 27 TCloud Computing <contact@tcloudcomputing.com>
# 10 22 GNU <contact@gnu.org>
#
99Cloud Inc <contact@99cloud.net> Yu Shengzuo <yu.shengzuo@99cloud.net>
Acaleph <contact@acale.ph> Alistair Israel <aisrael@gmail.com>
Alcatel Lucent <contact@alcatel-lucent.com> Joseph McDonald <joseph.mcdonald@alcatel-lucent.com>
Alcatel Lucent <contact@alcatel-lucent.com> Ker Liu <ker.liu@alcatel-lucent.com>
@ -66,14 +67,15 @@ China Mobile <contact@chinamobile.com> donglinpeng <donglinpeng@cmss.chinamobile
China Mobile <contact@chinamobile.com> Gui Hecheng <guihecheng@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> Guo Zhandong <guozhandong@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> Jing Wenjun <jingwenjun@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> lihongjie <lihongjie@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> Li Hongjie <lihongjie@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> Liu Hong <liuhong@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> lvshuhua <lvshuhua@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
China Mobile <contact@chinamobile.com> wangzhengyong <wangzhengyong@cmss.chinamobile.com>
Chinac <info@chinac.com> Huan Zhang <zhanghuan@chinac.com>
Chinac <info@chinac.com> Xinxin Shu <shuxinxin@chinac.com>
Choopa, LLC <contact@choopa.com> Adam Twardowski <adam.twardowski@gmail.com>
Chunghwa Telecom <contact@cht.com.tw> hrchu <hrchu@cht.com.tw>
Chunghwa Telecom <contact@cht.com.tw> Chu Hua-Rong <hrchu@cht.com.tw>
CISCO <contact@cisco.com> Johnu George <johnugeo@cisco.com>
CISCO <contact@cisco.com> Kai Zhang <zakir.exe@gmail.com>
CISCO <contact@cisco.com> Roland Mechler <rmechler@cisco.com>
@ -206,6 +208,7 @@ Hostplex Hosting <contact@hostplex.net> Andras Elso <elso.andras@gmail.com>
HP <contact@hp.com> Blaine Gardner <blaine.gardner@hp.com>
HP <contact@hp.com> Joe Handzik <joseph.t.handzik@hp.com>
Huazhong University of Science and Technology <contact@hust.edu.cn> Luo Runbing <runsisi@hust.edu.cn>
IBM <contact@IBM.com> Andrew Solomon <asolomon@us.ibm.com>
IBM <contact@IBM.com> Michel Normand <normand@linux.vnet.ibm.com>
IBM <contact@IBM.com> Samuel Matzek <smatzek@us.ibm.com>
ICT <contact@ict.ac.cn> Zhang Huan <zhanghuan@ict.ac.cn>
@ -214,6 +217,7 @@ Igalia <contact@igalia.com> Javier M. Mellid <jmunhoz@igalia.com>
Igalia <contact@igalia.com> Juan A. Suarez Romero <jasuarez@igalia.com>
Imagination Technologies Ltd. <contact@imgtec.com> Alistair Strachan <alistair.strachan@imgtec.com>
iNic <contact@inic.no> Bjørnar Ness <bjornar.ness@gmail.com>
Indraprastha Institute of Information Technology, Delhi <contact@iiitd.ac.in> Vedant Nanda <vedant15114@iiitd.ac.in>
Indiana University <contact@indiana.edu> Jayashree Candadai <jayaajay@indiana.edu>
Inktank <contact@inktank.com> Alexandre Marangone <alexandre.marangone@inktank.com>
Inktank <contact@inktank.com> Alex Elder <elder@inktank.com>
@ -278,6 +282,7 @@ Iron Systems Inc. <info@ironsystems.com> Harpreet Dhillon <harpreet@ironsystems.
Istuary Innovation Group <contact@istuary.com> Hongtong Liu <hongtong.liu@istuary.com>
Istuary Innovation Group <contact@istuary.com> Pan Liu <pan.liu@istuary.com>
Istuary Innovation Group <contact@istuary.com> Tang Jin <tang.jin@istuary.com>
Istuary Innovation Group <contact@istuary.com> Yanchun Bai <yanchun.bai@istuary.com>
IT Refined <contact@itrefined.com> Ron Allred <rallred@itrefined.com>
IWeb <contact@iweb.com> David Moreau Simard <dmsimard@iweb.com>
Johannes Gutenberg-Universität Mainz <contact@uni-mainz.de> Marcel Lauhoff <lauhoff@uni-mainz.de>
@ -296,6 +301,7 @@ LETV <contact@letv.com> Ji Chen <insomnia@139.com>
LETV <contact@letv.com> Wu Xingyi <wuxingyi@letv.com>
Linaro <contact@linaro.org> Steve Capper <steve.capper@linaro.org>
Linaro <contact@linaro.org> wei xiao <wei.xiao@linaro.org>
Linaro <contact@linaro.org> Tone Zhang <tone.zhang@linaro.org>
Linaro <contact@linaro.org> Yazen Ghannam <yazen.ghannam@linaro.org>
Linaro <contact@linaro.org> Yibo Cai <yibo.cai@linaro.org>
Los Alamos National Laboratory <contact@lanl.gov> Esteban Molina-Estolano <eestolan@lanl.gov>
@ -339,6 +345,7 @@ Pacific Northwest National Laboratory <contact@pnl.gov> Evan Felix <evan.felix@p
Pacific Northwest National Laboratory <contact@pnl.gov> Scott Devoid <devoid@anl.gov>
Piston Cloud Computing <info@pistoncloud.com> Mike Lundy <mike@fluffypenguin.org>
Pogoapp <contact@pogoapp.com> Paul Meserve <paul@pogodan.com>
Proxmox Server Solutions GmbH <office@proxmox.com> Fabian Grünbichler <f.gruenbichler@proxmox.com>
Purdue University <contact@purdue.edu> Mike Shuey <shuey@purdue.edu>
Quadrature Capital Limited <info@quadraturecapital.com> Jim Wright <jim@quadraturecapital.com>
Quantum Corporation <info@quantum.com> Bassam Tabbara <bassam.tabbara@quantum.com>
@ -481,14 +488,17 @@ SUSE <contact@suse.com> João Eduardo Luís <joao@suse.de>
SUSE <contact@suse.com> Karl Eichwalder <ke@suse.de>
SUSE <contact@suse.com> Karol Mroz <kmroz@suse.com>
SUSE <contact@suse.com> Kapil Sharma <ksharma@suse.com>
SUSE <contact@suse.com> Mohamad Gebai <mgebai@suse.com>
SUSE <contact@suse.com> Michal Koutný <mkoutny@suse.com>
SUSE <contact@suse.com> Nathan Cutler <ncutler@suse.com>
SUSE <contact@suse.com> Owen Synge <osynge@suse.com>
SUSE <contact@suse.com> Ricardo Dias <rdias@suse.com>
SUSE <contact@suse.com> Sven Seeberg <sseebergelverfeldt@suse.com>
SUSE <contact@suse.com> Thorsten Behrens <tbehrens@suse.com>
SUSE <contact@suse.com> Tim Serong <tserong@suse.com>
SWITCH <contact@switch.ch> Jens-Christian Fischer <jens-christian.fischer@switch.ch>
SWITCH <contact@switch.ch> Simon Leinen <simon.leinen@switch.ch>
T2Cloud <contact@t2cloud.net> Leo Zhang <nguzcf@gmail.com>
TCloud Computing <contact@tcloudcomputing.com> CC Lien <cc_lien@tcloudcomputing.com>
TCloud Computing <contact@tcloudcomputing.com> Henry C Chang <henry_c_chang@tcloudcomputing.com>
TCloud Computing <contact@tcloudcomputing.com> Herb Shiu <herb_shiu@tcloudcomputing.com>
@ -509,6 +519,7 @@ The University of Arizona <contact@arizona.edu> James Ryan Cresawn <jrcresawn@gm
Time Warner Cable Inc. <contact@twcable.com> Bryan Stillwell <bryan.stillwell@twcable.com>
Trendy Tech <contact@trendytech.com.cn> shiqi <m13913886148@gmail.com>
Trendy Tech <contact@trendytech.com.cn> Lei Zhang <243290414@qq.com>
Uber Technologies Inc. <contact@uber.com> Henrik Korkuc <henrik@uber.com>
Ubuntu Kylin <contact@ubuntukylin.com> Min Chen <minchen@ubuntukylin.com>
UMCloud <contact@umcloud.com> Jiaying Ren <mikulely@gmail.com>
UMCloud <contact@umcloud.com> Rongze Zhu <zrzhit@gmail.com>
@ -624,7 +635,6 @@ Unaffiliated <no@organization.net> koleosfuscus <koleosfuscus@yahoo.com>
Unaffiliated <no@organization.net> Kyr Shatskyy <kyrylo.shatskyy@gmail.com>
Unaffiliated <no@organization.net> Laurent Guerby <laurent@guerby.net>
Unaffiliated <no@organization.net> Lee Revell <rlrevell@gmail.com>
Unaffiliated <no@organization.net> Leo Zhang <nguzcf@gmail.com>
Unaffiliated <no@organization.net> Lucas Fantinel <lucas.fantinel@gmail.com>
Unaffiliated <no@organization.net> Коренберг Марк <socketpair@gmail.com>
Unaffiliated <no@organization.net> Matt Richards <mattjrichards@gmail.com>
@ -644,6 +654,7 @@ Unaffiliated <no@organization.net> oddomatik <bandrus+github@gmail.com>
Unaffiliated <no@organization.net> Oleh Prypin <oleh@pryp.in>
Unaffiliated <no@organization.net> optimistyzy <optimistyzy@gmail.com>
Unaffiliated <no@organization.net> Pascal Bach <pasci.bach@gmail.com>
Unaffiliated <no@organization.net> Patrick Dinnen <pdinnen@gmail.com>
Unaffiliated <no@organization.net> Patrick Donnelly <batrick@batbytes.com>
Unaffiliated <no@organization.net> Peter Maloney <peter.maloney@yahoo.ca>
Unaffiliated <no@organization.net> Petertc Chu <petertc.chu@gmail.com>
@ -699,6 +710,7 @@ Unaffiliated <no@organization.net> Xu Biao <xubiao.codeyz@gmail.com>
Unaffiliated <no@organization.net> Yang Honggang <joseph.yang@xtaotech.com>
Unaffiliated <no@organization.net> Yann Dupont <yann@objoo.org>
Unaffiliated <no@organization.net> Yannick Atchy Dalama <yannick.atchy.dalama@gmail.com>
Unaffiliated <no@organization.net> Yao Zongyou <yaozongyou@vip.qq.com>
Unaffiliated <no@organization.net> Yatin Kumbhare <yatinkumbhare@gmail.com>
Unaffiliated <no@organization.net> YiQiang Chen <cyqsign@163.com>
Unaffiliated <no@organization.net> Yongyue Sun <abioy.sun@gmail.com>
@ -735,6 +747,7 @@ University of Maryland <contact@umd.edu> Liam Monahan <liam@umiacs.umd.edu>
University of Maryland <contact@umd.edu> Padraig O'Sullivan <posulliv@umd.edu>
University of Toronto <contact@mail.utoronto.ca> Shehbaz Jaffer <shehbaz.jaffer@mail.utoronto.ca>
University of Utah <contact@utah.edu> Xing Lin <xinglin@cs.utah.edu>
Virtuozzo <contact@virtuozzo.com> Andrey Parfenov <aparfenov@virtuozzo.com>
VRT <contact@vrt.com.au> Stuart Longland <stuartl@vrt.com.au>
Walmart Labs <contact@walmartlabs.com> Pavan Rallabhandi <PRallabhandi@walmartlabs.com>
Walmart Labs <contact@walmartlabs.com> Sirisha Guduru <SGuduru@walmartlabs.com>
@ -748,11 +761,13 @@ X-ION <contact@x-ion.de> Mouad Benchchaoui <m.benchchaoui@x-ion.de>
X-ION <contact@x-ion.de> Stephan Renatus <s.renatus@x-ion.de>
XSky <contact@xsky.com> hechuang <hechuang@xsky.com>
XSky <contact@xsky.com> Haomai Wang <haomai@xsky.com>
XSky <contact@xsky.com> He Chuang <hechuang@xsky.com>
XSky <contact@xsky.com> Min Chen <chenmin@xsky.com>
XSky <contact@xsky.com> Mingxin Liu <mingxin@xsky.com>
XSky <contact@xsky.com> Tianshan Qu <tianshan@xsky.com>
XSky <contact@xsky.com> Xinze Chi <xinze@xsky.com>
XSky <contact@xsky.com> Zhiqiang Wang <zhiqiang@xsky.com>
XTao Co. Ltd. <contact@xtaotech.com> Yunfei Guan <yunfei.guan@xtaotech.com>
Yahoo! <contact@yahoo-inc.com> Guang Yang <yguang@yahoo-inc.com>
Yahoo! <contact@yahoo-inc.com> Haifeng Liu <haifeng@yahoo-inc.com>
Yahoo! <contact@yahoo-inc.com> Lei Dong <leidong@yahoo-inc.com>

15
ceph/AUTHORS
View File

@ -1,17 +1,14 @@
Maintainer
----------
Sage Weil <sage@redhat.com>
Component Technical Leads
-------------------------
Core, RADOS - Josh Durgin <jdurgin@redhat.com>
RBD - Jason Dillaman <dillaman@redhat.com>
RBD (kernel) - Ilya Dryomov <idryomov@redhat.com>
RGW - Yehuda Sadeh <yehuda@redhat.com>
Matt Benjamin <mbenjamin@redhat.com>
Matt Benjamin <mbenjami@redhat.com>
CephFS - John Spray <jspray@redhat.com>
CephFS (kernel) - Yan, Zheng <zyan@redhat.com>
Deployment - Alfredo Deza <adeza@redhat.com>
@ -19,10 +16,18 @@ Teuthology - Zack Cerza <zack@redhat.com>
Calamari - Gregory Meno <gmeno@redhat.com>
Chef Cookbook - Guilhem Lettron <guilhem@lettron.fr>
Release Manager
---------------
Abhishek Lekshmanan <abhishek@suse.com>
Backport Team
-------------
Nathan Cutler <ncutler@suse.cz>
Shinobu Kinjo <shinobu@redhat.com>
Abhishek Lekshmanan <abhishek@suse.com>
Contributors
------------
For a complete contributor list:
git shortlog -sn

121
ceph/CMakeLists.txt
View File

@ -1,7 +1,7 @@
cmake_minimum_required(VERSION 2.8.11)
project(ceph)
set(VERSION 12.0.3)
set(VERSION 12.1.0)
if(POLICY CMP0046)
# Tweak policies (this one disables "missing" dependency warning)
@ -22,7 +22,15 @@ endif()
if(POLICY CMP0065)
cmake_policy(SET CMP0065 NEW)
endif()
if(POLICY CMP0051)
# cmake 3.1 and higher include generator expressions in SOURCES property.
# in BuildBoost.cmake, get_target_property(<var> <target> SOURCES) is used
# to retrieve the source files of a target. in that case, we are only
# interested in the *source* files. and i don't want to bother stripping off
# the TARGET_OBJECTS elements from the returned SOURCES. so let's stick with
# the old behavior now.
cmake_policy(SET CMP0051 OLD)
endif()
list(APPEND CMAKE_MODULE_PATH "${CMAKE_SOURCE_DIR}/cmake/modules/")
if(CMAKE_SYSTEM_NAME MATCHES "Linux")
@ -105,11 +113,10 @@ CHECK_FUNCTION_EXISTS(pthread_set_name_np HAVE_PTHREAD_SET_NAME_NP)
CHECK_FUNCTION_EXISTS(pthread_setname_np HAVE_PTHREAD_SETNAME_NP)
CHECK_FUNCTION_EXISTS(pthread_getname_np HAVE_PTHREAD_GETNAME_NP)
CHECK_FUNCTION_EXISTS(eventfd HAVE_EVENTFD)
CHECK_FUNCTION_EXISTS(getprogname HAVE_GETPROGNAME)
CHECK_INCLUDE_FILES("inttypes.h" HAVE_INTTYPES_H)
CHECK_INCLUDE_FILES("linux/types.h" HAVE_LINUX_TYPES_H)
CHECK_INCLUDE_FILES("linux/version.h" HAVE_LINUX_VERSION_H)
CHECK_INCLUDE_FILES("stdint.h" HAVE_STDINT_H)
CHECK_INCLUDE_FILES("arpa/nameser_compat.h" HAVE_ARPA_NAMESER_COMPAT_H)
CHECK_INCLUDE_FILES("sys/mount.h" HAVE_SYS_MOUNT_H)
CHECK_INCLUDE_FILES("sys/param.h" HAVE_SYS_PARAM_H)
@ -218,6 +225,12 @@ if(WITH_SPDK)
set(HAVE_SPDK TRUE)
endif(WITH_SPDK)
option(WITH_PMEM "Enable PMEM" OFF)
if(WITH_PMEM)
find_package(pmem REQUIRED)
set(HAVE_PMEM ${PMEM_FOUND})
endif(WITH_PMEM)
# needs mds and? XXX
option(WITH_LIBCEPHFS "libcephfs client library" ON)
@ -251,14 +264,14 @@ if(WITH_LEVELDB)
find_file(HAVE_LEVELDB_FILTER_POLICY leveldb/filter_policy.h PATHS ${LEVELDB_INCLUDE_DIR})
endif(WITH_LEVELDB)
find_package(atomic_ops REQUIRED)
message(STATUS "${ATOMIC_OPS_LIBRARIES}")
if(NOT ${ATOMIC_OPS_FOUND})
set(NO_ATOMIC_OPS 1)
endif(NOT ${ATOMIC_OPS_FOUND})
find_package(snappy REQUIRED)
option(WITH_LZ4 "LZ4 compression support" OFF)
if(WITH_LZ4)
find_package(LZ4 REQUIRED)
set(HAVE_LZ4 ${LZ4_FOUND})
endif(WITH_LZ4)
#if allocator is set on command line make sure it matches below strings
if(ALLOCATOR)
if(${ALLOCATOR} MATCHES "tcmalloc(_minimal)?")
@ -345,7 +358,7 @@ endif(WITH_BLKIN)
#option for RGW
option(WITH_RADOSGW "Rados Gateway is enabled" ON)
option(WITH_RADOSGW_FCGI_FRONTEND "Rados Gateway's FCGI frontend is enabled" ON)
option(WITH_RADOSGW_FCGI_FRONTEND "Rados Gateway's FCGI frontend is enabled" OFF)
option(WITH_RADOSGW_BEAST_FRONTEND "Rados Gateway's Beast frontend is enabled" ON)
if(WITH_RADOSGW)
find_package(EXPAT REQUIRED)
@ -437,6 +450,8 @@ if(${WITH_LTTNG})
endif()
endif(${WITH_LTTNG})
option(WITH_OSD_INSTRUMENT_FUNCTIONS OFF)
#option for Babeltrace
option(HAVE_BABELTRACE "Babeltrace libraries are enabled" ON)
if(${HAVE_BABELTRACE})
@ -513,90 +528,42 @@ endif()
# Boost
option(WITH_SYSTEM_BOOST "require and build with system Boost" OFF)
# Boost::thread depends on Boost::atomic, so list it explicitly.
set(BOOST_COMPONENTS
container thread system regex random program_options date_time iostreams coroutine context)
atomic thread system regex random program_options date_time iostreams
filesystem)
set(BOOST_HEADER_COMPONENTS container)
if(WITH_MGR)
list(APPEND BOOST_COMPONENTS python)
endif()
if (WITH_SYSTEM_BOOST)
if(WITH_RADOSGW_BEAST_FRONTEND)
list(APPEND BOOST_COMPONENTS coroutine context)
endif()
set(Boost_USE_MULTITHREADED ON)
# require minimally the bundled version
if(WITH_SYSTEM_BOOST)
if(ENABLE_SHARED)
set(Boost_USE_STATIC_LIBS OFF)
else()
set(Boost_USE_STATIC_LIBS ON)
endif()
find_package(Boost 1.61 COMPONENTS ${BOOST_COMPONENTS} REQUIRED)
else()
set(BOOST_CFLAGS "-fPIC -w") # check on arm, etc <---XXX
set(BOOST_J 1 CACHE STRING
"max jobs for Boost build") # override w/-DBOOST_J=<n>
message(STATUS "BUILDING Boost Libraries at j ${BOOST_J}")
# 1. prep w/required components
set(BOOST_SOURCE_DIR "${PROJECT_SOURCE_DIR}/src/boost")
set(BOOST_PREFIX "${PROJECT_BINARY_DIR}/boost")
set(BOOST_BUILD "${PROJECT_BINARY_DIR}/boost-build")
string(REPLACE ";" "," BOOST_WITH_LIBS "${BOOST_COMPONENTS}")
execute_process(COMMAND "./bootstrap.sh"
"--prefix=${BOOST_PREFIX}"
"--with-libraries=${BOOST_WITH_LIBS}"
WORKING_DIRECTORY ${BOOST_SOURCE_DIR})
set(BOOST_ROOT "${BOOST_PREFIX}")
set(b2 ./b2
--build-dir=${BOOST_BUILD} -j${BOOST_J})
if(CMAKE_VERBOSE_MAKEFILE)
list(APPEND b2 -d1)
else()
list(APPEND b2 -d0)
endif()
list(APPEND b2
variant=release link=static threading=multi cxxflags=${BOOST_CFLAGS})
if(NOT CMAKE_HOST_SYSTEM_PROCESSOR STREQUAL CMAKE_SYSTEM_PROCESSOR)
# we are crosscompiling
if(CMAKE_CXX_COMPILER_ID STREQUAL GNU)
set(b2_cc gcc)
elseif(CMAKE_CXX_COMPILER_ID STREQUAL Clang)
set(b2_cc clang)
else()
message(SEND_ERROR "unknown compiler: ${CMAKE_CXX_COMPILER_ID}")
endif()
# edit the config.jam so, b2 will be able to use the specified toolset
execute_process(
COMMAND
sed -i
"s|using ${b2_cc} ;|using ${b2_cc} : ${CMAKE_SYSTEM_PROCESSOR} : ${CMAKE_CXX_COMPILER} ;|"
${PROJECT_SOURCE_DIR}/src/boost/project-config.jam)
# use ${CMAKE_SYSTEM_PROCESSOR} as the version identifier of compiler
list(APPEND b2 toolset=${b2_cc}-${CMAKE_SYSTEM_PROCESSOR})
endif()
# 2. install headers
execute_process(COMMAND
${b2}
headers
WORKING_DIRECTORY ${BOOST_SOURCE_DIR})
# 3. build and install libs
execute_process(COMMAND
${b2}
install
WORKING_DIRECTORY ${BOOST_SOURCE_DIR})
# 4. set hints for FindBoost.cmake
set(Boost_USE_STATIC_LIBS ON)
set(Boost_NO_SYSTEM_PATHS ON)
include_directories(BEFORE ${BOOST_PREFIX}/include)
# fixup for CheckIncludeFileCXX
set(HAVE_BOOST_ASIO_COROUTINE ON)
set(BOOST_ROOT ${BOOST_PREFIX})
set(Boost_NO_SYSTEM_PATHS ON)
include(BuildBoost)
build_boost(1.63
COMPONENTS ${BOOST_COMPONENTS} ${BOOST_HEADER_COMPONENTS})
include_directories(BEFORE SYSTEM ${Boost_INCLUDE_DIRS})
endif()
set(Boost_USE_MULTITHREADED ON)
# require minimally the bundled version
find_package(Boost 1.61 COMPONENTS ${BOOST_COMPONENTS} REQUIRED)
include_directories(SYSTEM ${Boost_INCLUDE_DIRS})
include_directories(SYSTEM ${PROJECT_BINARY_DIR}/include)
CHECK_INCLUDE_FILE_CXX("boost/asio/coroutine.hpp" HAVE_BOOST_ASIO_COROUTINE)
find_package(Threads REQUIRED)
option(WITH_SELINUX "build SELinux policy" OFF)
@ -636,5 +603,5 @@ add_tags(ctags
SRC_DIR src
TAG_FILE tags
EXCLUDE_OPTS ${CTAG_EXCLUDES}
EXCLUDES "*.js")
EXCLUDES "*.js" "*.css")
add_custom_target(tags DEPENDS ctags)

4
ceph/COPYING
View File

@ -30,10 +30,6 @@ Files: src/common/bloom_filter.hpp
Copyright: Copyright (C) 2000 Arash Partow <arash@partow.net>
License: Boost Software License, Version 1.0
Files: m4/acx_pthread.m4
Copyright: Steven G. Johnson <stevenj@alum.mit.edu>
License: GPLWithACException
Files: src/common/crc32c_intel*:
Copyright:
Copyright 2012-2013 Intel Corporation All Rights Reserved.

121
ceph/PendingReleaseNotes
View File

@ -1,36 +1,7 @@
>= 12.0.0
------
* The original librados rados_objects_list_open (C) and objects_begin
(C++) object listing API, deprecated in Hammer, has finally been
removed. Users of this interface must update their software to use
either the rados_nobjects_list_open (C) and nobjects_begin (C++) API or
the new rados_object_list_begin (C) and object_list_begin (C++) API
before updating the client-side librados library to Luminous.
Object enumeration (via any API) with the latest librados version
and pre-Hammer OSDs is no longer supported. Note that no in-tree
Ceph services rely on object enumeration via the deprecated APIs, so
only external librados users might be affected.
The newest (and recommended) rados_object_list_begin (C) and
object_list_begin (C++) API is only usable on clusters with the
SORTBITWISE flag enabled (Jewel and later). (Note that this flag is
required to be set before upgrading beyond Jewel.)
* CephFS clients without the 'p' flag in their authentication capability
string will no longer be able to set quotas or any layout fields. This
flag previously only restricted modification of the pool and namespace
fields in layouts.
* CephFS directory fragmentation (large directory support) is enabled
by default on new filesystems. To enable it on existing filesystems
use "ceph fs set <fs_name> allow_dirfrags".
* CephFS will generate a health warning if you have fewer standby daemons
than it thinks you wanted. By default this will be 1 if you ever had
a standby, and 0 if you did not. You can customize this using
``ceph fs set <fs> standby_count_wanted <number>``. Setting it
to zero will effectively disable the health check.
* The "ceph mds tell ..." command has been removed. It is superceded
by "ceph tell mds.<id> ..."
* The "journaler allow split entries" config setting has been removed.
* The 'apply' mode of cephfs-journal-tool has been removed
12.0.0
------
@ -121,3 +92,93 @@
* The 'rados df' JSON output now prints numeric values as numbers instead of
strings.
* There was a bug introduced in Jewel (#19119) that broke the mapping behavior
when an "out" OSD that still existed in the CRUSH map was removed with 'osd rm'.
This could result in 'misdirected op' and other errors. The bug is now fixed,
but the fix itself introduces the same risk because the behavior may vary between
clients and OSDs. To avoid problems, please ensure that all OSDs are removed
from the CRUSH map before deleting them. That is, be sure to do::
ceph osd crush rm osd.123
before::
ceph osd rm osd.123
12.0.2
------
* The original librados rados_objects_list_open (C) and objects_begin
(C++) object listing API, deprecated in Hammer, has finally been
removed. Users of this interface must update their software to use
either the rados_nobjects_list_open (C) and nobjects_begin (C++) API or
the new rados_object_list_begin (C) and object_list_begin (C++) API
before updating the client-side librados library to Luminous.
Object enumeration (via any API) with the latest librados version
and pre-Hammer OSDs is no longer supported. Note that no in-tree
Ceph services rely on object enumeration via the deprecated APIs, so
only external librados users might be affected.
The newest (and recommended) rados_object_list_begin (C) and
object_list_begin (C++) API is only usable on clusters with the
SORTBITWISE flag enabled (Jewel and later). (Note that this flag is
required to be set before upgrading beyond Jewel.)
* CephFS clients without the 'p' flag in their authentication capability
string will no longer be able to set quotas or any layout fields. This
flag previously only restricted modification of the pool and namespace
fields in layouts.
* CephFS directory fragmentation (large directory support) is enabled
by default on new filesystems. To enable it on existing filesystems
use "ceph fs set <fs_name> allow_dirfrags".
* CephFS will generate a health warning if you have fewer standby daemons
than it thinks you wanted. By default this will be 1 if you ever had
a standby, and 0 if you did not. You can customize this using
``ceph fs set <fs> standby_count_wanted <number>``. Setting it
to zero will effectively disable the health check.
* The "ceph mds tell ..." command has been removed. It is superceded
by "ceph tell mds.<id> ..."
12.1.0
------
* The ``mon_osd_max_op_age`` option has been renamed to
``mon_osd_warn_op_age`` (default: 32 seconds), to indicate we
generate a warning at this age. There is also a new
``mon_osd_err_op_age_ratio`` that is a expressed as a multitple of
``mon_osd_warn_op_age`` (default: 128, for roughly 60 minutes) to
control when an error is generated.
* The default maximum size for a single RADOS object has been reduced from
100GB to 128MB. The 100GB limit was completely impractical in practice
while the 128MB limit is a bit high but not unreasonable. If you have an
application written directly to librados that is using objects larger than
128MB you may need to adjust ``osd_max_object_size``.
* The semantics of the 'rados ls' and librados object listing
operations have always been a bit confusing in that "whiteout"
objects (which logically don't exist and will return ENOENT if you
try to access them) are included in the results. Previously
whiteouts only occurred in cache tier pools. In luminous, logically
deleted but snapshotted objects now result in a whiteout object, and
as a result they will appear in 'rados ls' results, even though
trying to read such an object will result in ENOENT. The 'rados
listsnaps' operation can be used in such a case to enumerate which
snapshots are present.
This may seem a bit strange, but is less strange than having a
deleted-but-snapshotted object not appear at all and be completely
hidden from librados's ability to enumerate objects. Future
versions of Ceph will likely include an alternative object
enumeration interface that makes it more natural and efficient to
enumerate all objects along with their snapshot and clone metadata.
* The deprecated 'crush_ruleset' property has finally been removed; please use
'crush_rule' instead for the 'osd pool get ...' and 'osd pool set ..' commands.
* The 'osd pool default crush replicated ruleset' option has been
removed and replaced by the 'osd pool default crush rule' option.
By default it is -1, which means the mon will pick the first type
replicated rule in the CRUSH map for replicated pools. Erasure
coded pools have rules that are automatically created for them if they are
not specified at pool creation time.

13
ceph/alpine/APKBUILD
View File

@ -1,7 +1,7 @@
# Contributor: John Coyle <dx9err@gmail.com>
# Maintainer: John Coyle <dx9err@gmail.com>
pkgname=ceph
pkgver=12.0.3
pkgver=12.1.0
pkgrel=0
pkgdesc="Ceph is a distributed object store and file system"
pkgusers="ceph"
@ -38,7 +38,6 @@ makedepends="
keyutils-dev
leveldb-dev
libaio-dev
libatomic_ops-dev
libedit-dev
libressl-dev
libtirpc-dev
@ -64,7 +63,7 @@ makedepends="
xmlstarlet
yasm
"
source="ceph-12.0.3.tar.bz2"
source="ceph-12.1.0.tar.bz2"
subpackages="
$pkgname-base
$pkgname-common
@ -117,7 +116,7 @@ _sysconfdir=/etc
_udevrulesdir=/etc/udev/rules.d
_python_sitelib=/usr/lib/python2.7/site-packages
builddir=$srcdir/ceph-12.0.3
builddir=$srcdir/ceph-12.1.0
build() {
export CEPH_BUILD_VIRTUALENV=$builddir
@ -167,7 +166,6 @@ package() {
|| return 1
chmod 0644 $pkgdir$_docdir/ceph/sample.ceph.conf || return 1
chmod 0644 $pkgdir$_docdir/ceph/sample.fetch_config || return 1
# udev rules
install -m 0644 -D udev/50-rbd.rules $pkgdir$_udevrulesdir/50-rbd.rules || return 1
@ -193,7 +191,6 @@ base() {
xfsprogs
"
_pkg $_docdir/ceph sample.ceph.conf sample.fetch_config
_pkg $_bindir crushtool monmaptool osdmaptool ceph-run ceph-detect-init
_pkg $_sbindir ceph-create-keys mount.ceph
_pkg $_libexecdir/ceph ceph_common.sh
@ -292,7 +289,7 @@ radosgw() {
pkgdesc="Rados REST gateway which implements Amazon's S3 and OpenStack's Swift APIs."
depends="ceph-common"
_pkg $_bindir radosgw radosgw-admin radosgw-token radosgw-object-expirer
_pkg $_bindir radosgw radosgw-admin radosgw-token radosgw-es radosgw-object-expirer
mkdir -p $subpkgdir$_localstatedir/lib/ceph/radosgw
}
@ -301,7 +298,7 @@ osd() {
depends="ceph-base parted gptfdisk"
_pkg $_bindir ceph-clsinfo ceph-bluefs-tool ceph-objectstore-tool ceph-osd
_pkg $_sbindir ceph-disk ceph-disk-udev
_pkg $_sbindir ceph-disk
_pkg $_libexecdir/ceph ceph-osd-prestart.sh
_pkg $_udevrulesdir 60-ceph-by-parttypeuuid.rules 95-ceph-osd.rules
install -m 750 -o $_ceph_uid -g $_ceph_gid -d \

7
ceph/alpine/APKBUILD.in
View File

@ -38,7 +38,6 @@ makedepends="
keyutils-dev
leveldb-dev
libaio-dev
libatomic_ops-dev
libedit-dev
libressl-dev
libtirpc-dev
@ -167,7 +166,6 @@ package() {
|| return 1
chmod 0644 $pkgdir$_docdir/ceph/sample.ceph.conf || return 1
chmod 0644 $pkgdir$_docdir/ceph/sample.fetch_config || return 1
# udev rules
install -m 0644 -D udev/50-rbd.rules $pkgdir$_udevrulesdir/50-rbd.rules || return 1
@ -193,7 +191,6 @@ base() {
xfsprogs
"
_pkg $_docdir/ceph sample.ceph.conf sample.fetch_config
_pkg $_bindir crushtool monmaptool osdmaptool ceph-run ceph-detect-init
_pkg $_sbindir ceph-create-keys mount.ceph
_pkg $_libexecdir/ceph ceph_common.sh
@ -292,7 +289,7 @@ radosgw() {
pkgdesc="Rados REST gateway which implements Amazon's S3 and OpenStack's Swift APIs."
depends="ceph-common"
_pkg $_bindir radosgw radosgw-admin radosgw-token radosgw-object-expirer
_pkg $_bindir radosgw radosgw-admin radosgw-token radosgw-es radosgw-object-expirer
mkdir -p $subpkgdir$_localstatedir/lib/ceph/radosgw
}
@ -301,7 +298,7 @@ osd() {
depends="ceph-base parted gptfdisk"
_pkg $_bindir ceph-clsinfo ceph-bluefs-tool ceph-objectstore-tool ceph-osd
_pkg $_sbindir ceph-disk ceph-disk-udev
_pkg $_sbindir ceph-disk
_pkg $_libexecdir/ceph ceph-osd-prestart.sh
_pkg $_udevrulesdir 60-ceph-by-parttypeuuid.rules 95-ceph-osd.rules
install -m 750 -o $_ceph_uid -g $_ceph_gid -d \

380
ceph/ceph.spec
View File

File diff suppressed because it is too large Load Diff

374
ceph/ceph.spec.in
View File

File diff suppressed because it is too large Load Diff

184
ceph/cmake/modules/BuildBoost.cmake Normal file
View File

@ -0,0 +1,184 @@
# This module builds Boost
# executables are. It sets the following variables:
#
# Boost_FOUND : boolean - system has Boost
# Boost_LIBRARIES : list(filepath) - the libraries needed to use Boost
# Boost_INCLUDE_DIRS : list(path) - the Boost include directories
#
# Following hints are respected
#
# Boost_USE_STATIC_LIBS : boolean (default: OFF)
# Boost_USE_MULTITHREADED : boolean (default: OFF)
# BOOST_J: integer (defanult 1)
function(do_build_boost version)
cmake_parse_arguments(Boost_BUILD "" "" COMPONENTS ${ARGN})
set(boost_features "variant=release")
if(Boost_USE_MULTITHREADED)
list(APPEND boost_features "threading=multi")
else()
list(APPEND boost_features "threading=single")
endif()
if(Boost_USE_STATIC_LIBS)
list(APPEND boost_features "link=static")
else()
list(APPEND boost_features "link=shared")
endif()
if(CMAKE_SIZEOF_VOID_P EQUAL 8)
list(APPEND boost_features "address-model=64")
else()
list(APPEND boost_features "address-model=32")
endif()
set(BOOST_CXXFLAGS "-fPIC -w") # check on arm, etc <---XXX
list(APPEND boost_features "cxxflags=${BOOST_CXXFLAGS}")
string(REPLACE ";" "," boost_with_libs "${Boost_BUILD_COMPONENTS}")
# build b2 and prepare the project-config.jam for boost
set(configure_command
./bootstrap.sh --prefix=<INSTALL_DIR>
--with-libraries=${boost_with_libs})
set(b2 ./b2)
if(BOOST_J)
message(STATUS "BUILDING Boost Libraries at j ${BOOST_J}")
list(APPEND b2 -j${BOOST_J})
endif()
if(CMAKE_VERBOSE_MAKEFILE)
list(APPEND b2 -d1)
else()
list(APPEND b2 -d0)
endif()
if(NOT CMAKE_HOST_SYSTEM_PROCESSOR STREQUAL CMAKE_SYSTEM_PROCESSOR)
# we are crosscompiling
if(CMAKE_CXX_COMPILER_ID STREQUAL GNU)
set(b2_cc gcc)
elseif(CMAKE_CXX_COMPILER_ID STREQUAL Clang)
set(b2_cc clang)
else()
message(SEND_ERROR "unknown compiler: ${CMAKE_CXX_COMPILER_ID}")
endif()
# edit the config.jam so, b2 will be able to use the specified toolset
execute_process(
COMMAND
sed -i
"s|using ${b2_cc} ;|using ${b2_cc} : ${CMAKE_SYSTEM_PROCESSOR} : ${CMAKE_CXX_COMPILER} ;|"
${PROJECT_SOURCE_DIR}/src/boost/project-config.jam)
# use ${CMAKE_SYSTEM_PROCESSOR} as the version identifier of compiler
list(APPEND b2 toolset=${b2_cc}-${CMAKE_SYSTEM_PROCESSOR})
endif()
set(build_command
${b2} headers stage
#"--buildid=ceph" # changes lib names--can omit for static
${boost_features})
set(install_command
${b2} install)
set(boost_root_dir "${CMAKE_BINARY_DIR}/boost")
if(EXISTS "${PROJECT_SOURCE_DIR}/src/boost/libs/config/include/boost/config.hpp")
message(STATUS "boost already in src")
set(source_dir
SOURCE_DIR "${PROJECT_SOURCE_DIR}/src/boost")
elseif(version VERSION_GREATER 1.63)
message(FATAL_ERROR "Unknown BOOST_REQUESTED_VERSION: ${version}")
else()
message(STATUS "boost will be downloaded from sf.net")
set(boost_version 1.63.0)
set(boost_md5 1c837ecd990bb022d07e7aab32b09847)
string(REPLACE "." "_" boost_version_underscore ${boost_version} )
set(boost_url http://downloads.sourceforge.net/project/boost/boost/${boost_version}/boost_${boost_version_underscore}.tar.bz2)
set(source_dir
URL ${boost_url}
URL_MD5 ${boost_md5})
if(CMAKE_VERSION VERSION_GREATER 3.0)
list(APPEND source_dir DOWNLOAD_NO_PROGRESS 1)
endif()
endif()
# build all components in a single shot
include(ExternalProject)
ExternalProject_Add(Boost
${source_dir}
CONFIGURE_COMMAND CC=${CMAKE_C_COMPILER} CXX=${CMAKE_CXX_COMPILER} ${configure_command}
BUILD_COMMAND CC=${CMAKE_C_COMPILER} CXX=${CMAKE_CXX_COMPILER} ${build_command}
BUILD_IN_SOURCE 1
INSTALL_COMMAND ${install_command}
PREFIX "${boost_root_dir}")
endfunction()
macro(build_boost version)
do_build_boost(version ${ARGN})
ExternalProject_Get_Property(Boost install_dir)
set(Boost_INCLUDE_DIRS ${install_dir}/include)
set(Boost_INCLUDE_DIR ${install_dir}/include)
# create the directory so cmake won't complain when looking at the imported
# target
file(MAKE_DIRECTORY ${Boost_INCLUDE_DIRS})
cmake_parse_arguments(Boost_BUILD "" "" COMPONENTS ${ARGN})
foreach(c ${Boost_BUILD_COMPONENTS})
string(TOUPPER ${c} upper_c)
if(Boost_USE_STATIC_LIBS)
add_library(Boost::${c} STATIC IMPORTED)
else()
add_library(Boost::${c} SHARED IMPORTED)
endif()
add_dependencies(Boost::${c} Boost)
if(Boost_USE_STATIC_LIBS)
set(Boost_${upper_c}_LIBRARY
${install_dir}/lib/${CMAKE_STATIC_LIBRARY_PREFIX}boost_${c}${CMAKE_STATIC_LIBRARY_SUFFIX})
else()
set(Boost_${upper_c}_LIBRARY
${install_dir}/lib/${CMAKE_SHARED_LIBRARY_PREFIX}boost_${c}${CMAKE_SHARED_LIBRARY_SUFFIX})
endif()
set_target_properties(Boost::${c} PROPERTIES
INTERFACE_INCLUDE_DIRECTORIES "${Boost_INCLUDE_DIRS}"
IMPORTED_LINK_INTERFACE_LANGUAGES "CXX"
IMPORTED_LOCATION "${Boost_${upper_c}_LIBRARY}")
list(APPEND Boost_LIBRARIES ${Boost_${upper_c}_LIBRARY})
endforeach()
# for header-only libraries
if(CMAKE_VERSION VERSION_LESS 3.3)
# only ALIAS and INTERFACE target names allow ":" in it, but
# INTERFACE library is not allowed until cmake 3.1
add_custom_target(Boost.boost DEPENDS Boost)
else()
add_library(Boost.boost INTERFACE IMPORTED)
set_target_properties(Boost.boost PROPERTIES
INTERFACE_INCLUDE_DIRECTORIES "${Boost_INCLUDE_DIRS}")
add_dependencies(Boost.boost Boost)
endif()
find_package_handle_standard_args(Boost DEFAULT_MSG
Boost_INCLUDE_DIRS Boost_LIBRARIES)
mark_as_advanced(Boost_LIBRARIES BOOST_INCLUDE_DIRS)
endmacro()
function(maybe_add_boost_dep target)
get_target_property(imported ${target} IMPORTED)
if(imported)
return()
endif()
get_target_property(type ${target} TYPE)
if(NOT type MATCHES "OBJECT_LIBRARY|STATIC_LIBRARY|SHARED_LIBRARY|EXECUTABLE")
return()
endif()
get_target_property(sources ${target} SOURCES)
foreach(src ${sources})
get_filename_component(ext ${src} EXT)
# assuming all cxx source files include boost header(s)
if(ext MATCHES ".cc|.cpp|.cxx")
add_dependencies(${target} Boost.boost)
return()
endif()
endforeach()
endfunction()
# override add_library() to add Boost headers dependency
function(add_library target)
_add_library(${target} ${ARGN})
maybe_add_boost_dep(${target})
endfunction()
function(add_executable target)
_add_executable(${target} ${ARGN})
maybe_add_boost_dep(${target})
endfunction()

28
ceph/cmake/modules/Findatomic_ops.cmake
View File

@ -1,28 +0,0 @@
# - Find atomic_ops
# Find the native ATOMIC_OPS headers and libraries.
#
# ATOMIC_OPS_INCLUDE_DIRS - where to find atomic_ops.h, etc.
# ATOMIC_OPS_LIBRARIES - List of libraries when using atomic_ops.
# ATOMIC_OPS_FOUND - True if atomic_ops found.
# Look for the header file.
FIND_PATH(ATOMIC_OPS_INCLUDE_DIR NAMES atomic_ops.h)
# Look for the library.
FIND_LIBRARY(ATOMIC_OPS_LIBRARY NAMES atomic_ops)
# handle the QUIETLY and REQUIRED arguments and set ATOMIC_OPS_FOUND to TRUE if
# all listed variables are TRUE
INCLUDE(FindPackageHandleStandardArgs)
FIND_PACKAGE_HANDLE_STANDARD_ARGS(atomic_ops DEFAULT_MSG ATOMIC_OPS_LIBRARY ATOMIC_OPS_INCLUDE_DIR)
# Copy the results to the output variables.
IF(ATOMIC_OPS_FOUND)
SET(ATOMIC_OPS_LIBRARIES ${ATOMIC_OPS_LIBRARY})
SET(ATOMIC_OPS_INCLUDE_DIRS ${ATOMIC_OPS_INCLUDE_DIR})
ELSE(ATOMIC_OPS_FOUND)
SET(ATOMIC_OPS_LIBRARIES)
SET(ATOMIC_OPS_INCLUDE_DIRS)
ENDIF(ATOMIC_OPS_FOUND)
MARK_AS_ADVANCED(ATOMIC_OPS_INCLUDE_DIR ATOMIC_OPS_LIBRARY)

15
ceph/cmake/modules/Findpmem.cmake Normal file
View File

@ -0,0 +1,15 @@
# Try to find libpmem
#
# Once done, this will define
#
# PMEM_FOUND
# PMEM_INCLUDE_DIR
# PMEM_LIBRARY
find_path(PMEM_INCLUDE_DIR NAMES libpmem.h)
find_library(PMEM_LIBRARY NAMES pmem)
include(FindPackageHandleStandardArgs)
find_package_handle_standard_args(pmem DEFAULT_MSG PMEM_LIBRARY PMEM_INCLUDE_DIR)
mark_as_advanced(PMEM_INCLUDE_DIR PMEM_LIBRARY)

60
ceph/cmake/modules/SIMDExt.cmake
View File

@ -75,34 +75,38 @@ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "arm|ARM")
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i386|i686|amd64|x86_64|AMD64")
set(HAVE_INTEL 1)
CHECK_C_COMPILER_FLAG(-msse HAVE_INTEL_SSE)
if(HAVE_INTEL_SSE)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse")
endif()
CHECK_C_COMPILER_FLAG(-msse2 HAVE_INTEL_SSE2)
if(HAVE_INTEL_SSE2)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse2")
endif()
CHECK_C_COMPILER_FLAG(-msse3 HAVE_INTEL_SSE3)
if(HAVE_INTEL_SSE3)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse3")
endif()
CHECK_C_COMPILER_FLAG(-mssse3 HAVE_INTEL_SSSE3)
if(HAVE_INTEL_SSSE3)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -mssse3")
endif()
CHECK_C_COMPILER_FLAG(-mpclmul HAVE_INTEL_PCLMUL)
if(HAVE_INTEL_PCLMUL)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -mpclmul")
endif()
CHECK_C_COMPILER_FLAG(-msse4.1 HAVE_INTEL_SSE4_1)
if(HAVE_INTEL_SSE4_1)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse4.1")
endif()
CHECK_C_COMPILER_FLAG(-msse4.2 HAVE_INTEL_SSE4_2)
if(HAVE_INTEL_SSE4_2)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse4.2")
endif()
if(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
CHECK_C_COMPILER_FLAG(-msse HAVE_INTEL_SSE)
if(HAVE_INTEL_SSE)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse")
endif()
if(CMAKE_SYSTEM_PROCESSOR MATCHES "amd64|x86_64|AMD64")
CHECK_C_COMPILER_FLAG(-msse2 HAVE_INTEL_SSE2)
if(HAVE_INTEL_SSE2)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse2")
endif()
CHECK_C_COMPILER_FLAG(-msse3 HAVE_INTEL_SSE3)
if(HAVE_INTEL_SSE3)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse3")
endif()
CHECK_C_COMPILER_FLAG(-mssse3 HAVE_INTEL_SSSE3)
if(HAVE_INTEL_SSSE3)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -mssse3")
endif()
CHECK_C_COMPILER_FLAG(-mpclmul HAVE_INTEL_PCLMUL)
if(HAVE_INTEL_PCLMUL)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -mpclmul")
endif()
CHECK_C_COMPILER_FLAG(-msse4.1 HAVE_INTEL_SSE4_1)
if(HAVE_INTEL_SSE4_1)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse4.1")
endif()
CHECK_C_COMPILER_FLAG(-msse4.2 HAVE_INTEL_SSE4_2)
if(HAVE_INTEL_SSE4_2)
set(SIMD_COMPILE_FLAGS "${SIMD_COMPILE_FLAGS} -msse4.2")
endif()
endif(CMAKE_SYSTEM_PROCESSOR MATCHES "amd64|x86_64|AMD64")
endif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(powerpc|ppc)64le")
set(HAVE_PPC64LE 1)
message(STATUS " we are ppc64le")

2
ceph/debian/ceph-base.install
View File

@ -1,4 +1,3 @@
etc/bash_completion.d/ceph
etc/init.d/ceph
usr/sbin/ceph-create-keys
usr/bin/ceph-detect-init
@ -11,7 +10,6 @@ usr/lib/ceph/ceph_common.sh
usr/lib/ceph/erasure-code/*
usr/lib/rados-classes/*
usr/share/doc/ceph/sample.ceph.conf
usr/share/doc/ceph/sample.fetch_config
usr/share/man/man8/ceph-debugpack.8
usr/share/man/man8/ceph-deploy.8
usr/share/man/man8/ceph-run.8

2
ceph/debian/ceph-base.maintscript Normal file
View File

@ -0,0 +1,2 @@
rm_conffile /etc/logrotate.d/ceph.logrotate -- "$@"
rm_conffile /etc/logrotate.d/ceph -- "$@"

1
ceph/debian/ceph-common.install
View File

@ -1,5 +1,6 @@
#! /usr/bin/dh-exec --with=install
etc/bash_completion.d/ceph
etc/bash_completion.d/rados
etc/bash_completion.d/radosgw-admin
etc/bash_completion.d/rbd

1
ceph/debian/ceph-osd.install
View File

@ -1,7 +1,6 @@
lib/udev/rules.d/95-ceph-osd.rules
lib/udev/rules.d/60-ceph-by-parttypeuuid.rules
usr/sbin/ceph-disk
usr/sbin/ceph-disk-udev
usr/bin/ceph-clsinfo
usr/bin/ceph-objectstore-tool
usr/bin/ceph-bluestore-tool

2
ceph/debian/ceph-test.install
View File

@ -26,7 +26,5 @@ usr/bin/ceph_xattr_bench
usr/bin/ceph-monstore-tool
usr/bin/ceph-osdomap-tool
usr/bin/ceph-kvstore-tool
usr/bin/dmclock-tests
usr/bin/dmclock-data-struct-tests
usr/share/java/libcephfs-test.jar
usr/lib/ceph/ceph-monstore-update-crush.sh

6
ceph/debian/changelog
View File

@ -1,3 +1,9 @@
ceph (12.1.0-1) stable; urgency=medium
* New upstream release
-- Ceph Release Team <ceph-maintainers@ceph.com> Thu, 22 Jun 2017 15:43:47 +0000
ceph (12.0.3-1) stable; urgency=medium
* New upstream release

14
ceph/debian/control
View File

@ -9,6 +9,7 @@ Uploaders: Ken Dreyer <kdreyer@redhat.com>,
Alfredo Deza <adeza@redhat.com>
Build-Depends: bc,
btrfs-tools,
gperf,
cmake,
cpio,
cryptsetup-bin | cryptsetup,
@ -25,13 +26,11 @@ Build-Depends: bc,
jq,
junit4,
libaio-dev,
libatomic-ops-dev,
libbabeltrace-ctf-dev,
libbabeltrace-dev,
libblkid-dev (>= 2.17),
libcurl4-gnutls-dev,
libexpat1-dev,
libfcgi-dev,
libfuse-dev,
libgoogle-perftools-dev [i386 amd64 arm64],
libibverbs-dev,
@ -50,10 +49,13 @@ Build-Depends: bc,
pkg-config,
python (>= 2.7),
python-all-dev,
python-cherrypy3,
python-nose,
python-pecan,
python-prettytable,
python-setuptools,
python-sphinx,
python-werkzeug,
python3-all-dev,
python3-setuptools,
virtualenv | python-virtualenv,
@ -159,12 +161,16 @@ Description: debugging symbols for ceph-mds
Package: ceph-mgr
Architecture: linux-any
Depends: ceph-base (= ${binary:Version}),
python-openssl,
python-pecan,
python-werkzeug,
${misc:Depends},
${python:Depends},
python-cherrypy3,
${shlibs:Depends}
Replaces: ceph (<< 0.93-417)
Breaks: ceph (<< 0.93-417)
Description: metadata server for the ceph distributed file system
Description: manager for the ceph distributed storage system
Ceph is a massively scalable, open-source, distributed
storage system that runs on commodity hardware and delivers object,
block and file system storage.
@ -646,7 +652,7 @@ Description: debugging symbols for radosgw
Package: ceph-test
Architecture: linux-any
Depends: ceph-common, curl, xmlstarlet, ${misc:Depends}, ${shlibs:Depends}
Depends: ceph-common, curl, xmlstarlet, jq, ${misc:Depends}, ${shlibs:Depends}
Description: Ceph test and benchmarking tools
This package contains tools for testing and benchmarking Ceph.

4
ceph/debian/copyright
View File

@ -33,10 +33,6 @@ Files: src/common/bloom_filter.hpp
Copyright: Copyright (C) 2000 Arash Partow
License: Boost Software License, Version 1.0
Files: m4/acx_pthread.m4
Copyright: Steven G. Johnson <stevenj@alum.mit.edu>
License: GPLWithACException
Files: src/common/crc32c_intel*:
Copyright:
Copyright 2012-2013 Intel Corporation All Rights Reserved.

1
ceph/debian/radosgw.install
View File

@ -1,4 +1,5 @@
usr/bin/radosgw
usr/bin/radosgw-token
usr/bin/radosgw-es
usr/bin/radosgw-object-expirer
usr/share/man/man8/radosgw.8

11
ceph/debian/rules
View File

@ -22,6 +22,13 @@ ifeq ($(DEB_HOST_ARCH), armel)
extraopts += -DWITH_ATOMIC_OPS=OFF
endif
ifneq (,$(filter $(DEB_HOST_ARCH), arm armel armhf arm64 i386 amd64 mips mipsel powerpc ppc64))
# beast depends on libboost_context which only support the archs above
extraopts += -DWITH_RADOSGW_BEAST_FRONTEND=ON
else
extraopts += -DWITH_RADOSGW_BEAST_FRONTEND=OFF
endif
%:
dh $@ --buildsystem=cmake --with javahelper,python2,python3,systemd --parallel
@ -52,8 +59,8 @@ override_dh_installdocs:
dh_installdocs -a --all ChangeLog
override_dh_installlogrotate:
cp src/logrotate.conf debian/ceph-base.ceph.logrotate
dh_installlogrotate -pceph-base --name=ceph
cp src/logrotate.conf debian/ceph-common.logrotate
dh_installlogrotate -pceph-common
override_dh_installinit:
# dh_installinit is only set up to handle one upstart script

9
ceph/do_cmake.sh
View File

@ -4,9 +4,16 @@ if test -e build; then
echo 'build dir already exists; rm -rf build and re-run'
exit 1
fi
ARGS=""
if which ccache ; then
echo "enabling ccache"
ARGS+="-DWITH_CCACHE=ON"
fi
mkdir build
cd build
cmake -DBOOST_J=$(nproc) "$@" ..
cmake -DBOOST_J=$(nproc) $ARGS "$@" ..
# minimal config to find plugins
cat <<EOF > ceph.conf

8
ceph/doc/api/index.rst
View File

@ -33,14 +33,6 @@ See `librgw-py`_.
.. _librgw-py: ../radosgw/api
Calamari APIs
=============
See `Calamari API`_.
.. _Calamari API: http://ceph.com/calamari/docs/calamari_rest/index.html
Ceph Object Store APIs
======================

24
ceph/doc/architecture.rst
View File

@ -207,10 +207,10 @@ of failure or bottleneck when using ``cephx``. The monitor returns an
authentication data structure similar to a Kerberos ticket that contains a
session key for use in obtaining Ceph services. This session key is itself
encrypted with the user's permanent secret key, so that only the user can
request services from the Ceph monitor(s). The client then uses the session key
request services from the Ceph Monitor(s). The client then uses the session key
to request its desired services from the monitor, and the monitor provides the
client with a ticket that will authenticate the client to the OSDs that actually
handle data. Ceph monitors and OSDs share a secret, so the client can use the
handle data. Ceph Monitors and OSDs share a secret, so the client can use the
ticket provided by the monitor with any OSD or metadata server in the cluster.
Like Kerberos, ``cephx`` tickets expire, so an attacker cannot use an expired
ticket or session key obtained surreptitiously. This form of authentication will
@ -338,7 +338,7 @@ dispatch--which is a **huge** bottleneck at the petabyte-to-exabyte scale.
Ceph eliminates the bottleneck: Ceph's OSD Daemons AND Ceph Clients are cluster
aware. Like Ceph clients, each Ceph OSD Daemon knows about other Ceph OSD
Daemons in the cluster. This enables Ceph OSD Daemons to interact directly with
other Ceph OSD Daemons and Ceph monitors. Additionally, it enables Ceph Clients
other Ceph OSD Daemons and Ceph Monitors. Additionally, it enables Ceph Clients
to interact directly with Ceph OSD Daemons.
The ability of Ceph Clients, Ceph Monitors and Ceph OSD Daemons to interact with
@ -360,13 +360,15 @@ ability to leverage this computing power leads to several major benefits:
Ceph Client requests. If a Ceph OSD Daemon is ``down`` and ``in`` the Ceph
Storage Cluster, this status may indicate the failure of the Ceph OSD
Daemon. If a Ceph OSD Daemon is not running (e.g., it crashes), the Ceph OSD
Daemon cannot notify the Ceph Monitor that it is ``down``. The Ceph Monitor
can ping a Ceph OSD Daemon periodically to ensure that it is running.
However, Ceph also empowers Ceph OSD Daemons to determine if a neighboring
OSD is ``down``, to update the cluster map and to report it to the Ceph
monitor(s). This means that Ceph monitors can remain light weight processes.
See `Monitoring OSDs`_ and `Heartbeats`_ for additional details.
Daemon cannot notify the Ceph Monitor that it is ``down``. The OSDs
periodically send messages to the Ceph Monitor (``MPGStats`` pre-luminous,
and a new ``MOSDBeacon`` in luminous). If the Ceph Monitor doesn't see that
message after a configurable period of time then it marks the OSD down.
This mechanism is a failsafe, however. Normally, Ceph OSD Daemons will
determine if a neighboring OSD is down and report it to the Ceph Monitor(s).
This assures that Ceph Monitors are lightweight processes. See `Monitoring
OSDs`_ and `Heartbeats`_ for additional details.
#. **Data Scrubbing:** As part of maintaining data consistency and cleanliness,
Ceph OSD Daemons can scrub objects within placement groups. That is, Ceph
OSD Daemons can compare object metadata in one placement group with its
@ -1579,7 +1581,7 @@ instance for high availability.
.. _Monitoring OSDs and PGs: ../rados/operations/monitoring-osd-pg
.. _Heartbeats: ../rados/configuration/mon-osd-interaction
.. _Monitoring OSDs: ../rados/operations/monitoring-osd-pg/#monitoring-osds
.. _CRUSH - Controlled, Scalable, Decentralized Placement of Replicated Data: http://ceph.com/papers/weil-crush-sc06.pdf
.. _CRUSH - Controlled, Scalable, Decentralized Placement of Replicated Data: https://ceph.com/wp-content/uploads/2016/08/weil-crush-sc06.pdf
.. _Data Scrubbing: ../rados/configuration/osd-config-ref#scrubbing
.. _Report Peering Failure: ../rados/configuration/mon-osd-interaction#osds-report-peering-failure
.. _Troubleshooting Peering Failure: ../rados/troubleshooting/troubleshooting-pg#placement-group-down-peering-failure

37
ceph/doc/cephfs/administration.rst
View File

@ -42,6 +42,38 @@ creation of multiple filesystems use ``ceph fs flag set enable_multiple true``.
fs rm_data_pool <filesystem name> <pool name/id>
Settings
--------
::
fs set <fs name> max_file_size <size in bytes>
CephFS has a configurable maximum file size, and it's 1TB by default.
You may wish to set this limit higher if you expect to store large files
in CephFS. It is a 64-bit field.
Setting ``max_file_size`` to 0 does not disable the limit. It would
simply limit clients to only creating empty files.
Maximum file sizes and performance
----------------------------------
CephFS enforces the maximum file size limit at the point of appending to
files or setting their size. It does not affect how anything is stored.
When users create a file of an enormous size (without necessarily
writing any data to it), some operations (such as deletes) cause the MDS
to have to do a large number of operations to check if any of the RADOS
objects within the range that could exist (according to the file size)
really existed.
The ``max_file_size`` setting prevents users from creating files that
appear to be eg. exabytes in size, causing load on the MDS as it tries
to enumerate the objects during operations like stats or deletes.
Daemons
-------
@ -126,6 +158,11 @@ filesystem.
Legacy
------
The ``ceph mds set`` command is the deprecated version of ``ceph fs set``,
from before there was more than one filesystem per cluster. It operates
on whichever filesystem is marked as the default (see ``ceph fs
set-default``.)
::
mds stat

204
ceph/doc/cephfs/eviction.rst
View File

@ -1,4 +1,5 @@
===============================
Ceph filesystem client eviction
===============================
@ -6,114 +7,139 @@ When a filesystem client is unresponsive or otherwise misbehaving, it
may be necessary to forcibly terminate its access to the filesystem. This
process is called *eviction*.
This process is somewhat thorough in order to protect against data inconsistency
resulting from misbehaving clients.
Evicting a CephFS client prevents it from communicating further with MDS
daemons and OSD daemons. If a client was doing buffered IO to the filesystem,
any un-flushed data will be lost.
OSD blacklisting
----------------
Clients may either be evicted automatically (if they fail to communicate
promptly with the MDS), or manually (by the system administrator).
First, prevent the client from performing any more data operations by *blacklisting*
it at the RADOS level. You may be familiar with this concept as *fencing* in other
storage systems.
The client eviction process applies to clients of all kinds, this includes
FUSE mounts, kernel mounts, nfs-ganesha gateways, and any process using
libcephfs.
Identify the client to evict from the MDS session list:
Automatic client eviction
=========================
There are two situations in which a client may be evicted automatically:
On an active MDS daemon, if a client has not communicated with the MDS for
over ``mds_session_autoclose`` seconds (300 seconds by default), then it
will be evicted automatically.
During MDS startup (including on failover), the MDS passes through a
state called ``reconnect``. During this state, it waits for all the
clients to connect to the new MDS daemon. If any clients fail to do
so within the time window (``mds_reconnect_timeout``, 45 seconds by default)
then they will be evicted.
A warning message is sent to the cluster log if either of these situations
arises.
Manual client eviction
======================
Sometimes, the administrator may want to evict a client manually. This
could happen if a client is died and the administrator does not
want to wait for its session to time out, or it could happen if
a client is misbehaving and the administrator does not have access to
the client node to unmount it.
It is useful to inspect the list of clients first:
::
# ceph daemon mds.a session ls
ceph tell mds.0 client ls
[
{ "id": 4117,
"num_leases": 0,
"num_caps": 1,
"state": "open",
"replay_requests": 0,
"reconnecting": false,
"inst": "client.4117 172.16.79.251:0\/3271",
"client_metadata": { "entity_id": "admin",
"hostname": "fedoravm.localdomain",
"mount_point": "\/home\/user\/mnt"}}]
{
"id": 4305,
"num_leases": 0,
"num_caps": 3,
"state": "open",
"replay_requests": 0,
"completed_requests": 0,
"reconnecting": false,
"inst": "client.4305 172.21.9.34:0/422650892",
"client_metadata": {
"ceph_sha1": "ae81e49d369875ac8b569ff3e3c456a31b8f3af5",
"ceph_version": "ceph version 12.0.0-1934-gae81e49 (ae81e49d369875ac8b569ff3e3c456a31b8f3af5)",
"entity_id": "0",
"hostname": "senta04",
"mount_point": "/tmp/tmpcMpF1b/mnt.0",
"pid": "29377",
"root": "/"
}
}
]
In this case the 'fedoravm' client has address ``172.16.79.251:0/3271``, so we blacklist
it as follows:
Once you have identified the client you want to evict, you can
do that using its unique ID, or various other attributes to identify it:
::
# These all work
ceph tell mds.0 client evict id=4305
ceph tell mds.0 client evict client_metadata.=4305
Advanced: Un-blacklisting a client
==================================
Ordinarily, a blacklisted client may not reconnect to the servers: it
must be unmounted and then mounted anew.
However, in some situations it may be useful to permit a client that
was evicted to attempt to reconnect.
Because CephFS uses the RADOS OSD blacklist to control client eviction,
CephFS clients can be permitted to reconnect by removing them from
the blacklist:
::
# ceph osd blacklist add 172.16.79.251:0/3271
blacklisting 172.16.79.251:0/3271 until 2014-12-09 13:09:56.569368 (3600 sec)
ceph osd blacklist ls
# ... identify the address of the client ...
ceph osd blacklist rm <address>
OSD epoch barrier
-----------------
Doing this may put data integrity at risk if other clients have accessed
files that the blacklisted client was doing buffered IO to. It is also not
guaranteed to result in a fully functional client -- the best way to get
a fully healthy client back after an eviction is to unmount the client
and do a fresh mount.
While the evicted client is now marked as blacklisted in the central (mon) copy of the OSD
map, it is now necessary to ensure that this OSD map update has propagated to all daemons
involved in subsequent filesystem I/O. To do this, use the ``osdmap barrier`` MDS admin
socket command.
If you are trying to reconnect clients in this way, you may also
find it useful to set ``client_reconnect_stale`` to true in the
FUSE client, to prompt the client to try to reconnect.
First read the latest OSD epoch:
Advanced: Configuring blacklisting
==================================
::
If you are experiencing frequent client evictions, due to slow
client hosts or an unreliable network, and you cannot fix the underlying
issue, then you may want to ask the MDS to be less strict.
# ceph osd dump
epoch 12
fsid fd61ca96-53ff-4311-826c-f36b176d69ea
created 2014-12-09 12:03:38.595844
modified 2014-12-09 12:09:56.619957
...
It is possible to respond to slow clients by simply dropping their
MDS sessions, but permit them to re-open sessions and permit them
to continue talking to OSDs. To enable this mode, set
``mds_session_blacklist_on_timeout`` to false on your MDS nodes.
In this case it is 12. Now request the MDS to barrier on this epoch:
For the equivalent behaviour on manual evictions, set
``mds_session_blacklist_on_evict`` to false.
::
Note that if blacklisting is disabled, then evicting a client will
only have an effect on the MDS you send the command to. On a system
with multiple active MDS daemons, you would need to send an
eviction command to each active daemon. When blacklisting is enabled
(the default), sending an eviction to command to just a single
MDS is sufficient, because the blacklist propagates it to the others.
# ceph daemon mds.a osdmap barrier 12
Advanced options
================
MDS session eviction
--------------------
``mds_blacklist_interval`` - this setting controls how many seconds
entries will remain in the blacklist for.
Finally, it is safe to evict the client's MDS session, such that any capabilities it held
may be issued to other clients. The ID here is the ``id`` attribute from the ``session ls``
output:
::
# ceph daemon mds.a session evict 4117
That's it! The client has now been evicted, and any resources it had locked will
now be available for other clients.
Background: OSD epoch barrier
-----------------------------
The purpose of the barrier is to ensure that when we hand out any
capabilities which might allow touching the same RADOS objects, the
clients we hand out the capabilities to must have a sufficiently recent
OSD map to not race with cancelled operations (from ENOSPC) or
blacklisted clients (from evictions)
More specifically, the cases where we set an epoch barrier are:
* Client eviction (where the client is blacklisted and other clients
must wait for a post-blacklist epoch to touch the same objects)
* OSD map full flag handling in the client (where the client may
cancel some OSD ops from a pre-full epoch, so other clients must
wait until the full epoch or later before touching the same objects).
* MDS startup, because we don't persist the barrier epoch, so must
assume that latest OSD map is always required after a restart.
Note that this is a global value for simplicity: we could maintain this on
a per-inode basis. We don't, because:
* It would be more complicated
* It would use an extra 4 bytes of memory for every inode
* It would not be much more efficient as almost always everyone has the latest
OSD map anyway, in most cases everyone will breeze through this barrier
rather than waiting.
* We only do this barrier in very rare cases, so any benefit from per-inode
granularity would only very rarely be seen.
The epoch barrier is transmitted along with all capability messages, and
instructs the receiver of the message to avoid sending any more RADOS
operations to OSDs until it has seen this OSD epoch. This mainly applies
to clients (doing their data writes directly to files), but also applies
to the MDS because things like file size probing and file deletion are
done directly from the MDS.

8
ceph/doc/cephfs/file-layouts.rst
View File

@ -34,6 +34,14 @@ stripe_count
object_size
Integer in bytes. File data is chunked into RADOS objects of this size.
.. tip::
RADOS enforces a configurable limit on object sizes: if you increase CephFS
object sizes beyond that limit then writes may not succeed. The OSD
setting is ``rados_max_object_size``, which is 128MB by default.
Very large RADOS objects may prevent smooth operation of the cluster,
so increasing the object size limit past the default is not recommended.
Reading layouts with ``getfattr``
---------------------------------

3
ceph/doc/cephfs/multimds.rst
View File

@ -122,6 +122,7 @@ extended attribute of directories. The name of this extended attribute is
``ceph.dir.pin``. Users can set this attribute using standard commands:
::
setfattr -n ceph.dir.pin -v 2 path/to/dir
The value of the extended attribute is the rank to assign the directory subtree
@ -133,9 +134,11 @@ children. However, the parents pin can be overriden by setting the child
directory's export pin. For example:
::
mkdir -p a/b
# "a" and "a/b" both start without an export pin set
setfattr -n ceph.dir.pin -v 1 a/
# a and b are now pinned to rank 1
setfattr -n ceph.dir.pin -v 0 a/b
# a/b is now pinned to rank 0 and a/ and the rest of its children are still pinned to rank 1

2
ceph/doc/dev/blkin.rst
View File

@ -16,7 +16,7 @@ in realtime. The LTTng traces can then be visualized with Twitter's
Zipkin_.
.. _Dapper: http://static.googleusercontent.com/media/research.google.com/el//pubs/archive/36356.pdf
.. _Zipkin: http://twitter.github.io/zipkin/
.. _Zipkin: http://zipkin.io/
Installing Blkin

8
ceph/doc/dev/dev_cluster_deployement.rst
View File

@ -45,10 +45,6 @@ Options
Add *config* to all sections in the ceph configuration.
.. option:: -r
Start radosgw on port starting from 8000.
.. option:: --nodaemon
Use ceph-run as wrapper for mon/osd/mds.
@ -73,10 +69,6 @@ Options
Launch the osd/mds/mon/all the ceph binaries using valgrind with the specified tool and arguments.
.. option:: --{mon,osd,mds}_num
Set the count of mon/osd/mds daemons
.. option:: --bluestore
Use bluestore as the objectstore backend for osds

4
ceph/doc/dev/development-workflow.rst
View File

@ -55,7 +55,7 @@ Release Cycle
Four times a year, the development roadmap is discussed online during
the `Ceph Developer Summit <http://wiki.ceph.com/Planning/CDS/>`_. A
the `Ceph Developer Summit <http://tracker.ceph.com/projects/ceph/wiki/Planning#Ceph-Developer-Summit>`_. A
new stable release (hammer, infernalis, jewel ...) is published at the same
frequency. Every other release (firefly, hammer, jewel...) is a `Long Term
Stable (LTS) <../../releases>`_. See `Understanding the release cycle
@ -126,7 +126,7 @@ Running and interpreting teuthology integration tests
The :doc:`/dev/sepia` runs `teuthology
<https://github.com/ceph/teuthology/>`_ integration tests `on a regular basis <http://tracker.ceph.com/projects/ceph-releases/wiki/HOWTO_monitor_the_automated_tests_AKA_nightlies#Automated-tests-AKA-nightlies>`_ and the
results are posted on `pulpito <http://pulpito.ceph.com/>`_ and the
`ceph-qa mailing list <http://ceph.com/resources/mailing-list-irc/>`_.
`ceph-qa mailing list <https://ceph.com/irc/>`_.
* The job failures are `analyzed by quality engineers and developers
<http://tracker.ceph.com/projects/ceph-releases/wiki/HOWTO_monitor_the_automated_tests_AKA_nightlies#List-of-suites-and-watchers>`_

2
ceph/doc/dev/documenting.rst
View File

@ -30,7 +30,7 @@ API`_ provides a complete example. It is pulled into Sphinx by
`librados.rst`_, which is rendered at :doc:`/rados/api/librados`.
.. _`librados C API`: https://github.com/ceph/ceph/blob/master/src/include/rados/librados.h
.. _`librados.rst`: https://raw.github.com/ceph/ceph/master/doc/api/librados.rst
.. _`librados.rst`: https://github.com/ceph/ceph/raw/master/doc/rados/api/librados.rst
Drawing diagrams
================

91
ceph/doc/dev/index.rst
View File

@ -135,7 +135,7 @@ in the body of the message.
There are also `other Ceph-related mailing lists`_.
.. _`other Ceph-related mailing lists`: https://ceph.com/resources/mailing-list-irc/
.. _`other Ceph-related mailing lists`: https://ceph.com/irc/
IRC
---
@ -145,7 +145,7 @@ time using `Internet Relay Chat`_.
.. _`Internet Relay Chat`: http://www.irchelp.org/
See https://ceph.com/resources/mailing-list-irc/ for how to set up your IRC
See https://ceph.com/irc/ for how to set up your IRC
client and a list of channels.
Submitting patches
@ -750,7 +750,7 @@ The results of the nightlies are published at http://pulpito.ceph.com/ and
http://pulpito.ovh.sepia.ceph.com:8081/. The developer nick shows in the
test results URL and in the first column of the Pulpito dashboard. The
results are also reported on the `ceph-qa mailing list
<http://ceph.com/resources/mailing-list-irc/>`_ for analysis.
<https://ceph.com/irc/>`_ for analysis.
Suites inventory
----------------
@ -1132,11 +1132,14 @@ Reducing the number of tests
----------------------------
The ``rados`` suite generates thousands of tests out of a few hundred
files. For instance, all tests in the `rados/thrash suite
<https://github.com/ceph/ceph/tree/master/qa/suites/rados/thrash>`_
run for ``xfs``, ``btrfs`` and ``ext4`` because they are combined (via
special file ``%``) with the `fs directory
<https://github.com/ceph/ceph/tree/master/qa/suites/rados/thrash/fs>`_
files. This happens because teuthology constructs test matrices from
subdirectories wherever it encounters a file named ``%``. For instance,
all tests in the `rados/basic suite
<https://github.com/ceph/ceph/tree/master/qa/suites/rados/basic>`_
run with different messenger types: ``simple``, ``async`` and
``random``, because they are combined (via the special file ``%``) with
the `msgr directory
<https://github.com/ceph/ceph/tree/master/qa/suites/rados/basic/msgr>`_
All integration tests are required to be run before a Ceph release is published.
When merely verifying whether a contribution can be merged without
@ -1199,9 +1202,9 @@ Getting ceph-workbench
Since testing in the cloud is done using the `ceph-workbench
ceph-qa-suite`_ tool, you will need to install that first. It is designed
to be installed via Docker, so if you don't have Docker running on your
development machine, take care of that first. The Docker project has a good
tutorial called `Get Started with Docker Engine for Linux
<https://docs.docker.com/linux/>`_ if you unsure how to proceed.
development machine, take care of that first. You can follow `the official
tutorial<https://docs.docker.com/engine/installation/>`_ to install if
you have not installed yet.
Once Docker is up and running, install ``ceph-workbench`` by following the
`Installation instructions in the ceph-workbench documentation
@ -1434,28 +1437,14 @@ Refer to :doc:`install/build-ceph`.
You can do step 2 separately while it is building.
Step 2 - s3-tests
-----------------
The test suite is in a separate git repo, and is written in python. Perform the
following steps for jewel::
git clone git://github.com/ceph/s3-tests
cd s3-tests
git checkout ceph-jewel
./bootstrap
For kraken, checkout the ``ceph-kraken`` branch instead of ``ceph-jewel``. For
master, use ``ceph-master``.
Step 3 - vstart
Step 2 - vstart
---------------
When the build completes, and still in the top-level directory of the git
clone where you built Ceph, do the following::
clone where you built Ceph, do the following, for cmake builds::
cd src/
./vstart.sh -n -r --mds_num 0
cd build/
RGW=1 ../vstart.sh -n
This will produce a lot of output as the vstart cluster is started up. At the
end you should see a message like::
@ -1464,51 +1453,13 @@ end you should see a message like::
This means the cluster is running.
Step 4 - prepare S3 environment
-------------------------------
The s3-tests suite expects to run in a particular environment (S3 users, keys,
configuration file).
Before you try to prepare the environment, make sure you don't have any
existing keyring or ``ceph.conf`` files in ``/etc/ceph``.
For jewel, Abhishek Lekshmanan wrote a script that can be used for this
purpose. Assuming you are testing jewel, run the following commands from the
``src/`` directory of your ceph clone (where you just started the vstart
cluster)::
pushd ~
wget https://gist.githubusercontent.com/theanalyst/2fee6bc2780f67c79cad7802040fcddc/raw/b497ddba053d9a6fb5d91b73924cbafcfc32f137/s3tests-bootstrap.sh
popd
sh ~/s3tests-bootstrap.sh
If the script is successful, it will display a blob of JSON and create a file
called ``s3.conf`` in the current directory.
Step 5 - run s3-tests
Step 3 - run s3-tests
---------------------
To actually run the tests, take note of the full path to the ``s3.conf`` file
created in the previous step and then move to the directory where you cloned
``s3-tests`` in Step 2.
First, verify that the test suite is there and can be run::
S3TEST_CONF=/path/to/s3.conf ./virtualenv/bin/nosetests -a '!fails_on_rgw' -v --collect-only
This should complete quickly - it is like a "dry run" of all the tests in the
suite.
Finally, run the test suite itself::
S3TEST_CONF=/path/to/s3.conf ./virtualenv/bin/nosetests -a '!fails_on_rgw' -v
Note: the following test is expected to error - this is a problem in the test
setup (WIP), not an actual test failure::
ERROR: s3tests.functional.test_s3.test_bucket_acl_grant_email
To run the s3tests suite do the following::
$ ../qa/workunits/rgw/run-s3tests.sh
.. WIP
.. ===

2
ceph/doc/dev/osd_internals/erasure_coding/developer_notes.rst
View File

@ -189,7 +189,7 @@ in the registry. The `ErasureCodePluginExample <https://github.com/ceph/ceph/blo
The *ErasureCodePlugin* derived object must provide a factory method
from which the concrete implementation of the *ErasureCodeInterface*
object can be generated. The `ErasureCodePluginExample plugin <https://github.com/ceph/ceph/blob/v0.78/src/test/osd/ErasureCodePluginExample.cc>`_ reads:
object can be generated. The `ErasureCodePluginExample plugin <https://github.com/ceph/ceph/blob/v0.78/src/test/erasure-code/ErasureCodePluginExample.cc>`_ reads:
::

182
ceph/doc/dev/perf_counters.rst
View File

@ -39,89 +39,121 @@ The ``perf schema`` command dumps a json description of which values are availab
+------+-------------------------------------+
| 2 | unsigned 64-bit integer value |
+------+-------------------------------------+
| 4 | average (sum + count pair) |
| 4 | average (sum + count pair), where |
+------+-------------------------------------+
| 8 | counter (vs gauge) |
+------+-------------------------------------+
Every value will have either bit 1 or 2 set to indicate the type (float or integer). If bit 8 is set (counter), the reader may want to subtract off the previously read value to get the delta during the previous interval.
Every value will have either bit 1 or 2 set to indicate the type
(float or integer).
If bit 4 is set (average), there will be two values to read, a sum and a count. If it is a counter, the average for the previous interval would be sum delta (since the previous read) divided by the count delta. Alternatively, dividing the values outright would provide the lifetime average value. Normally these are used to measure latencies (number of requests and a sum of request latencies), and the average for the previous interval is what is interesting.
If bit 8 is set (counter), the value is monotonically increasing and
the reader may want to subtract off the previously read value to get
the delta during the previous interval.
If bit 4 is set (average), there will be two values to read, a sum and
a count. If it is a counter, the average for the previous interval
would be sum delta (since the previous read) divided by the count
delta. Alternatively, dividing the values outright would provide the
lifetime average value. Normally these are used to measure latencies
(number of requests and a sum of request latencies), and the average
for the previous interval is what is interesting.
Instead of interpreting the bit fields, the ``metric type`` has a
value of either ``guage`` or ``counter``, and the ``value type``
property will be one of ``real``, ``integer``, ``real-integer-pair``
(for a sum + real count pair), or ``integer-integer-pair`` (for a
sum + integer count pair).
Here is an example of the schema output::
{
"throttle-msgr_dispatch_throttler-hbserver" : {
"get_or_fail_fail" : {
"type" : 10
},
"get_sum" : {
"type" : 10
},
"max" : {
"type" : 10
},
"put" : {
"type" : 10
},
"val" : {
"type" : 10
},
"take" : {
"type" : 10
},
"get_or_fail_success" : {
"type" : 10
},
"wait" : {
"type" : 5
},
"get" : {
"type" : 10
},
"take_sum" : {
"type" : 10
},
"put_sum" : {
"type" : 10
}
},
"throttle-msgr_dispatch_throttler-client" : {
"get_or_fail_fail" : {
"type" : 10
},
"get_sum" : {
"type" : 10
},
"max" : {
"type" : 10
},
"put" : {
"type" : 10
},
"val" : {
"type" : 10
},
"take" : {
"type" : 10
},
"get_or_fail_success" : {
"type" : 10
},
"wait" : {
"type" : 5
},
"get" : {
"type" : 10
},
"take_sum" : {
"type" : 10
},
"put_sum" : {
"type" : 10
}
}
}
{
"throttle-bluestore_throttle_bytes": {
"val": {
"type": 2,
"metric_type": "gauge",
"value_type": "integer",
"description": "Currently available throttle",
"nick": ""
},
"max": {
"type": 2,
"metric_type": "gauge",
"value_type": "integer",
"description": "Max value for throttle",
"nick": ""
},
"get_started": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Number of get calls, increased before wait",
"nick": ""
},
"get": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Gets",
"nick": ""
},
"get_sum": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Got data",
"nick": ""
},
"get_or_fail_fail": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Get blocked during get_or_fail",
"nick": ""
},
"get_or_fail_success": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Successful get during get_or_fail",
"nick": ""
},
"take": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Takes",
"nick": ""
},
"take_sum": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Taken data",
"nick": ""
},
"put": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Puts",
"nick": ""
},
"put_sum": {
"type": 10,
"metric_type": "counter",
"value_type": "integer",
"description": "Put data",
"nick": ""
},
"wait": {
"type": 5,
"metric_type": "gauge",
"value_type": "real-integer-pair",
"description": "Waiting latency",
"nick": ""
}
}
Dump

677
ceph/doc/dev/perf_histograms.rst Normal file
View File

@ -0,0 +1,677 @@
=================
Perf histograms
=================
The perf histograms build on perf counters infrastructure. Histograms are built for a number of counters and simplify gathering data on which groups of counter values occur most often over time.
Perf histograms are currently unsigned 64-bit integer counters, so they're mostly useful for time and sizes. Data dumped by perf histogram can then be feed into other analysis tools/scripts.
Access
------
The perf histogram data are accessed via the admin socket. For example::
ceph daemon osd.0 perf histogram schema
ceph daemon osd.0 perf histogram dump
Collections
-----------
The histograms are grouped into named collections, normally representing a subsystem or an instance of a subsystem. For example, the internal ``throttle`` mechanism reports statistics on how it is throttling, and each instance is named something like::
op_r_latency_out_bytes_histogram
op_rw_latency_in_bytes_histogram
op_rw_latency_out_bytes_histogram
...
Schema
------
The ``perf histogram schema`` command dumps a json description of which values are available, and what their type is. Each named value as a ``type`` bitfield, with the 5-th bit always set and following bits defined.
+------+-------------------------------------+
| bit | meaning |
+======+=====================================+
| 1 | floating point value |
+------+-------------------------------------+
| 2 | unsigned 64-bit integer value |
+------+-------------------------------------+
| 4 | average (sum + count pair) |
+------+-------------------------------------+
| 8 | counter (vs gauge) |
+------+-------------------------------------+
In other words, histogram of type "18" is a histogram of unsigned 64-bit integer values (16 + 2).
Here is an example of the schema output::
{
"AsyncMessenger::Worker-0": {},
"AsyncMessenger::Worker-1": {},
"AsyncMessenger::Worker-2": {},
"mutex-WBThrottle::lock": {},
"objecter": {},
"osd": {
"op_r_latency_out_bytes_histogram": {
"type": 18,
"description": "Histogram of operation latency (including queue time) + data read",
"nick": ""
},
"op_w_latency_in_bytes_histogram": {
"type": 18,
"description": "Histogram of operation latency (including queue time) + data written",
"nick": ""
},
"op_rw_latency_in_bytes_histogram": {
"type": 18,
"description": "Histogram of rw operation latency (including queue time) + data written",
"nick": ""
},
"op_rw_latency_out_bytes_histogram": {
"type": 18,
"description": "Histogram of rw operation latency (including queue time) + data read",
"nick": ""
}
}
}
Dump
----
The actual dump is similar to the schema, except that there are actual value groups. For example::
"osd": {
"op_r_latency_out_bytes_histogram": {
"axes": [
{
"name": "Latency (usec)",
"min": 0,
"quant_size": 100000,
"buckets": 32,
"scale_type": "log2",
"ranges": [
{
"max": -1
},
{
"min": 0,
"max": 99999
},
{
"min": 100000,
"max": 199999
},
{
"min": 200000,
"max": 399999
},
{
"min": 400000,
"max": 799999
},
{
"min": 800000,
"max": 1599999
},
{
"min": 1600000,
"max": 3199999
},
{
"min": 3200000,
"max": 6399999
},
{
"min": 6400000,
"max": 12799999
},
{
"min": 12800000,
"max": 25599999
},
{
"min": 25600000,
"max": 51199999
},
{
"min": 51200000,
"max": 102399999
},
{
"min": 102400000,
"max": 204799999
},
{
"min": 204800000,
"max": 409599999
},
{
"min": 409600000,
"max": 819199999
},
{
"min": 819200000,
"max": 1638399999
},
{
"min": 1638400000,
"max": 3276799999
},
{
"min": 3276800000,
"max": 6553599999
},
{
"min": 6553600000,
"max": 13107199999
},
{
"min": 13107200000,
"max": 26214399999
},
{
"min": 26214400000,
"max": 52428799999
},
{
"min": 52428800000,
"max": 104857599999
},
{
"min": 104857600000,
"max": 209715199999
},
{
"min": 209715200000,
"max": 419430399999
},
{
"min": 419430400000,
"max": 838860799999
},
{
"min": 838860800000,
"max": 1677721599999
},
{
"min": 1677721600000,
"max": 3355443199999
},
{
"min": 3355443200000,
"max": 6710886399999
},
{
"min": 6710886400000,
"max": 13421772799999
},
{
"min": 13421772800000,
"max": 26843545599999
},
{
"min": 26843545600000,
"max": 53687091199999
},
},
{
"min": 53687091200000
}
]
},
{
"name": "Request size (bytes)",
"min": 0,
"quant_size": 512,
"buckets": 32,
"scale_type": "log2",
"ranges": [
{
"max": -1
},
{
"min": 0,
"max": 511
},
{
"min": 512,
"max": 1023
},
{
"min": 1024,
"max": 2047
},
{
"min": 2048,
"max": 4095
},
{
"min": 4096,
"max": 8191
},
{
"min": 8192,
"max": 16383
},
{
"min": 16384,
"max": 32767
},
{
"min": 32768,
"max": 65535
},
{
"min": 65536,
"max": 131071
},
{
"min": 131072,
"max": 262143
},
{
"min": 262144,
"max": 524287
},
{
"min": 524288,
"max": 1048575
},
{
"min": 1048576,
"max": 2097151
},
{
"min": 2097152,
"max": 4194303
},
{
"min": 4194304,
"max": 8388607
},
{
"min": 8388608,
"max": 16777215
},
{
"min": 16777216,
"max": 33554431
},
{
"min": 33554432,
"max": 67108863
},
{
"min": 67108864,
"max": 134217727
},
{
"min": 134217728,
"max": 268435455
},
{
"min": 268435456,
"max": 536870911
},
{
"min": 536870912,
"max": 1073741823
},
{
"min": 1073741824,
"max": 2147483647
},
{
"min": 2147483648,
"max": 4294967295
},
{
"min": 4294967296,
"max": 8589934591
},
{
"min": 8589934592,
"max": 17179869183
},
{
"min": 17179869184,
"max": 34359738367
},
{
"min": 34359738368,
"max": 68719476735
},
{
"min": 68719476736,
"max": 137438953471
},
{
"min": 137438953472,
"max": 274877906943
},
{
"min": 274877906944
}
]
}
],
"values": [
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
[
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
]
}
},
This represents the 2d histogram, consisting of 9 history entrires and 32 value groups per each history entry.
"Ranges" element denote value bounds for each of value groups. "Buckets" denote amount of value groups ("buckets"),
"Min" is a minimum accepted valaue, "quant_size" is quantization unit and "scale_type" is either "log2" (logarhitmic
scale) or "linear" (linear scale).
You can use histogram_dump.py tool (see src/tools/histogram_dump.py) for quick visualisation of existing histogram
data.

11
ceph/doc/dev/placement-group.rst
View File

@ -45,12 +45,11 @@ is the primary and the rest are replicas.
Many PGs can map to one OSD.
A PG represents nothing but a grouping of objects; you configure the
number of PGs you want (see
http://ceph.com/wiki/Changing_the_number_of_PGs ), number of
OSDs * 100 is a good starting point, and all of your stored objects
are pseudo-randomly evenly distributed to the PGs. So a PG explicitly
does NOT represent a fixed amount of storage; it represents 1/pg_num
'th of the storage you happen to have on your OSDs.
number of PGs you want, number of OSDs * 100 is a good starting point
, and all of your stored objects are pseudo-randomly evenly distributed
to the PGs. So a PG explicitly does NOT represent a fixed amount of
storage; it represents 1/pg_num'th of the storage you happen to have
on your OSDs.
Ignoring the finer points of CRUSH and custom placement, it goes
something like this in pseudocode::

5
ceph/doc/dev/quick_guide.rst
View File

@ -106,12 +106,13 @@ you might do something like this:
Running a RadosGW development environment
-----------------------------------------
Add the ``-r`` to vstart.sh to enable the RadosGW
Set the ``RGW`` environment variable when running vstart.sh to enable the RadosGW.
.. code::
$ cd build
$ ../src/vstart.sh -d -n -x -r
$ RGW=1 ../src/vstart.sh -d -n -x
You can now use the swift python client to communicate with the RadosGW.

3
ceph/doc/install/build-ceph.rst
View File

@ -28,9 +28,8 @@ Ceph is built using cmake. To build Ceph, navigate to your cloned Ceph
repository and execute the following::
cd ceph
mkdir build
./do_cmake.sh
cd build
cmake ..
make
.. topic:: Hyperthreading

4
ceph/doc/install/get-packages.rst
View File

@ -107,9 +107,7 @@ You may find releases for CentOS/RHEL and others (installed with YUM) at::
https://download.ceph.com/rpm-{release-name}
The major releases of Ceph are summarized at:
http://docs.ceph.com/docs/master/releases/
The major releases of Ceph are summarized at: :doc:`/releases`.
Every second major release is considered Long Term Stable (LTS). Critical
bugfixes are backported to LTS releases until their retirement. Since retired

7
ceph/doc/install/index.rst
View File

@ -21,7 +21,7 @@ repository and build Ceph yourself.
Get Tarballs <get-tarballs>
Clone Source <clone-source>
Build Ceph <build-ceph>
Ceph Mirrors <mirrors>
Ceph Mirrors <mirrors>
Install Software
@ -38,7 +38,7 @@ QEMU.
:maxdepth: 1
Install ceph-deploy <install-ceph-deploy>
Install Ceph Storage Cluster <install-storage-cluster>
Install Ceph Storage Cluster <install-storage-cluster>
Install Ceph Object Gateway <install-ceph-gateway>
Install Virtualization for Block <install-vm-cloud>
@ -52,7 +52,8 @@ deployment scripts with Chef, Juju, Puppet, etc.
.. toctree::
Manual Deployment <manual-deployment>
Manual Deployment <manual-deployment>
Manual Deployment on FreeBSD <manual-freebsd-deployment>
Upgrade Software
================

8
ceph/doc/install/manual-deployment.rst
View File

@ -58,7 +58,7 @@ a number of things:
For example, when you run multiple clusters in a `federated architecture`_,
the cluster name (e.g., ``us-west``, ``us-east``) identifies the cluster for
the current CLI session. **Note:** To identify the cluster name on the
command line interface, specify the a Ceph configuration file with the
command line interface, specify the Ceph configuration file with the
cluster name (e.g., ``ceph.conf``, ``us-west.conf``, ``us-east.conf``, etc.).
Also see CLI usage (``ceph --cluster {cluster-name}``).
@ -290,6 +290,12 @@ The procedure is as follows:
**Note:** Once you add OSDs and start them, the placement group health errors
should disappear. See the next section for details.
Manager daemon configuration
============================
On each node where you run a ceph-mon daemon, you should also set up a ceph-mgr daemon.
See `../mgr/administrator`_
Adding OSDs
===========

624
ceph/doc/install/manual-freebsd-deployment.rst Normal file
View File

@ -0,0 +1,624 @@
==============================
Manual Deployment on FreeBSD
==============================
This a largely a copy of the regular Manual Deployment with FreeBSD specifics.
The difference lies in two parts: The underlying diskformat, and the way to use
the tools.
All Ceph clusters require at least one monitor, and at least as many OSDs as
copies of an object stored on the cluster. Bootstrapping the initial monitor(s)
is the first step in deploying a Ceph Storage Cluster. Monitor deployment also
sets important criteria for the entire cluster, such as the number of replicas
for pools, the number of placement groups per OSD, the heartbeat intervals,
whether authentication is required, etc. Most of these values are set by
default, so it's useful to know about them when setting up your cluster for
production.
Following the same configuration as `Installation (Quick)`_, we will set up a
cluster with ``node1`` as the monitor node, and ``node2`` and ``node3`` for
OSD nodes.
.. ditaa::
/------------------\ /----------------\
| Admin Node | | node1 |
| +-------->+ |
| | | cCCC |
\---------+--------/ \----------------/
|
| /----------------\
| | node2 |
+----------------->+ |
| | cCCC |
| \----------------/
|
| /----------------\
| | node3 |
+----------------->| |
| cCCC |
\----------------/
Disklayout on FreeBSD
=====================
Current implementation works on ZFS pools
* All Ceph data is created in /var/lib/ceph
* Log files go into /var/log/ceph
* PID files go into /var/log/run
* One ZFS pool is allocated per OSD, like::
gpart create -s GPT ada1
gpart add -t freebsd-zfs -l osd1 ada1
zpool create -o mountpoint=/var/lib/ceph/osd/osd.1 osd
* Some cache and log (ZIL) can be attached.
Please note that this is different from the Ceph journals. Cache and log are
totally transparent for Ceph, and help the filesystem to keep the system
consistant and help performance.
Assuming that ada2 is an SSD::
gpart create -s GPT ada2
gpart add -t freebsd-zfs -l osd1-log -s 1G ada2
zpool add osd1 log gpt/osd1-log
gpart add -t freebsd-zfs -l osd1-cache -s 10G ada2
zpool add osd1 log gpt/osd1-cache
* Note: *UFS2 does not allow large xattribs*
Configuration
-------------
As per FreeBSD default parts of extra software go into ``/usr/local/``. Which
means that for ``/etc/ceph.conf`` the default location is
``/usr/local/etc/ceph/ceph.conf``. Smartest thing to do is to create a softlink
from ``/etc/ceph`` to ``/usr/local/etc/ceph``::
ln -s /usr/local/etc/ceph /etc/ceph
A sample file is provided in ``/usr/local/share/doc/ceph/sample.ceph.conf``
Note that ``/usr/local/etc/ceph/ceph.conf`` will be found by most tools,
linking it to ``/etc/ceph/ceph.conf`` will help with any scripts that are found
in extra tools, scripts, and/or discussionlists.
Monitor Bootstrapping
=====================
Bootstrapping a monitor (a Ceph Storage Cluster, in theory) requires
a number of things:
- **Unique Identifier:** The ``fsid`` is a unique identifier for the cluster,
and stands for File System ID from the days when the Ceph Storage Cluster was
principally for the Ceph Filesystem. Ceph now supports native interfaces,
block devices, and object storage gateway interfaces too, so ``fsid`` is a
bit of a misnomer.
- **Cluster Name:** Ceph clusters have a cluster name, which is a simple string
without spaces. The default cluster name is ``ceph``, but you may specify
a different cluster name. Overriding the default cluster name is
especially useful when you are working with multiple clusters and you need to
clearly understand which cluster your are working with.
For example, when you run multiple clusters in a `federated architecture`_,
the cluster name (e.g., ``us-west``, ``us-east``) identifies the cluster for
the current CLI session. **Note:** To identify the cluster name on the
command line interface, specify the a Ceph configuration file with the
cluster name (e.g., ``ceph.conf``, ``us-west.conf``, ``us-east.conf``, etc.).
Also see CLI usage (``ceph --cluster {cluster-name}``).
- **Monitor Name:** Each monitor instance within a cluster has a unique name.
In common practice, the Ceph Monitor name is the host name (we recommend one
Ceph Monitor per host, and no commingling of Ceph OSD Daemons with
Ceph Monitors). You may retrieve the short hostname with ``hostname -s``.
- **Monitor Map:** Bootstrapping the initial monitor(s) requires you to
generate a monitor map. The monitor map requires the ``fsid``, the cluster
name (or uses the default), and at least one host name and its IP address.
- **Monitor Keyring**: Monitors communicate with each other via a
secret key. You must generate a keyring with a monitor secret and provide
it when bootstrapping the initial monitor(s).
- **Administrator Keyring**: To use the ``ceph`` CLI tools, you must have
a ``client.admin`` user. So you must generate the admin user and keyring,
and you must also add the ``client.admin`` user to the monitor keyring.
The foregoing requirements do not imply the creation of a Ceph Configuration
file. However, as a best practice, we recommend creating a Ceph configuration
file and populating it with the ``fsid``, the ``mon initial members`` and the
``mon host`` settings.
You can get and set all of the monitor settings at runtime as well. However,
a Ceph Configuration file may contain only those settings that override the
default values. When you add settings to a Ceph configuration file, these
settings override the default settings. Maintaining those settings in a
Ceph configuration file makes it easier to maintain your cluster.
The procedure is as follows:
#. Log in to the initial monitor node(s)::
ssh {hostname}
For example::
ssh node1
#. Ensure you have a directory for the Ceph configuration file. By default,
Ceph uses ``/etc/ceph``. When you install ``ceph``, the installer will
create the ``/etc/ceph`` directory automatically. ::
ls /etc/ceph
**Note:** Deployment tools may remove this directory when purging a
cluster (e.g., ``ceph-deploy purgedata {node-name}``, ``ceph-deploy purge
{node-name}``).
#. Create a Ceph configuration file. By default, Ceph uses
``ceph.conf``, where ``ceph`` reflects the cluster name. ::
sudo vim /etc/ceph/ceph.conf
#. Generate a unique ID (i.e., ``fsid``) for your cluster. ::
uuidgen
#. Add the unique ID to your Ceph configuration file. ::
fsid = {UUID}
For example::
fsid = a7f64266-0894-4f1e-a635-d0aeaca0e993
#. Add the initial monitor(s) to your Ceph configuration file. ::
mon initial members = {hostname}[,{hostname}]
For example::
mon initial members = node1
#. Add the IP address(es) of the initial monitor(s) to your Ceph configuration
file and save the file. ::
mon host = {ip-address}[,{ip-address}]
For example::
mon host = 192.168.0.1
**Note:** You may use IPv6 addresses instead of IPv4 addresses, but
you must set ``ms bind ipv6`` to ``true``. See `Network Configuration
Reference`_ for details about network configuration.
#. Create a keyring for your cluster and generate a monitor secret key. ::
ceph-authtool --create-keyring /tmp/ceph.mon.keyring --gen-key -n mon. --cap mon 'allow *'
#. Generate an administrator keyring, generate a ``client.admin`` user and add
the user to the keyring. ::
sudo ceph-authtool --create-keyring /etc/ceph/ceph.client.admin.keyring --gen-key -n client.admin --set-uid=0 --cap mon 'allow *' --cap osd 'allow *' --cap mds 'allow'
#. Add the ``client.admin`` key to the ``ceph.mon.keyring``. ::
ceph-authtool /tmp/ceph.mon.keyring --import-keyring /etc/ceph/ceph.client.admin.keyring
#. Generate a monitor map using the hostname(s), host IP address(es) and the FSID.
Save it as ``/tmp/monmap``::
monmaptool --create --add {hostname} {ip-address} --fsid {uuid} /tmp/monmap
For example::
monmaptool --create --add node1 192.168.0.1 --fsid a7f64266-0894-4f1e-a635-d0aeaca0e993 /tmp/monmap
#. Create a default data directory (or directories) on the monitor host(s). ::
sudo mkdir /var/lib/ceph/mon/{cluster-name}-{hostname}
For example::
sudo mkdir /var/lib/ceph/mon/ceph-node1
See `Monitor Config Reference - Data`_ for details.
#. Populate the monitor daemon(s) with the monitor map and keyring. ::
sudo -u ceph ceph-mon [--cluster {cluster-name}] --mkfs -i {hostname} --monmap /tmp/monmap --keyring /tmp/ceph.mon.keyring
For example::
sudo -u ceph ceph-mon --mkfs -i node1 --monmap /tmp/monmap --keyring /tmp/ceph.mon.keyring
#. Consider settings for a Ceph configuration file. Common settings include
the following::
[global]
fsid = {cluster-id}
mon initial members = {hostname}[, {hostname}]
mon host = {ip-address}[, {ip-address}]
public network = {network}[, {network}]
cluster network = {network}[, {network}]
auth cluster required = cephx
auth service required = cephx
auth client required = cephx
osd journal size = {n}
osd pool default size = {n} # Write an object n times.
osd pool default min size = {n} # Allow writing n copy in a degraded state.
osd pool default pg num = {n}
osd pool default pgp num = {n}
osd crush chooseleaf type = {n}
In the foregoing example, the ``[global]`` section of the configuration might
look like this::
[global]
fsid = a7f64266-0894-4f1e-a635-d0aeaca0e993
mon initial members = node1
mon host = 192.168.0.1
public network = 192.168.0.0/24
auth cluster required = cephx
auth service required = cephx
auth client required = cephx
osd journal size = 1024
osd pool default size = 2
osd pool default min size = 1
osd pool default pg num = 333
osd pool default pgp num = 333
osd crush chooseleaf type = 1
#. Touch the ``done`` file.
Mark that the monitor is created and ready to be started::
sudo touch /var/lib/ceph/mon/ceph-node1/done
#. And for FreeBSD an entry for every monitor needs to be added to the config
file. (The requirement will be removed in future releases).
The entry should look like::
[mon]
[mon.node1]
host = node1 # this name can be resolve
#. Start the monitor(s).
For Ubuntu, use Upstart::
sudo start ceph-mon id=node1 [cluster={cluster-name}]
In this case, to allow the start of the daemon at each reboot you
must create two empty files like this::
sudo touch /var/lib/ceph/mon/{cluster-name}-{hostname}/upstart
For example::
sudo touch /var/lib/ceph/mon/ceph-node1/upstart
For Debian/CentOS/RHEL, use sysvinit::
sudo /etc/init.d/ceph start mon.node1
For FreeBSD we use the rc.d init scripts (called bsdrc in Ceph)::
sudo service ceph start start mon.node1
For this to work /etc/rc.conf also needs the entry to enable ceph::
cat 'ceph_enable="YES"' >> /etc/rc.conf
#. Verify that Ceph created the default pools. ::
ceph osd lspools
You should see output like this::
0 data,1 metadata,2 rbd,
#. Verify that the monitor is running. ::
ceph -s
You should see output that the monitor you started is up and running, and
you should see a health error indicating that placement groups are stuck
inactive. It should look something like this::
cluster a7f64266-0894-4f1e-a635-d0aeaca0e993
health HEALTH_ERR 192 pgs stuck inactive; 192 pgs stuck unclean; no osds
monmap e1: 1 mons at {node1=192.168.0.1:6789/0}, election epoch 1, quorum 0 node1
osdmap e1: 0 osds: 0 up, 0 in
pgmap v2: 192 pgs, 3 pools, 0 bytes data, 0 objects
0 kB used, 0 kB / 0 kB avail
192 creating
**Note:** Once you add OSDs and start them, the placement group health errors
should disappear. See the next section for details.
Adding OSDs
===========
Once you have your initial monitor(s) running, you should add OSDs. Your cluster
cannot reach an ``active + clean`` state until you have enough OSDs to handle the
number of copies of an object (e.g., ``osd pool default size = 2`` requires at
least two OSDs). After bootstrapping your monitor, your cluster has a default
CRUSH map; however, the CRUSH map doesn't have any Ceph OSD Daemons mapped to
a Ceph Node.
Short Form
----------
Ceph provides the ``ceph-disk`` utility, which can prepare a disk, partition or
directory for use with Ceph. The ``ceph-disk`` utility creates the OSD ID by
incrementing the index. Additionally, ``ceph-disk`` will add the new OSD to the
CRUSH map under the host for you. Execute ``ceph-disk -h`` for CLI details.
The ``ceph-disk`` utility automates the steps of the `Long Form`_ below. To
create the first two OSDs with the short form procedure, execute the following
on ``node2`` and ``node3``:
#. Prepare the OSD. ::
On FreeBSD only existing directories can be use to create OSDs in::
ssh {node-name}
sudo ceph-disk prepare --cluster {cluster-name} --cluster-uuid {uuid} {path-to-ceph-osd-directory}
For example::
ssh node1
sudo ceph-disk prepare --cluster ceph --cluster-uuid a7f64266-0894-4f1e-a635-d0aeaca0e993 /var/lib/ceph/osd/osd.1
#. Activate the OSD::
sudo ceph-disk activate {data-path} [--activate-key {path}]
For example::
sudo ceph-disk activate /var/lib/ceph/osd/osd.1
**Note:** Use the ``--activate-key`` argument if you do not have a copy
of ``/var/lib/ceph/bootstrap-osd/{cluster}.keyring`` on the Ceph Node.
FreeBSD does not auto start the OSDs, but also requires a entry in
``ceph.conf``. One for each OSD::
[osd]
[osd.1]
host = node1 # this name can be resolve
Long Form
---------
Without the benefit of any helper utilities, create an OSD and add it to the
cluster and CRUSH map with the following procedure. To create the first two
OSDs with the long form procedure, execute the following on ``node2`` and
``node3``:
#. Connect to the OSD host. ::
ssh {node-name}
#. Generate a UUID for the OSD. ::
uuidgen
#. Create the OSD. If no UUID is given, it will be set automatically when the
OSD starts up. The following command will output the OSD number, which you
will need for subsequent steps. ::
ceph osd create [{uuid} [{id}]]
#. Create the default directory on your new OSD. ::
ssh {new-osd-host}
sudo mkdir /var/lib/ceph/osd/{cluster-name}-{osd-number}
Above are the ZFS instructions to do this for FreeBSD.
#. If the OSD is for a drive other than the OS drive, prepare it
for use with Ceph, and mount it to the directory you just created.
#. Initialize the OSD data directory. ::
ssh {new-osd-host}
sudo ceph-osd -i {osd-num} --mkfs --mkkey --osd-uuid [{uuid}]
The directory must be empty before you can run ``ceph-osd`` with the
``--mkkey`` option. In addition, the ceph-osd tool requires specification
of custom cluster names with the ``--cluster`` option.
#. Register the OSD authentication key. The value of ``ceph`` for
``ceph-{osd-num}`` in the path is the ``$cluster-$id``. If your
cluster name differs from ``ceph``, use your cluster name instead.::
sudo ceph auth add osd.{osd-num} osd 'allow *' mon 'allow profile osd' -i /var/lib/ceph/osd/{cluster-name}-{osd-num}/keyring
#. Add your Ceph Node to the CRUSH map. ::
ceph [--cluster {cluster-name}] osd crush add-bucket {hostname} host
For example::
ceph osd crush add-bucket node1 host
#. Place the Ceph Node under the root ``default``. ::
ceph osd crush move node1 root=default
#. Add the OSD to the CRUSH map so that it can begin receiving data. You may
also decompile the CRUSH map, add the OSD to the device list, add the host as a
bucket (if it's not already in the CRUSH map), add the device as an item in the
host, assign it a weight, recompile it and set it. ::
ceph [--cluster {cluster-name}] osd crush add {id-or-name} {weight} [{bucket-type}={bucket-name} ...]
For example::
ceph osd crush add osd.0 1.0 host=node1
#. After you add an OSD to Ceph, the OSD is in your configuration. However,
it is not yet running. The OSD is ``down`` and ``in``. You must start
your new OSD before it can begin receiving data.
For Ubuntu, use Upstart::
sudo start ceph-osd id={osd-num} [cluster={cluster-name}]
For example::
sudo start ceph-osd id=0
sudo start ceph-osd id=1
For Debian/CentOS/RHEL, use sysvinit::
sudo /etc/init.d/ceph start osd.{osd-num} [--cluster {cluster-name}]
For example::
sudo /etc/init.d/ceph start osd.0
sudo /etc/init.d/ceph start osd.1
In this case, to allow the start of the daemon at each reboot you
must create an empty file like this::
sudo touch /var/lib/ceph/osd/{cluster-name}-{osd-num}/sysvinit
For example::
sudo touch /var/lib/ceph/osd/ceph-0/sysvinit
sudo touch /var/lib/ceph/osd/ceph-1/sysvinit
Once you start your OSD, it is ``up`` and ``in``.
For FreeBSD using rc.d init::
After adding the OSD to ``ceph.conf``
sudo service ceph start osd.{osd-num}
For example::
sudo service ceph start osd.0
sudo service ceph start osd.1
In this case, to allow the start of the daemon at each reboot you
must create an empty file like this::
sudo touch /var/lib/ceph/osd/{cluster-name}-{osd-num}/bsdrc
For example::
sudo touch /var/lib/ceph/osd/ceph-0/bsdrc
sudo touch /var/lib/ceph/osd/ceph-1/bsdrc
Once you start your OSD, it is ``up`` and ``in``.
Adding MDS
==========
In the below instructions, ``{id}`` is an arbitrary name, such as the hostname of the machine.
#. Create the mds data directory.::
mkdir -p /var/lib/ceph/mds/{cluster-name}-{id}
#. Create a keyring.::
ceph-authtool --create-keyring /var/lib/ceph/mds/{cluster-name}-{id}/keyring --gen-key -n mds.{id}
#. Import the keyring and set caps.::
ceph auth add mds.{id} osd "allow rwx" mds "allow" mon "allow profile mds" -i /var/lib/ceph/mds/{cluster}-{id}/keyring
#. Add to ceph.conf.::
[mds.{id}]
host = {id}
#. Start the daemon the manual way.::
ceph-mds --cluster {cluster-name} -i {id} -m {mon-hostname}:{mon-port} [-f]
#. Start the daemon the right way (using ceph.conf entry).::
service ceph start
#. If starting the daemon fails with this error::
mds.-1.0 ERROR: failed to authenticate: (22) Invalid argument
Then make sure you do not have a keyring set in ceph.conf in the global section; move it to the client section; or add a keyring setting specific to this mds daemon. And verify that you see the same key in the mds data directory and ``ceph auth get mds.{id}`` output.
#. Now you are ready to `create a Ceph filesystem`_.
Summary
=======
Once you have your monitor and two OSDs up and running, you can watch the
placement groups peer by executing the following::
ceph -w
To view the tree, execute the following::
ceph osd tree
You should see output that looks something like this::
# id weight type name up/down reweight
-1 2 root default
-2 2 host node1
0 1 osd.0 up 1
-3 1 host node2
1 1 osd.1 up 1
To add (or remove) additional monitors, see `Add/Remove Monitors`_.
To add (or remove) additional Ceph OSD Daemons, see `Add/Remove OSDs`_.
.. _federated architecture: ../../radosgw/federated-config
.. _Installation (Quick): ../../start
.. _Add/Remove Monitors: ../../rados/operations/add-or-rm-mons
.. _Add/Remove OSDs: ../../rados/operations/add-or-rm-osds
.. _Network Configuration Reference: ../../rados/configuration/network-config-ref
.. _Monitor Config Reference - Data: ../../rados/configuration/mon-config-ref#data
.. _create a Ceph filesystem: ../../cephfs/createfs

1
ceph/doc/install/mirrors.rst
View File

@ -24,6 +24,7 @@ These mirrors are available on the following locations:
- **UK: UK**: http://uk.ceph.com
- **US-East: US East Coast**: http://us-east.ceph.com/
- **US-West: US West Coast**: http://us-west.ceph.com/
- **CN: China**: http://cn.ceph.com/
You can replace all download.ceph.com URLs with any of the mirrors, for example:

11
ceph/doc/man/8/ceph-detect-init.rst
View File

@ -41,6 +41,17 @@ Options
Display additional information for debugging.
Bugs
====
:program:`ceph-detect-init` is used by :program:`ceph-disk` to figure out the init system to manage the mount directory of an OSD. But only following combinations are fully tested:
- `upstart` on `Ubuntu 14.04`
- `systemd` on `Ubuntu 15.04` and up
- `systemd` on `Debian 8` and up
- `systemd` on `RHEL/CentOS 7` and up
- `systemd` on `Fedora 22` and up
Availability
============

6
ceph/doc/man/8/ceph-disk.rst
View File

@ -78,6 +78,11 @@ the subcommands ``deactivate`` and ``destroy``.
The documentation for each subcommand (prepare, activate, etc.) can be displayed
with its ``--help`` option. For instance ``ceph-disk prepare --help``.
Bugs
====
See also the ``Bugs`` section in :doc:`ceph-detect-init <ceph-detect-init>`\(8).
Availability
============
@ -87,5 +92,6 @@ the Ceph documentation at http://ceph.com/docs for more information.
See also
========
:doc:`ceph-detect-init <ceph-detect-init>`\(8)
:doc:`ceph-osd <ceph-osd>`\(8),
:doc:`ceph-deploy <ceph-deploy>`\(8)

14
ceph/doc/man/8/ceph-osd.rst
View File

@ -10,7 +10,7 @@ Synopsis
========
| **ceph-osd** -i *osdnum* [ --osd-data *datapath* ] [ --osd-journal
*journal* ] [ --mkfs ] [ --mkjournal ] [ --mkkey ]
*journal* ] [ --mkfs ] [ --mkjournal ] [--flush-journal] [--check-allows-journal] [--check-wants-journal] [--check-needs-journal] [ --mkkey ]
Description
@ -57,6 +57,18 @@ Options
Journal updates to *journal*.
.. option:: --check-wants-journal
Check whether a journal is desired.
.. option:: --check-allows-journal
Check whether a journal is allowed.
.. option:: --check-needs-journal
Check whether a journal is required.
.. option:: --mkfs
Create an empty object repository. This also initializes the journal

62
ceph/doc/man/8/ceph.rst
View File

@ -39,7 +39,7 @@ Synopsis
| **ceph** **mon_status**
| **ceph** **osd** [ *blacklist* \| *blocked-by* \| *create* \| *deep-scrub* \| *df* \| *down* \| *dump* \| *erasure-code-profile* \| *find* \| *getcrushmap* \| *getmap* \| *getmaxosd* \| *in* \| *lspools* \| *map* \| *metadata* \| *out* \| *pause* \| *perf* \| *pg-temp* \| *primary-affinity* \| *primary-temp* \| *repair* \| *reweight* \| *reweight-by-pg* \| *rm* \| *scrub* \| *set* \| *setcrushmap* \| *setmaxosd* \| *stat* \| *tree* \| *unpause* \| *unset* ] ...
| **ceph** **osd** [ *blacklist* \| *blocked-by* \| *create* \| *new* \| *deep-scrub* \| *df* \| *down* \| *dump* \| *erasure-code-profile* \| *find* \| *getcrushmap* \| *getmap* \| *getmaxosd* \| *in* \| *lspools* \| *map* \| *metadata* \| *out* \| *pause* \| *perf* \| *pg-temp* \| *primary-affinity* \| *primary-temp* \| *repair* \| *reweight* \| *reweight-by-pg* \| *rm* \| *destroy* \| *purge* \| *scrub* \| *set* \| *setcrushmap* \| *setmaxosd* \| *stat* \| *tree* \| *unpause* \| *unset* ] ...
| **ceph** **osd** **crush** [ *add* \| *add-bucket* \| *create-or-move* \| *dump* \| *get-tunable* \| *link* \| *move* \| *remove* \| *rename-bucket* \| *reweight* \| *reweight-all* \| *reweight-subtree* \| *rm* \| *rule* \| *set* \| *set-tunable* \| *show-tunables* \| *tunables* \| *unlink* ] ...
@ -478,10 +478,40 @@ Usage::
Subcommand ``create`` creates new osd (with optional UUID and ID).
This command is DEPRECATED as of the Luminous release, and will be removed in
a future release.
Subcommand ``new`` should instead be used.
Usage::
ceph osd create {<uuid>} {<id>}
Subcommand ``new`` reuses a previously destroyed OSD *id*. The new OSD will
have the specified *uuid*, and the command expects a JSON file containing
the base64 cephx key for auth entity *client.osd.<id>*, as well as optional
base64 cepx key for dm-crypt lockbox access and a dm-crypt key. Specifying
a dm-crypt requires specifying the accompanying lockbox cephx key.
Usage::
ceph osd new {<id>} {<uuid>} -i {<secrets.json>}
The secrets JSON file is expected to maintain a form of the following format::
{
"cephx_secret": "AQBWtwhZdBO5ExAAIDyjK2Bh16ZXylmzgYYEjg=="
}
Or::
{
"cephx_secret": "AQBWtwhZdBO5ExAAIDyjK2Bh16ZXylmzgYYEjg==",
"cephx_lockbox_secret": "AQDNCglZuaeVCRAAYr76PzR1Anh7A0jswkODIQ==",
"dmcrypt_key": "<dm-crypt key>"
}
Subcommand ``crush`` is used for CRUSH management. It uses some additional
subcommands.
@ -938,6 +968,29 @@ Usage::
ceph osd rm <ids> [<ids>...]
Subcommand ``destroy`` marks OSD *id* as *destroyed*, removing its cephx
entity's keys and all of its dm-crypt and daemon-private config key
entries.
This command will not remove the OSD from crush, nor will it remove the
OSD from the OSD map. Instead, once the command successfully completes,
the OSD will show marked as *destroyed*.
In order to mark an OSD as destroyed, the OSD must first be marked as
**lost**.
Usage::
ceph osd destroy <id> {--yes-i-really-mean-it}
Subcommand ``purge`` performs a combination of ``osd destroy``,
``osd rm`` and ``osd crush remove``.
Usage::
ceph osd purge <id> {--yes-i-really-mean-it}
Subcommand ``scrub`` initiates scrub on specified osd.
Usage::
@ -1265,6 +1318,13 @@ Usage::
ceph tell <name (type.id)> <args> [<args>...]
List all available commands.
Usage::
ceph tell <name (type.id)> help
version
-------

12
ceph/doc/man/8/rados.rst
View File

@ -138,6 +138,18 @@ Pool specific commands
:command:`cleanup`
:command:`listxattr` *name*
List all extended attributes of an object.
:command:`getxattr` *name* *attr*
Dump the extended attribute value of *attr* of an object.
:command:`setxattr` *name* *attr* *value*
Set the value of *attr* in the extended attributes of an object.
:command:`rmxattr` *name* *attr*
Remove *attr* from the extended attributes of an object.
:command:`listomapkeys` *name*
List all the keys stored in the object map of object name.

27
ceph/doc/mgr/administrator.rst
View File

@ -2,12 +2,16 @@
ceph-mgr administrator's guide
==============================
Setup
-----
Manual setup
------------
Create an authentication key for your daemon::
Usually, you would set up a ceph-mgr daemon using a tool such
as ceph-ansible. These instructions describe how to set up
a ceph-mgr daemon manually.
ceph auth get-or-create mgr.$name mon 'allow *'
First, create an authentication key for your daemon::
ceph auth get-or-create mgr.$name mon 'allow profile mgr' osd 'allow *' mds 'allow *'
Place that key into ``mgr data`` path, which for a cluster "ceph"
and mgr $name "foo" would be ``/var/lib/ceph/mgr/ceph-foo``.
@ -21,6 +25,15 @@ of ``ceph status``, which should now include a mgr status line::
mgr active: $name
Client authentication
---------------------
The manager is a new daemon which requires new CephX capabilities. If you upgrade
a cluster from an old version of Ceph, or use the default install/deploy tools,
your admin client should get this capability automatically. If you use tooling from
elsewhere, you may get EACCES errors when invoking certain ceph cluster commands.
To fix that, add a "mgr allow *" stanza to your client's cephx capabilities by
`Modifying User Capabilities`_.
High availability
-----------------
@ -75,9 +88,10 @@ OPTION(mgr_module_path, OPT_STR, CEPH_PKGLIBDIR "/mgr") // where to load python
:Type: String
:Default: ``"/var/lib/ceph/mgr/$cluster-$id"``
``mgr beacon period``
``mgr tick period``
:Description: How many seconds between mgr beacons to monitors
:Description: How many seconds between mgr beacons to monitors, and other
periodic checks.
:Type: Integer
:Default: ``5``
@ -87,3 +101,4 @@ OPTION(mgr_module_path, OPT_STR, CEPH_PKGLIBDIR "/mgr") // where to load python
:Type: Integer
:Default: ``30``
.. _Modifying User Capabilities: ../rados/operations/user-management#modify-user-capabilities

15
ceph/doc/mgr/dashboard.rst Normal file
View File

@ -0,0 +1,15 @@
dashboard plugin
================
Dashboard plugin visualizes the statistics of the cluster using a web server
hosted by ``ceph-mgr``. Like most web applications, dashboard binds to a host
name and port. Since each ``ceph-mgr`` hosts its own instance of dashboard, we
need to configure them separately. The hostname and port are stored using the
configuration key facility. So we can configure them like::
ceph config-key put mgr/dashboard/$name/server_addr $IP
ceph config-key put mgr/dashboard/$name/server_port $PORT
where ``$name`` is the ID of the ceph-mgr who is hosting this dashboard web app.
If they are not configured, the web app will be bound to ``127.0.0.1:7000``.

16
ceph/doc/mgr/index.rst
View File

@ -8,9 +8,25 @@ The :term:`Ceph Manager` daemon (ceph-mgr) runs alongside monitor daemons,
to provide additional monitoring and interfaces to external monitoring
and management systems.
Since the 12.x (*luminous*) Ceph release, the ceph-mgr daemon is required for
normal operations. The ceph-mgr daemon is an optional component in
the 11.x (*kraken*) Ceph release.
By default, the manager daemon requires no additional configuration, beyond
ensuring it is running. If there is no mgr daemon running, you will
see a health warning to that effect, and some of the other information
in the output of `ceph status` will be missing or stale until a mgr is started.
Use your normal deployment tools, such as ceph-ansible or ceph-deploy, to
set up ceph-mgr daemons on each of your mon nodes. It is not mandatory
to place mgr daemons on the same nodes as mons, but it is almost always
sensible.
.. toctree::
:maxdepth: 1
Installation and Configuration <administrator>
Dashboard <dashboard>
RESTful <restful>
Writing plugins <plugins>

25
ceph/doc/mgr/restful.rst Normal file
View File

@ -0,0 +1,25 @@
restful plugin
==============
RESTful plugin offers the REST API access to the status of the cluster. RESTful
plugin enables you to secure the API endpoints via SSL. If you don't have a
security certificate and key already, you need to create them first::
openssl req -new -nodes -x509 \
-subj "/O=IT/CN=ceph-mgr-restful" \
-days 3650 -keyout $PKEY -out $CERT -extensions v3_ca
where ``$PKEY`` and ``$CERT`` are the paths to the private key and the
certificate. And then you need to import the keystore to the cluster using the
configuration key facility, so RESTful plugin can read them at startup::
ceph config-key put mgr/restful/$name/crt -i $CERT
ceph config-key put mgr/restful/$name/key -i $PKEY
Also, like other web applications, RESTful plugin is bound to the a hostname and
a port::
ceph config-key put mgr/restful/$name/server_addr $IP
ceph config-key put mgr/restful/$name/server_port $PORT
If not specified, the plugin uses ``127.0.0.1:8003`` by default.

16
ceph/doc/rados/configuration/filestore-config-ref.rst
View File

@ -192,7 +192,7 @@ The following settings provide limits on the size of filestore queue.
:Description: Defines the maximum number of in progress operations the file store accepts before blocking on queuing new operations.
:Type: Integer
:Required: No. Minimal impact on performance.
:Default: ``500``
:Default: ``50``
``filestore queue max bytes``
@ -203,20 +203,6 @@ The following settings provide limits on the size of filestore queue.
:Default: ``100 << 20``
``filestore queue committing max ops``
:Description: The maximum number of operations the filestore can commit.
:Type: Integer
:Required: No
:Default: ``500``
``filestore queue committing max bytes``
:Description: The maximum number of bytes the filestore can commit.
:Type: Integer
:Required: No
:Default: ``100 << 20``
.. index:: filestore; timeouts

437
ceph/doc/rados/configuration/mon-config-ref.rst
View File

@ -106,6 +106,11 @@ A consensus requires a majority of monitors running to establish a quorum for
consensus about the cluster map (e.g., 1; 2 out of 3; 3 out of 5; 4 out of 6;
etc.).
``mon force quorum join``
:Description: Force monitor to join quorum even if it has been previously removed from the map
:Type: Boolean
:Default: ``False``
.. index:: Ceph Monitor; consistency
@ -314,6 +319,126 @@ by setting it in the ``[mon]`` section of the configuration file.
:Default: ``/var/lib/ceph/mon/$cluster-$id``
``mon data size warn``
:Description: Issue a ``HEALTH_WARN`` in cluster log when the monitor's data
store goes over 15GB.
:Type: Integer
:Default: 15*1024*1024*1024*
``mon data avail warn``
:Description: Issue a ``HEALTH_WARN`` in cluster log when the available disk
space of monitor's data store is lower or equal to this
percentage.
:Type: Integer
:Default: 30
``mon data avail crit``
:Description: Issue a ``HEALTH_ERR`` in cluster log when the available disk
space of monitor's data store is lower or equal to this
percentage.
:Type: Integer
:Default: 5
``mon warn on cache pools without hit sets``
:Description: Issue a ``HEALTH_WARN`` in cluster log if a cache pool does not
have the hitset type set set.
See `hit set type <../operations/pools#hit-set-type>`_ for more
details.
:Type: Boolean
:Default: True
``mon warn on crush straw calc version zero``
:Description: Issue a ``HEALTH_WARN`` in cluster log if the CRUSH's
``straw_calc_version`` is zero. See
`CRUSH map tunables <../operations/crush-map#tunables>`_ for
details.
:Type: Boolean
:Default: True
``mon warn on legacy crush tunables``
:Description: Issue a ``HEALTH_WARN`` in cluster log if
CRUSH tunables are too old (older than ``mon_min_crush_required_version``)
:Type: Boolean
:Default: True
``mon crush min required version``
:Description: The minimum tunable profile version required by the cluster.
See
`CRUSH map tunables <../operations/crush-map#tunables>`_ for
details.
:Type: String
:Default: ``firefly``
``mon warn on osd down out interval zero``
:Description: Issue a ``HEALTH_WARN`` in cluster log if
``mon osd down out interval`` is zero. Having this option set to
zero on the leader acts much like the ``noout`` flag. It's hard
to figure out what's going wrong with clusters witout the
``noout`` flag set but acting like that just the same, so we
report a warning in this case.
:Type: Boolean
:Default: True
``mon cache target full warn ratio``
:Description: Position between pool's ``cache_target_full`` and
``target_max_object`` where we start warning
:Type: Float
:Default: ``0.66``
``mon health data update interval``
:Description: How often (in seconds) the monitor in quorum shares its health
status with its peers. (negative number disables it)
:Type: Float
:Default: ``60``
``mon health to clog``
:Description: Enable sending health summary to cluster log periodically.
:Type: Boolean
:Default: True
``mon health to clog tick interval``
:Description: How often (in seconds) the monitor send health summary to cluster
log (a non-positive number disables it). If current health summary
is empty or identical to the last time, monitor will not send it
to cluster log.
:Type: Integer
:Default: 3600
``mon health to clog interval``
:Description: How often (in seconds) the monitor send health summary to cluster
log (a non-positive number disables it). Monitor will always
send the summary to cluster log no matter if the summary changes
or not.
:Type: Integer
:Default: 60
.. index:: Ceph Storage Cluster; capacity planning, Ceph Monitor; capacity planning
Storage Capacity
@ -546,7 +671,9 @@ Trimming requires that the placement groups are ``active + clean``.
``mon sync timeout``
:Description:
:Description: Number of seconds the monitor will wait for the next update
message from its sync provider before it gives up and bootstrap
again.
:Type: Double
:Default: ``30.0``
@ -560,39 +687,123 @@ Trimming requires that the placement groups are ``active + clean``.
``mon sync max payload size``
:Description: The maximum size for a sync payload.
:Description: The maximum size for a sync payload (in bytes).
:Type: 32-bit Integer
:Default: ``1045676``
``mon accept timeout``
``paxos max join drift``
:Description: Number of seconds the Leader will wait for the Requester(s) to
accept a Paxos update. It is also used during the Paxos recovery
phase for similar purposes.
:Description: The maximum Paxos iterations before we must first sync the
monitor data stores. When a monitor finds that its peer is too
far ahead of it, it will first sync with data stores before moving
on.
:Type: Integer
:Default: ``10``
:Type: Float
:Default: ``10.0``
``paxos stash full interval``
:Description: How often (in commits) to stash a full copy of the PaxosService state.
Current this setting only affects ``mds``, ``mon``, ``auth`` and ``mgr``
PaxosServices.
:Type: Integer
:Default: 25
``paxos propose interval``
:Description: Gather updates for this time interval before proposing
a map update.
a map update.
:Type: Double
:Default: ``1.0``
``paxos min``
:Description: The minimum number of paxos states to keep around
:Type: Integer
:Default: 500
``paxos min wait``
:Description: The minimum amount of time to gather updates after a period of
inactivity.
:Type: Double
:Default: ``0.05``
``paxos trim min``
:Description: Number of extra proposals tolerated before trimming
:Type: Integer
:Default: 250
``paxos trim max``
:Description: The maximum number of extra proposals to trim at a time
:Type: Integer
:Default: 500
``paxos service trim min``
:Description: The minimum amount of versions to trigger a trim (0 disables it)
:Type: Integer
:Default: 250
``paxos service trim max``
:Description: The maximum amount of versions to trim during a single proposal (0 disables it)
:Type: Integer
:Default: 500
``mon max log epochs``
:Description: The maximum amount of log epochs to trim during a single proposal
:Type: Integer
:Default: 500
``mon max pgmap epochs``
:Description: The maximum amount of pgmap epochs to trim during a single proposal
:Type: Integer
:Default: 500
``mon mds force trim to``
:Description: Force monitor to trim mdsmaps to this point (0 disables it.
dangerous, use with care)
:Type: Integer
:Default: 0
``mon osd force trim to``
:Description: Force monitor to trim osdmaps to this point, even if there is
PGs not clean at the specified epoch (0 disables it. dangerous,
use with care)
:Type: Integer
:Default: 0
``mon osd cache size``
:Description: The size of osdmaps cache, not to rely on underlying store's cache
:Type: Integer
:Default: 10
``mon election timeout``
:Description: On election proposer, maximum waiting time for all ACKs in seconds.
:Type: Float
:Default: ``5``
``mon lease``
:Description: The length (in seconds) of the lease on the monitor's versions.
@ -600,22 +811,30 @@ Trimming requires that the placement groups are ``active + clean``.
:Default: ``5``
``mon lease renew interval``
``mon lease renew interval factor``
:Description: The interval (in seconds) for the Leader to renew the other
monitor's leases.
:Description: ``mon lease`` \* ``mon lease renew interval factor`` will be the
interval for the Leader to renew the other monitor's leases. The
factor should be less than ``1.0``.
:Type: Float
:Default: ``3``
:Default: ``0.6``
``mon lease ack timeout``
``mon lease ack timeout factor``
:Description: The number of seconds the Leader will wait for the Providers to
acknowledge the lease extension.
:Description: The Leader will wait ``mon lease`` \* ``mon lease ack timeout factor``
for the Providers to acknowledge the lease extension.
:Type: Float
:Default: ``10.0``
:Default: ``2.0``
``mon accept timeout factor``
:Description: The Leader will wait ``mon lease`` \* ``mon accept timeout factor``
for the Requester(s) to accept a Paxos update. It is also used
during the Paxos recovery phase for similar purposes.
:Type: Float
:Default: ``2.0``
``mon min osdmap epochs``
@ -640,42 +859,6 @@ Trimming requires that the placement groups are ``active + clean``.
Slurp
-----
In Ceph version 0.58 and earlier, when a Paxos service drifts beyond a given
number of versions, Ceph triggers the `slurp` mechanism, which establishes a
connection with the quorum Leader and obtains every single version the Leader
has for every service that has drifted. In Ceph versions 0.59 and later, slurp
will not work, because there is a single Paxos instance for all services.
.. deprecated:: 0.58
``paxos max join drift``
:Description: The maximum Paxos iterations before we must first sync the
monitor data stores.
:Type: Integer
:Default: ``10``
``mon slurp timeout``
:Description: The number of seconds the monitor has to recover using slurp
before the process is aborted and the monitor bootstraps.
:Type: Double
:Default: ``10.0``
``mon slurp bytes``
:Description: Limits the slurp messages to the specified number of bytes.
:Type: 32-bit Integer
:Default: ``256 * 1024``
.. index:: Ceph Monitor; clock
Clock
@ -688,7 +871,7 @@ are not synchronized, it can lead to a number of anomalies. For example:
- Daemons ignoring received messages (e.g., timestamps outdated)
- Timeouts triggered too soon/late when a message wasn't received in time.
See `Monitor Store Synchronization`_ and `Slurp`_ for details.
See `Monitor Store Synchronization`_ for details.
.. tip:: You SHOULD install NTP on your Ceph monitor hosts to
@ -739,12 +922,19 @@ acceptable values.
``mon timecheck interval``
:Description: The time check interval (clock drift check) in seconds
for the leader.
for the Leader.
:Type: Float
:Default: ``300.0``
``mon timecheck skew interval``
:Description: The time check interval (clock drift check) in seconds when in
presence of a skew in seconds for the Leader.
:Type: Float
:Default: ``30.0``
Client
------
@ -835,12 +1025,6 @@ Miscellaneous
:Type: 32-bit Integer
:Default: ``100``
``mon sync fs threshold``
:Description: Synchronize with the filesystem when writing the specified number of objects. Set it to ``0`` to disable it.
:Type: 32-bit Integer
:Default: ``5``
``mon subscribe interval``
:Description: The refresh interval (in seconds) for subscriptions. The
@ -897,6 +1081,129 @@ Miscellaneous
:Default: ``0.5``
``mon osd prime pg temp max time estimate``
:Description: Maximum estimate of time spent on each PG before we prime all PGs
in parallel.
:Type: Float
:Default: ``0.25``
``mon osd allow primary affinity``
:Description: allow ``primary_affinity`` to be set in the osdmap.
:Type: Boolean
:Default: False
``mon osd pool ec fast read``
:Description: Whether turn on fast read on the pool or not. It will be used as
the default setting of newly created erasure pools if ``fast_read``
is not specified at create time.
:Type: Boolean
:Default: False
``mon mds skip sanity``
:Description: Skip safety assertions on FSMap (in case of bugs where we want to
continue anyway). Monitor terminates if the FSMap sanity check
fails, but we can disable it by enabling this option.
:Type: Boolean
:Default: False
``mon max mdsmap epochs``
:Description: The maximum amount of mdsmap epochs to trim during a single proposal.
:Type: Integer
:Default: 500
``mon config key max entry size``
:Description: The maximum size of config-key entry (in bytes)
:Type: Integer
:Default: 4096
``mon scrub interval``
:Description: How often (in seconds) the monitor scrub its store by comparing
the stored checksums with the computed ones of all the stored
keys.
:Type: Integer
:Default: 3600*24
``mon scrub max keys``
:Description: The maximum number of keys to scrub each time.
:Type: Integer
:Default: 100
``mon compact on start``
:Description: Compact the database used as Ceph Monitor store on
``ceph-mon`` start. A manual compaction helps to shrink the
monitor database and improve the performance of it if the regular
compaction fails to work.
:Type: Boolean
:Default: False
``mon compact on bootstrap``
:Description: Compact the database used as Ceph Monitor store on
on bootstrap. Monitor starts probing each other for creating
a quorum after bootstrap. If it times out before joining the
quorum, it will start over and bootstrap itself again.
:Type: Boolean
:Default: False
``mon compact on trim``
:Description: Compact a certain prefix (including paxos) when we trim its old states.
:Type: Boolean
:Default: True
``mon cpu threads``
:Description: Number of threads for performing CPU intensive work on monitor.
:Type: Boolean
:Default: True
``mon osd mapping pgs per chunk``
:Description: We calculate the mapping from placement group to OSDs in chunks.
This option specifies the number of placement groups per chunk.
:Type: Integer
:Default: 4096
``mon osd max split count``
:Description: Largest number of PGs per "involved" OSD to let split create.
When we increase the ``pg_num`` of a pool, the placement groups
will be splitted on all OSDs serving that pool. We want to avoid
extreme multipliers on PG splits.
:Type: Integer
:Default: 300
``mon session timeout``
:Description: Monitor will terminate inactive sessions stay idle over this
time limit.
:Type: Integer
:Default: 300
.. _Paxos: http://en.wikipedia.org/wiki/Paxos_(computer_science)
.. _Monitor Keyrings: ../../../dev/mon-bootstrap#secret-keys

7
ceph/doc/rados/configuration/mon-lookup-dns.rst
View File

@ -12,6 +12,13 @@ This allows for less configuration on clients and monitors. Using a DNS update c
By default clients and daemons will look for the TCP service called *ceph-mon* which is configured by the *mon_dns_srv_name* configuration directive.
``mon dns srv name``
:Description: the service name used querying the DNS for the monitor hosts/addresses
:Type: String
:Default: ``ceph-mon``
Example
-------
When the DNS search domain is set to *example.com* a DNS zone file might contain the following elements.

161
ceph/doc/rados/configuration/mon-osd-interaction.rst
View File

@ -66,39 +66,46 @@ or by setting the value at runtime.
|----+ Mark |
| | OSD 2 |
|<---+ Down |
.. index:: OSD down report
OSDs Report Down OSDs
=====================
By default, a Ceph OSD Daemon must report to the Ceph Monitors that another Ceph
OSD Daemon is ``down`` three times before the Ceph Monitors acknowledge that the
reported Ceph OSD Daemon is ``down``. By default, only one
Ceph OSD Daemon is required to report another Ceph OSD Daemon ``down``. You can
change the number of Ceph OSD Daemones required to report a Ceph OSD Daemon
``down`` to a Ceph Monitor by adding an ``mon osd min down reporters`` setting
(``osd min down reporters`` prior to v0.62) under the ``[mon]`` section of your
Ceph configuration file, or by setting the value at runtime.
By default, two Ceph OSD Daemons from different hosts must report to the Ceph
Monitors that another Ceph OSD Daemon is ``down`` before the Ceph Monitors
acknowledge that the reported Ceph OSD Daemon is ``down``. But there is chance
that all the OSDs reporting the failure are hosted in a rack with a bad switch
which has trouble connecting to another OSD. To avoid this sort of false alarm,
we consider the peers reporting a failure a proxy for a potential "subcluster"
over the overall cluster that is similarly laggy. This is clearly not true in
all cases, but will sometimes help us localize the grace correction to a subset
of the system that is unhappy. ``mon osd reporter subtree level`` is used to
group the peers into the "subcluster" by their common ancestor type in CRUSH
map. By default, only two reports from different subtree are required to report
another Ceph OSD Daemon ``down``. You can change the number of reporters from
unique subtrees and the common ancestor type required to report a Ceph OSD
Daemon ``down`` to a Ceph Monitor by adding an ``mon osd min down reporters``
and ``mon osd reporter subtree level`` settings under the ``[mon]`` section of
your Ceph configuration file, or by setting the value at runtime.
.. ditaa:: +---------+ +---------+
| OSD 1 | | Monitor |
+---------+ +---------+
| |
| OSD 2 Is Down |
|-------------->|
| |
| OSD 2 Is Down |
|-------------->|
| |
| OSD 2 Is Down |
|-------------->|
| |
| |----------+ Mark
| | | OSD 2
| |<---------+ Down
.. ditaa:: +---------+ +---------+ +---------+
| OSD 1 | | OSD 2 | | Monitor |
+---------+ +---------+ +---------+
| | |
| OSD 3 Is Down | |
|---------------+--------------->|
| | |
| | |
| | OSD 3 Is Down |
| |--------------->|
| | |
| | |
| | |---------+ Mark
| | | | OSD 3
| | |<--------+ Down
.. index:: peering failure
@ -118,13 +125,13 @@ setting the value at runtime.
+---------+ +---------+ +-------+ +---------+
| | | |
| Request To | | |
| Peer | | |
| Peer | | |
|-------------->| | |
|<--------------| | |
| Peering | |
| | |
| Request To | |
| Peer | |
| Peer | |
|----------------------------->| |
| |
|----+ OSD Monitor |
@ -135,7 +142,7 @@ setting the value at runtime.
|-------------------------------------------->|
|<--------------------------------------------|
| Receive New Cluster Map |
.. index:: OSD status
@ -149,10 +156,10 @@ event such as a failure, a change in placement group stats, a change in
``up_thru`` or when it boots within 5 seconds. You can change the Ceph OSD
Daemon minimum report interval by adding an ``osd mon report interval min``
setting under the ``[osd]`` section of your Ceph configuration file, or by
setting the value at runtime. A Ceph OSD Daemon sends a report to a Ceph
Monitor every 120 seconds irrespective of whether any notable changes occur.
You can change the Ceph Monitor report interval by adding an ``osd mon report
interval max`` setting under the ``[osd]`` section of your Ceph configuration
setting the value at runtime. A Ceph OSD Daemon sends a report to a Ceph
Monitor every 120 seconds irrespective of whether any notable changes occur.
You can change the Ceph Monitor report interval by adding an ``osd mon report
interval max`` setting under the ``[osd]`` section of your Ceph configuration
file, or by setting the value at runtime.
@ -207,18 +214,18 @@ Monitor Settings
``mon osd min up ratio``
:Description: The minimum ratio of ``up`` Ceph OSD Daemons before Ceph will
:Description: The minimum ratio of ``up`` Ceph OSD Daemons before Ceph will
mark Ceph OSD Daemons ``down``.
:Type: Double
:Default: ``.3``
``mon osd min in ratio``
:Description: The minimum ratio of ``in`` Ceph OSD Daemons before Ceph will
:Description: The minimum ratio of ``in`` Ceph OSD Daemons before Ceph will
mark Ceph OSD Daemons ``out``.
:Type: Double
:Default: ``.75``
@ -237,6 +244,15 @@ Monitor Settings
:Default: ``0.3``
``mon osd laggy max interval``
:Description: Maximum value of ``laggy_interval`` in laggy estimations (in seconds).
Monitor uses an adaptive approach to evaluate the ``laggy_interval`` of
a certain OSD. This value will be used to calculate the grace time for
that OSD.
:Type: Integer
:Default: 300
``mon osd adjust heartbeat grace``
:Description: If set to ``true``, Ceph will scale based on laggy estimations.
@ -251,38 +267,38 @@ Monitor Settings
:Default: ``true``
``mon osd auto mark in``
``mon osd auto mark in``
:Description: Ceph will mark any booting Ceph OSD Daemons as ``in``
:Description: Ceph will mark any booting Ceph OSD Daemons as ``in``
the Ceph Storage Cluster.
:Type: Boolean
:Default: ``false``
``mon osd auto mark auto out in``
``mon osd auto mark auto out in``
:Description: Ceph will mark booting Ceph OSD Daemons auto marked ``out``
:Description: Ceph will mark booting Ceph OSD Daemons auto marked ``out``
of the Ceph Storage Cluster as ``in`` the cluster.
:Type: Boolean
:Default: ``true``
:Default: ``true``
``mon osd auto mark new in``
``mon osd auto mark new in``
:Description: Ceph will mark booting new Ceph OSD Daemons as ``in`` the
:Description: Ceph will mark booting new Ceph OSD Daemons as ``in`` the
Ceph Storage Cluster.
:Type: Boolean
:Default: ``true``
:Default: ``true``
``mon osd down out interval``
``mon osd down out interval``
:Description: The number of seconds Ceph waits before marking a Ceph OSD Daemon
``down`` and ``out`` if it doesn't respond.
:Type: 32-bit Integer
:Default: ``600``
@ -298,21 +314,30 @@ Monitor Settings
:Default: ``rack``
``mon osd report timeout``
``mon osd report timeout``
:Description: The grace period in seconds before declaring
:Description: The grace period in seconds before declaring
unresponsive Ceph OSD Daemons ``down``.
:Type: 32-bit Integer
:Default: ``900``
``mon osd min down reporters``
``mon osd min down reporters``
:Description: The minimum number of Ceph OSD Daemons required to report a
:Description: The minimum number of Ceph OSD Daemons required to report a
``down`` Ceph OSD Daemon.
:Type: 32-bit Integer
:Default: ``1``
:Default: ``2``
``mon osd reporter subtree level``
:Description: In which level of parent bucket the reporters are counted. The OSDs
send failure reports to monitor if they find its peer is not responsive.
And monitor mark the reported OSD out and then down after a grace period.
:Type: String
:Default: ``host``
.. index:: OSD hearbeat
@ -322,63 +347,61 @@ OSD Settings
``osd heartbeat address``
:Description: An Ceph OSD Daemon's network address for heartbeats.
:Description: An Ceph OSD Daemon's network address for heartbeats.
:Type: Address
:Default: The host address.
``osd heartbeat interval``
``osd heartbeat interval``
:Description: How often an Ceph OSD Daemon pings its peers (in seconds).
:Type: 32-bit Integer
:Default: ``6``
``osd heartbeat grace``
``osd heartbeat grace``
:Description: The elapsed time when a Ceph OSD Daemon hasn't shown a heartbeat
that the Ceph Storage Cluster considers it ``down``.
This setting has to be set in both the [mon] and [osd] or [global]
section so that it is read by both the MON and OSD daemons.
:Type: 32-bit Integer
:Default: ``20``
``osd mon heartbeat interval``
``osd mon heartbeat interval``
:Description: How often the Ceph OSD Daemon pings a Ceph Monitor if it has no
:Description: How often the Ceph OSD Daemon pings a Ceph Monitor if it has no
Ceph OSD Daemon peers.
:Type: 32-bit Integer
:Default: ``30``
:Default: ``30``
``osd mon report interval max``
``osd mon report interval max``
:Description: The maximum time in seconds that a Ceph OSD Daemon can wait before
it must report to a Ceph Monitor.
:Type: 32-bit Integer
:Default: ``120``
:Default: ``120``
``osd mon report interval min``
``osd mon report interval min``
:Description: The minimum number of seconds a Ceph OSD Daemon may wait
from startup or another reportable event before reporting
from startup or another reportable event before reporting
to a Ceph Monitor.
:Type: 32-bit Integer
:Default: ``5``
:Valid Range: Should be less than ``osd mon report interval max``
:Valid Range: Should be less than ``osd mon report interval max``
``osd mon ack timeout``
``osd mon ack timeout``
:Description: The number of seconds to wait for a Ceph Monitor to acknowledge a
:Description: The number of seconds to wait for a Ceph Monitor to acknowledge a
request for statistics.
:Type: 32-bit Integer
:Default: ``30``
:Default: ``30``

22
ceph/doc/rados/configuration/osd-config-ref.rst
View File

@ -420,7 +420,7 @@ recovery operations to ensure optimal performance during recovery.
``osd client op priority``.
:Type: 32-bit Integer
:Default: ``10``
:Default: ``3``
:Valid Range: 1-63
@ -654,7 +654,7 @@ perform well in a degraded state.
increased load on the cluster.
:Type: 32-bit Integer
:Default: ``15``
:Default: ``3``
``osd recovery max chunk``
@ -664,6 +664,14 @@ perform well in a degraded state.
:Default: ``8 << 20``
``osd recovery max single start``
:Description: The maximum number of recovery operations per OSD that will be
newly started when an OSD is recovering.
:Type: 64-bit Integer Unsigned
:Default: ``1``
``osd recovery thread timeout``
:Description: The maximum time in seconds before timing out a recovery thread.
@ -679,6 +687,16 @@ perform well in a degraded state.
:Type: Boolean
:Default: ``true``
``osd recovery sleep``
:Description: Time to sleep before next recovery. Increasing this value will
slow down recovery operation while client operations will be
less impacted.
:Type: Float
:Default: ``0.01``
Tiering
=======

84
ceph/doc/rados/configuration/pool-pg-config-ref.rst
View File

@ -42,15 +42,95 @@ Ceph configuration file.
:Type: 32-bit Integer
:Default: ``300``
``mon pg min inactive``
``osd pg bits``
:Description: Issue a ``HEALTH_ERR`` in cluster log if the number of PGs stay
inactive longer than ``mon_pg_stuck_threshold`` exceeds this
setting. A non-positive number means disabled, never go into ERR.
:Type: Integer
:Default: ``1``
``mon pg warn min per osd``
:Description: Issue a ``HEALTH_WARN`` in cluster log if the average number
of PGs per (in) OSD is under this number. (a non-positive number
disables this)
:Type: Integer
:Default: ``30``
``mon pg warn max per osd``
:Description: Issue a ``HEALTH_WARN`` in cluster log if the average number
of PGs per (in) OSD is above this number. (a non-positive number
disables this)
:Type: Integer
:Default: ``300``
``mon pg warn min objects``
:Description: Do not warn if the total number of objects in cluster is below
this number
:Type: Integer
:Default: ``1000``
``mon pg warn min pool objects``
:Description: Do not warn on pools whose object number is below this number
:Type: Integer
:Default: ``1000``
``mon pg check down all threshold``
:Description: Threshold of down OSDs percentage after which we check all PGs
for stale ones.
:Type: Float
:Default: ``0.5``
``mon pg warn max object skew``
:Description: Issue a ``HEALTH_WARN`` in cluster log if the average object number
of a certain pool is greater than ``mon pg warn max object skew`` times
the average object number of the whole pool. (a non-positive number
disables this)
:Type: Float
:Default: ``10``
``mon delta reset interval``
:Description: Seconds of inactivity before we reset the pg delta to 0. We keep
track of the delta of the used space of each pool, so, for
example, it would be easier for us to understand the progress of
recovery or the performance of cache tier. But if there's no
activity reported for a certain pool, we just reset the history of
deltas of that pool.
:Type: Integer
:Default: ``10``
``mon osd max op age``
:Description: Maximum op age before we get concerned (make it a power of 2).
A ``HEALTH_WARN`` will be issued if a request has been blocked longer
than this limit.
:Type: Float
:Default: ``32.0``
``osd pg bits``
:Description: Placement group bits per Ceph OSD Daemon.
:Type: 32-bit Integer
:Default: ``6``
``osd pgp bits``
``osd pgp bits``
:Description: The number of bits per Ceph OSD Daemon for PGPs.
:Type: 32-bit Integer

2
ceph/doc/rados/operations/crush-map.rst
View File

@ -1257,4 +1257,4 @@ Further, as noted above, be careful running old versions of the
``ceph-osd`` daemon after reverting to legacy values as the feature
bit is not perfectly enforced.
.. _CRUSH - Controlled, Scalable, Decentralized Placement of Replicated Data: http://ceph.com/papers/weil-crush-sc06.pdf
.. _CRUSH - Controlled, Scalable, Decentralized Placement of Replicated Data: https://ceph.com/wp-content/uploads/2016/08/weil-crush-sc06.pdf

2
ceph/doc/rados/operations/erasure-code-jerasure.rst
View File

@ -6,7 +6,7 @@ The *jerasure* plugin is the most generic and flexible plugin, it is
also the default for Ceph erasure coded pools.
The *jerasure* plugin encapsulates the `Jerasure
<https://bitbucket.org/jimplank/jerasure/>`_ library. It is
<http://jerasure.org>`_ library. It is
recommended to read the *jerasure* documentation to get a better
understanding of the parameters.

2
ceph/doc/rados/operations/erasure-code.rst
View File

@ -139,7 +139,7 @@ erasure coded pool as the ``--data-pool`` during image creation::
rbd create --size 1G --data-pool ec_pool replicated_pool/image_name
For Cephfs, using an erasure coded pool means setting that pool in
a `file layout<../../cephfs/file-layouts>`_.
a `file layout <../../../cephfs/file-layouts>`_.
Erasure coded pool and cache tiering

1
ceph/doc/rados/operations/user-management.rst
View File

@ -360,6 +360,7 @@ are often restricted to accessing a particular pool. ::
pools in the cluster!
.. _modify-user-capabilities:
Modify User Capabilities
------------------------

2
ceph/doc/rados/troubleshooting/memory-profiling.rst
View File

@ -139,4 +139,4 @@ For example::
ceph tell osd.0 heap stop_profiler
.. _Logging and Debugging: ../log-and-debug
.. _Google Heap Profiler: http://google-perftools.googlecode.com/svn/trunk/doc/heapprofile.html
.. _Google Heap Profiler: http://goog-perftools.sourceforge.net/doc/heap_profiler.html

50
ceph/doc/rados/troubleshooting/troubleshooting-osd.rst
View File

@ -417,7 +417,57 @@ Possible solutions
- Upgrade Ceph
- Restart OSDs
Debugging Slow Requests
-----------------------
If you run "ceph daemon osd.<id> dump_historic_ops" or "dump_ops_in_flight",
you will see a set of operations and a list of events each operation went
through. These are briefly described below.
Events from the Messenger layer:
- header_read: when the messenger first started reading the message off the wire
- throttled: when the messenger tried to acquire memory throttle space to read
the message into memory
- all_read: when the messenger finished reading the message off the wire
- dispatched: when the messenger gave the message to the OSD
- Initiated: <This is identical to header_read. The existence of both is a
historical oddity.
Events from the OSD as it prepares operations
- queued_for_pg: the op has been put into the queue for processing by its PG
- reached_pg: the PG has started doing the op
- waiting for *: the op is waiting for some other work to complete before it
can proceed (a new OSDMap; for its object target to scrub; for the PG to
finish peering; all as specified in the message)
- started: the op has been accepted as something the OSD should actually do
(reasons not to do it: failed security/permission checks; out-of-date local
state; etc) and is now actually being performed
- waiting for subops from: the op has been sent to replica OSDs
Events from the FileStore
- commit_queued_for_journal_write: the op has been given to the FileStore
- write_thread_in_journal_buffer: the op is in the journal's buffer and waiting
to be persisted (as the next disk write)
- journaled_completion_queued: the op was journaled to disk and its callback
queued for invocation
Events from the OSD after stuff has been given to local disk
- op_commit: the op has been committed (ie, written to journal) by the
primary OSD
- op_applied: The op has been write()'en to the backing FS (ie, applied in
memory but not flushed out to disk) on the primary
- sub_op_applied: op_applied, but for a replica's "subop"
- sub_op_committed: op_commited, but for a replica's subop (only for EC pools)
- sub_op_commit_rec/sub_op_apply_rec from <X>: the primary marks this when it
hears about the above, but for a particular replica <X>
- commit_sent: we sent a reply back to the client (or primary OSD, for sub ops)
Many of these events are seemingly redundant, but cross important boundaries in
the internal code (such as passing data across locks into new threads).
Flapping OSDs
=============

14
ceph/doc/radosgw/admin.rst
View File

@ -438,6 +438,20 @@ new user, and that quota is enabled. See ``rgw bucket default quota max objects
``rgw bucket default quota max size``, ``rgw user default quota max objects``, and
``rgw user default quota max size`` in `Ceph Object Gateway Config Reference`_
Quota Cache
-----------
Quota statistics are cached on each RGW instance. If there are multiple
instances, then the cache can keep quotas from being perfectly enforced, as
each instance will have a different view of quotas. The options that control
this are ``rgw bucket quota ttl``, ``rgw user quota bucket sync interval`` and
``rgw user quota sync interval``. The higher these values are, the more
efficient quota operations are, but the more out-of-sync multiple instances
will be. The lower these values are, the closer to perfect enforcement
multiple instances will achieve. If all three are 0, then quota caching is
effectively disabled, and multiple instances will have perfect quota
enforcement. See `Ceph Object Gateway Config Reference`_
Reading / Writing Global Quotas
-------------------------------

5
ceph/doc/radosgw/adminops.rst
View File

@ -1818,11 +1818,6 @@ Special Error Responses
:Description: User does not possess specified capability.
:Code: 404 Not Found
Special Error Responses
~~~~~~~~~~~~~~~~~~~~~~~
None.
Quotas
======

3
ceph/doc/radosgw/barbican.rst
View File

@ -67,7 +67,8 @@ In the response, ``d1e7ef3b-f841-4b7c-90b2-b7d90ca2d723`` is the key id that
can be used in any `SSE-KMS`_ request.
This newly created key is not accessible by user ``rgwcrypt-user``. This
privilege must be added with an ACL.
privilege must be added with an ACL. See `How to Set/Replace ACL`_ for more
details.
Example request (assuming that the Keystone id of ``rgwcrypt-user`` is
``906aa90bd8a946c89cdff80d0869460f``)::

133
ceph/doc/radosgw/bucketpolicy.rst Normal file
View File

@ -0,0 +1,133 @@
===============
Bucket Policies
===============
.. versionadded:: Luminous
The Ceph Object Gateway supports a subset of the Amazon S3 policy
language applied to buckets.
Creation and Removal
====================
Bucket policies are managed through standard S3 operations rather than
radosgw-admin.
For example, one may use s3cmd to set or delete a policy thus::
$ cat > examplepol
{
"Version": "2012-10-17",
"Statement": [{
"Effect": "Allow",
"Principal": {"AWS": ["arn:aws:iam::usfolks:user/fred"]},
"Action": "s3PutObjectAcl",
"Resource": [
"arn:aws:s3:::happybucket/*"
]
}]
}
$ s3cmd setpolicy examplepol s3://happybucket
$ s3cmd delpolicy s3://happybucket
Limitations
===========
Currently, we support only the following actions:
- s3:AbortMultipartUpload
- s3:CreateBucket
- s3:DeleteBucketPolicy
- s3:DeleteBucket
- s3:DeleteBucketWebsite
- s3:DeleteObject
- s3:DeleteObjectVersion
- s3:DeleteReplicationConfiguration
- s3:GetAccelerateConfiguration
- s3:GetBucketAcl
- s3:GetBucketCORS
- s3:GetBucketLocation
- s3:GetBucketLogging
- s3:GetBucketNotification
- s3:GetBucketPolicy
- s3:GetBucketRequestPayment
- s3:GetBucketTagging
- s3:GetBucketVersioning
- s3:GetBucketWebsite
- s3:GetLifecycleConfiguration
- s3:GetObjectAcl
- s3:GetObject
- s3:GetObjectTorrent
- s3:GetObjectVersionAcl
- s3:GetObjectVersion
- s3:GetObjectVersionTorrent
- s3:GetReplicationConfiguration
- s3:ListAllMyBuckets
- s3:ListBucketMultiPartUploads
- s3:ListBucket
- s3:ListBucketVersions
- s3:ListMultipartUploadParts
- s3:PutAccelerateConfiguration
- s3:PutBucketAcl
- s3:PutBucketCORS
- s3:PutBucketLogging
- s3:PutBucketNotification
- s3:PutBucketPolicy
- s3:PutBucketRequestPayment
- s3:PutBucketTagging
- s3:PutBucketVersioning
- s3:PutBucketWebsite
- s3:PutLifecycleConfiguration
- s3:PutObjectAcl
- s3:PutObject
- s3:PutObjectVersionAcl
- s3:PutReplicationConfiguration
- s3:RestoreObject
We do not yet support setting policies on users, groups, or roles.
We use the RGW ‘tenant’ identifier in place of the Amazon twelve-digit
account ID. In the future we may allow you to assign an account ID to
a tenant, but for now if you want to use policies between AWS S3 and
RGW S3 you will have to use the Amazon account ID as the tenant ID when
creating users.
Under AWS, all tenants share a single namespace. RGW gives every
tenant its own namespace of buckets. There may be an option to enable
an AWS-like 'flat' bucket namespace in future versions. At present, to
access a bucket belonging to another tenant, address it as
"tenant:bucket" in the S3 request.
In AWS, a bucket policy can grant access to another account, and that
account owner can then grant access to individual users with user
permissions. Since we do not yet support user, role, and group
permissions, account owners will currently need to grant access
directly to individual users, and granting an entire account access to
a bucket grants access to all users in that account.
Bucket policies do not yet support string interpolation.
Currently, the only condition keys we support are:
- aws:CurrentTime
- aws:EpochTime
- aws:PrincipalType
- aws:Referer
- aws:SecureTransport
- aws:SourceIp
- aws:UserAgent
- aws:username
More may be supported soon as we integrate with the recently rewritten
Authentication/Authorization subsystem.
Swift
=====
There is no way to set bucket policies under Swift, but bucket
policies that have been set govern Swift as well as S3 operations.
Swift credentials are matched against Principals specified in a policy
in a way specific to whatever backend is being used.

595
ceph/doc/radosgw/config-fcgi.rst
View File

@ -1,595 +0,0 @@
=====================================================
Configuring Ceph Object Gateway with Apache/FastCGI
=====================================================
Configuring a Ceph Object Gateway requires a running Ceph Storage Cluster.
Since it contains an embedded webserver (civetweb), the Ceph Object Gateway
does not require an external web server, but it can be configured to use
Apache with the FastCGI module.
.. note:: CGI can pose a security risk.
The Ceph Object Gateway is a client of the Ceph Storage Cluster. As a
Ceph Storage Cluster client, it requires:
- A name for the gateway instance. We use ``gateway`` in this guide.
- A storage cluster user name with appropriate permissions in a keyring.
- Pools to store its data.
- A data directory for the gateway instance.
- An instance entry in the Ceph Configuration file.
- A configuration file for the web server to interact with FastCGI.
Create a User and Keyring
=========================
Each instance must have a user name and key to communicate with a Ceph Storage
Cluster. In the following steps, we use an admin node to create a keyring.
Then, we create a client user name and key. Next, we add the
key to the Ceph Storage Cluster. Finally, we distribute the key ring to
the node containing the gateway instance.
.. topic:: Monitor Key CAPS
When you provide CAPS to the key, you MUST provide read capability.
However, you have the option of providing write capability for the monitor.
This is an important choice. If you provide write capability to the key,
the Ceph Object Gateway will have the ability to create pools automatically;
however, it will create pools with either the default number of placement
groups (not ideal) or the number of placement groups you specified in your
Ceph configuration file. If you allow the Ceph Object Gateway to create
pools automatically, ensure that you have reasonable defaults for the number
of placement groups first. See `Pool Configuration`_ for details.
See `User Management`_ for additional details on Ceph authentication.
#. Generate a Ceph Object Gateway user name and key for each instance. For
exemplary purposes, we will use the name ``gateway`` after ``client.radosgw``::
sudo ceph auth get-or-create client.radosgw.gateway osd 'allow rwx' mon 'allow rwx' -o /etc/ceph/ceph.client.radosgw.keyring
#. Distribute the keyring to the node with the gateway instance. ::
sudo scp /etc/ceph/ceph.client.radosgw.keyring ceph@{hostname}:/home/ceph
ssh {hostname}
sudo mv ceph.client.radosgw.keyring /etc/ceph/ceph.client.radosgw.keyring
.. note:: The 2nd step is optional if ``admin node`` is the ``gateway host``.
Create Pools
============
Ceph Object Gateways require Ceph Storage Cluster pools to store specific
gateway data. If the user you created has permissions, the gateway
will create the pools automatically. However, you should ensure that you have
set an appropriate default number of placement groups per pool into your Ceph
configuration file.
.. note:: Ceph Object Gateways have multiple pools, so don't make the number of
PGs too high considering all of the pools assigned to the same CRUSH
hierarchy, or performance may suffer.
When configuring a gateway with the default region and zone, the naming
convention for pools typically omits region and zone naming, but you can use any
naming convention you prefer. For example:
- ``.rgw.root``
- ``.rgw.control``
- ``.rgw.gc``
- ``.rgw.buckets``
- ``.rgw.buckets.index``
- ``.rgw.buckets.extra``
- ``.log``
- ``.intent-log``
- ``.usage``
- ``.users``
- ``.users.email``
- ``.users.swift``
- ``.users.uid``
See `Configuration Reference - Pools`_ for details on the default pools for
gateways. See `Pools`_ for details on creating pools. As already said, if
write permission is given, Ceph Object Gateway will create pools automatically.
To create a pool manually, execute the following::
ceph osd pool create {poolname} {pg-num} {pgp-num} {replicated | erasure} [{erasure-code-profile}] {ruleset-name} {ruleset-number}
.. tip:: Ceph supports multiple CRUSH hierarchies and CRUSH rulesets, enabling
great flexibility in the way you configure your gateway. Pools such as
``rgw.buckets.index`` may benefit from a pool of SSDs for fast performance.
Backing storage may benefit from the increased economy of erasure-coded
storage, and/or the improved performance from cache tiering.
When you have completed this step, execute the following to ensure that
you have created all of the foregoing pools::
rados lspools
Add a Gateway Configuration to Ceph
===================================
Add the Ceph Object Gateway configuration to your Ceph Configuration file in
``admin node``. The Ceph Object Gateway configuration requires you to
identify the Ceph Object Gateway instance. Then, you must specify the host name
where you installed the Ceph Object Gateway daemon, a keyring (for use with
cephx), the socket path for FastCGI and a log file.
For distros with Apache 2.2 and early versions of Apache 2.4 (RHEL 6, Ubuntu
12.04, 14.04 etc), append the following configuration to ``/etc/ceph/ceph.conf``
in your ``admin node``::
[client.radosgw.gateway]
host = {hostname}
keyring = /etc/ceph/ceph.client.radosgw.keyring
rgw socket path = ""
log file = /var/log/radosgw/client.radosgw.gateway.log
rgw frontends = fastcgi socket_port=9000 socket_host=0.0.0.0
rgw print continue = false
.. note:: Apache 2.2 and early versions of Apache 2.4 do not use Unix Domain
Sockets but use localhost TCP.
For distros with Apache 2.4.9 or later (RHEL 7, CentOS 7 etc), append the
following configuration to ``/etc/ceph/ceph.conf`` in your ``admin node``::
[client.radosgw.gateway]
host = {hostname}
keyring = /etc/ceph/ceph.client.radosgw.keyring
rgw socket path = /var/run/ceph/ceph.radosgw.gateway.fastcgi.sock
log file = /var/log/radosgw/client.radosgw.gateway.log
rgw print continue = false
.. note:: ``Apache 2.4.9`` supports Unix Domain Socket (UDS) but as
``Ubuntu 14.04`` ships with ``Apache 2.4.7`` it doesn't have UDS support and
has to be configured for use with localhost TCP. A bug has been filed for
backporting UDS support in ``Apache 2.4.7`` for ``Ubuntu 14.04``.
See: `Backport support for UDS in Ubuntu Trusty`_
Here, ``{hostname}`` is the short hostname (output of command ``hostname -s``)
of the node that is going to provide the gateway service i.e., the
``gateway host``.
The ``[client.radosgw.gateway]`` portion of the gateway instance identifies this
portion of the Ceph configuration file as configuring a Ceph Storage Cluster
client where the client type is a Ceph Object Gateway (i.e., ``radosgw``).
.. note:: The last line in the configuration i.e., ``rgw print continue = false``
is added to avoid issues with ``PUT`` operations.
Once you finish the setup procedure, if you encounter issues with your
configuration, you can add debugging to the ``[global]`` section of your Ceph
configuration file and restart the gateway to help troubleshoot any
configuration issues. For example::
[global]
#append the following in the global section.
debug ms = 1
debug rgw = 20
Distribute updated Ceph configuration file
==========================================
The updated Ceph configuration file needs to be distributed to all Ceph cluster
nodes from the ``admin node``.
It involves the following steps:
#. Pull the updated ``ceph.conf`` from ``/etc/ceph/`` to the root directory of
the cluster in admin node (e.g. ``my-cluster`` directory). The contents of
``ceph.conf`` in ``my-cluster`` will get overwritten. To do so, execute the
following::
ceph-deploy --overwrite-conf config pull {hostname}
Here, ``{hostname}`` is the short hostname of the Ceph admin node.
#. Push the updated ``ceph.conf`` file from the admin node to all other nodes in
the cluster including the ``gateway host``::
ceph-deploy --overwrite-conf config push [HOST] [HOST...]
Give the hostnames of the other Ceph nodes in place of ``[HOST] [HOST...]``.
Copy ceph.client.admin.keyring from admin node to gateway host
==============================================================
As the ``gateway host`` can be a different node that is not part of the cluster,
the ``ceph.client.admin.keyring`` needs to be copied from the ``admin node`` to
the ``gateway host``. To do so, execute the following on ``admin node``::
sudo scp /etc/ceph/ceph.client.admin.keyring ceph@{hostname}:/home/ceph
ssh {hostname}
sudo mv ceph.client.admin.keyring /etc/ceph/ceph.client.admin.keyring
.. note:: The above step need not be executed if ``admin node`` is the
``gateway host``.
Create Data Directory
=====================
Deployment scripts may not create the default Ceph Object Gateway data
directory. Create data directories for each instance of a ``radosgw``
daemon (if you haven't done so already). The ``host`` variables in the
Ceph configuration file determine which host runs each instance of a
``radosgw`` daemon. The typical form specifies the ``radosgw`` daemon,
the cluster name and the daemon ID.
To create the directory on the ``gateway host``, execute the following::
sudo mkdir -p /var/lib/ceph/radosgw/ceph-radosgw.gateway
Adjust Socket Directory Permissions
===================================
On some distros, the ``radosgw`` daemon runs as the unprivileged ``apache``
UID, and this UID must have write access to the location where it will write
its socket file.
To grant permissions to the default socket location, execute the following on
the ``gateway host``::
sudo chown apache:apache /var/run/ceph
Change Log File Owner
=====================
On some distros, the ``radosgw`` daemon runs as the unprivileged ``apache`` UID,
but the ``root`` user owns the log file by default. You must change it to the
``apache`` user so that Apache can populate the log file. To do so, execute
the following::
sudo chown apache:apache /var/log/radosgw/client.radosgw.gateway.log
Start radosgw service
=====================
The Ceph Object gateway daemon needs to be started. To do so, execute the
following on the ``gateway host``:
On Debian-based distros::
sudo /etc/init.d/radosgw start
On RPM-based distros::
sudo /etc/init.d/ceph-radosgw start
Create a Gateway Configuration file
===================================
On the host where you installed the Ceph Object Gateway i.e., ``gateway host``,
create an ``rgw.conf`` file. Place the file in ``/etc/apache2/conf-available``
directory for ``Debian-based`` distros and in ``/etc/httpd/conf.d`` directory
for ``RPM-based`` distros. It is a Apache configuration file which is needed
for the ``radosgw`` service. This file must be readable by the web server.
Execute the following steps:
#. Create the file:
For Debian-based distros, execute::
sudo vi /etc/apache2/conf-available/rgw.conf
For RPM-based distros, execute::
sudo vi /etc/httpd/conf.d/rgw.conf
#. For distros with Apache 2.2 and early versions of Apache 2.4 that use
localhost TCP and do not support Unix Domain Socket, add the following
contents to the file::
<VirtualHost *:80>
ServerName localhost
DocumentRoot /var/www/html
ErrorLog /var/log/httpd/rgw_error.log
CustomLog /var/log/httpd/rgw_access.log combined
# LogLevel debug
RewriteEngine On
RewriteRule .* - [E=HTTP_AUTHORIZATION:%{HTTP:Authorization},L]
SetEnv proxy-nokeepalive 1
ProxyPass / fcgi://localhost:9000/
</VirtualHost>
.. note:: For Debian-based distros replace ``/var/log/httpd/``
with ``/var/log/apache2``.
#. For distros with Apache 2.4.9 or later that support Unix Domain Socket,
add the following contents to the file::
<VirtualHost *:80>
ServerName localhost
DocumentRoot /var/www/html
ErrorLog /var/log/httpd/rgw_error.log
CustomLog /var/log/httpd/rgw_access.log combined
# LogLevel debug
RewriteEngine On
RewriteRule .* - [E=HTTP_AUTHORIZATION:%{HTTP:Authorization},L]
SetEnv proxy-nokeepalive 1
ProxyPass / unix:///var/run/ceph/ceph.radosgw.gateway.fastcgi.sock|fcgi://localhost:9000/
</VirtualHost>
Restart Apache
==============
The Apache service needs to be restarted to accept the new configuration.
For Debian-based distros, run::
sudo service apache2 restart
For RPM-based distros, run::
sudo service httpd restart
Or::
sudo systemctl restart httpd
Using The Gateway
=================
To use the REST interfaces, first create an initial Ceph Object Gateway
user for the S3 interface. Then, create a subuser for the Swift interface.
See the `Admin Guide`_ for more details on user management.
Create a radosgw user for S3 access
------------------------------------
A ``radosgw`` user needs to be created and granted access. The command
``man radosgw-admin`` will provide information on additional command options.
To create the user, execute the following on the ``gateway host``::
sudo radosgw-admin user create --uid="testuser" --display-name="First User"
The output of the command will be something like the following::
{"user_id": "testuser",
"display_name": "First User",
"email": "",
"suspended": 0,
"max_buckets": 1000,
"auid": 0,
"subusers": [],
"keys": [
{ "user": "testuser",
"access_key": "I0PJDPCIYZ665MW88W9R",
"secret_key": "dxaXZ8U90SXydYzyS5ivamEP20hkLSUViiaR+ZDA"}],
"swift_keys": [],
"caps": [],
"op_mask": "read, write, delete",
"default_placement": "",
"placement_tags": [],
"bucket_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"user_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"temp_url_keys": []}
.. note:: The values of ``keys->access_key`` and ``keys->secret_key`` are
needed for access validation.
Create a Swift user
-------------------
A Swift subuser needs to be created if this kind of access is needed. Creating
a Swift user is a two step process. The first step is to create the user.
The second is to create the secret key.
Execute the following steps on the ``gateway host``:
Create the Swift user::
sudo radosgw-admin subuser create --uid=testuser --subuser=testuser:swift --access=full
The output will be something like the following::
{ "user_id": "testuser",
"display_name": "First User",
"email": "",
"suspended": 0,
"max_buckets": 1000,
"auid": 0,
"subusers": [
{ "id": "testuser:swift",
"permissions": "full-control"}],
"keys": [
{ "user": "testuser:swift",
"access_key": "3Y1LNW4Q6X0Y53A52DET",
"secret_key": ""},
{ "user": "testuser",
"access_key": "I0PJDPCIYZ665MW88W9R",
"secret_key": "dxaXZ8U90SXydYzyS5ivamEP20hkLSUViiaR+ZDA"}],
"swift_keys": [],
"caps": [],
"op_mask": "read, write, delete",
"default_placement": "",
"placement_tags": [],
"bucket_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"user_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"temp_url_keys": []}
Create the secret key::
sudo radosgw-admin key create --subuser=testuser:swift --key-type=swift --gen-secret
The output will be something like the following::
{ "user_id": "testuser",
"display_name": "First User",
"email": "",
"suspended": 0,
"max_buckets": 1000,
"auid": 0,
"subusers": [
{ "id": "testuser:swift",
"permissions": "full-control"}],
"keys": [
{ "user": "testuser:swift",
"access_key": "3Y1LNW4Q6X0Y53A52DET",
"secret_key": ""},
{ "user": "testuser",
"access_key": "I0PJDPCIYZ665MW88W9R",
"secret_key": "dxaXZ8U90SXydYzyS5ivamEP20hkLSUViiaR+ZDA"}],
"swift_keys": [
{ "user": "testuser:swift",
"secret_key": "244+fz2gSqoHwR3lYtSbIyomyPHf3i7rgSJrF\/IA"}],
"caps": [],
"op_mask": "read, write, delete",
"default_placement": "",
"placement_tags": [],
"bucket_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"user_quota": { "enabled": false,
"max_size_kb": -1,
"max_objects": -1},
"temp_url_keys": []}
Access Verification
===================
You then need to verify if the created users are able to access the gateway.
Test S3 access
--------------
You need to write and run a Python test script for verifying S3 access. The S3
access test script will connect to the ``radosgw``, create a new bucket and list
all buckets. The values for ``aws_access_key_id`` and ``aws_secret_access_key``
are taken from the values of ``access_key`` and ``secret_key`` returned by the
``radosgw_admin`` command.
Execute the following steps:
#. You will need to install the ``python-boto`` package.
For Debian-based distros, run::
sudo apt-get install python-boto
For RPM-based distros, run::
sudo yum install python-boto
#. Create the Python script::
vi s3test.py
#. Add the following contents to the file::
import boto
import boto.s3.connection
access_key = 'I0PJDPCIYZ665MW88W9R'
secret_key = 'dxaXZ8U90SXydYzyS5ivamEP20hkLSUViiaR+ZDA'
conn = boto.connect_s3(
aws_access_key_id = access_key,
aws_secret_access_key = secret_key,
host = '{hostname}',
is_secure=False,
calling_format = boto.s3.connection.OrdinaryCallingFormat(),
)
bucket = conn.create_bucket('my-new-bucket')
for bucket in conn.get_all_buckets():
print "{name}\t{created}".format(
name = bucket.name,
created = bucket.creation_date,
)
Replace ``{hostname}`` with the hostname of the host where you have
configured the gateway service i.e., the ``gateway host``.
#. Run the script::
python s3test.py
The output will be something like the following::
my-new-bucket 2015-02-16T17:09:10.000Z
Test swift access
-----------------
Swift access can be verified via the ``swift`` command line client. The command
``man swift`` will provide more information on available command line options.
To install ``swift`` client, execute the following:
For Debian-based distros::
sudo apt-get install python-setuptools
sudo easy_install pip
sudo pip install --upgrade setuptools
sudo pip install --upgrade python-swiftclient
For RPM-based distros::
sudo yum install python-setuptools
sudo easy_install pip
sudo pip install --upgrade setuptools
sudo pip install --upgrade python-swiftclient
To test swift access, execute the following::
swift -A http://{IP ADDRESS}/auth/1.0 -U testuser:swift -K ‘{swift_secret_key}’ list
Replace ``{IP ADDRESS}`` with the public IP address of the gateway server and
``{swift_secret_key}`` with its value from the output of
``radosgw-admin key create`` command executed for the ``swift`` user.
For example::
swift -A http://10.19.143.116/auth/1.0 -U testuser:swift -K ‘244+fz2gSqoHwR3lYtSbIyomyPHf3i7rgSJrF/IA’ list
The output should be::
my-new-bucket
.. _Configuration Reference - Pools: ../config-ref#pools
.. _Pool Configuration: ../../rados/configuration/pool-pg-config-ref/
.. _Pools: ../../rados/operations/pools
.. _User Management: ../../rados/operations/user-management
.. _Backport support for UDS in Ubuntu Trusty: https://bugs.launchpad.net/ubuntu/+source/apache2/+bug/1411030
.. _Admin Guide: ../admin

491
ceph/doc/radosgw/config-ref.rst
View File

@ -7,6 +7,11 @@ The following settings may added to the Ceph configuration file (i.e., usually
settings may contain default values. If you do not specify each setting in the
Ceph configuration file, the default value will be set automatically.
Configuration variables set under the ``[client.radosgw.{instance-name}]``
section will not apply to rgw or radosgw-admin commands without an instance-name
specified in the command. Thus variables meant to be applied to all RGW
instances or all radosgw-admin commands can be put into the ``[global]`` or the
``[client]`` section to avoid specifying instance-name.
``rgw data``
@ -294,29 +299,13 @@ Ceph configuration file, the default value will be set automatically.
a value of zero indicates there is no sharding. It is not
recommended to set a value too large (e.g. thousand) as it
increases the cost for bucket listing.
This variable should be set in the client or global sections
so that it is automatically applied to radosgw-admin commands.
:Type: Integer
:Default: ``0``
``rgw num zone opstate shards``
:Description: The maximum number of shards for keeping inter-region copy
progress information.
:Type: Integer
:Default: ``128``
``rgw opstate ratelimit sec``
:Description: The minimum time between opstate updates on a single upload.
``0`` disables the ratelimit.
:Type: Integer
:Default: ``30``
``rgw curl wait timeout ms``
:Description: The timeout in milliseconds for certain ``curl`` calls.
@ -352,10 +341,41 @@ Ceph configuration file, the default value will be set automatically.
:Default: ``false``
``rgw bucket quota ttl``
:Description: The amount of time in seconds cached quota information is
trusted. After this timeout, the quota information will be
re-fetched from the cluster.
:Type: Integer
:Default: ``600``
``rgw user quota bucket sync interval``
:Description: The amount of time in seconds bucket quota information is
accumulated before syncing to the cluster. During this time,
other RGW instances will not see the changes in bucket quota
stats from operations on this instance.
:Type: Integer
:Default: ``180``
``rgw user quota sync interval``
:Description: The amount of time in seconds user quota information is
accumulated before syncing to the cluster. During this time,
other RGW instances will not see the changes in user quota stats
from operations on this instance.
:Type: Integer
:Default: ``180``
``rgw bucket default quota max objects``
:Description: Default max number of objects per bucket. Set on new users,
if no other quota is specified. Has no effect on existing users.
This variable should be set in the client or global sections
so that it is automatically applied to radosgw-admin commands.
:Type: Integer
:Default: ``-1``
@ -385,389 +405,96 @@ Ceph configuration file, the default value will be set automatically.
:Default: ``-1``
Regions
=======
``rgw verify ssl``
In Ceph v0.67 and beyond, Ceph Object Gateway supports federated deployments and
a global namespace via the notion of regions. A region defines the geographic
location of one or more Ceph Object Gateway instances within one or more zones.
:Description: Verify SSL certificates while making requests.
:Type: Boolean
:Default: ``true``
Configuring regions differs from typical configuration procedures, because not
all of the settings end up in a Ceph configuration file. In Ceph v0.67 and
beyond, you can list regions, get a region configuration and set a region
configuration.
Multisite Settings
==================
List Regions
------------
A Ceph cluster contains a list of regions. To list the regions, execute::
sudo radosgw-admin region list
The ``radosgw-admin`` returns a JSON formatted list of regions.
.. code-block:: javascript
{ "default_info": { "default_region": "default"},
"regions": [
"default"]}
Get a Region Map
----------------
To list the details of each region, execute::
sudo radosgw-admin region-map get
.. note:: If you receive a ``failed to read region map`` error, run
``sudo radosgw-admin region-map update`` first.
Get a Region
------------
To view the configuration of a region, execute::
radosgw-admin region get [--rgw-region=<region>]
The ``default`` region looks like this:
.. code-block:: javascript
{"name": "default",
"api_name": "",
"is_master": "true",
"endpoints": [],
"hostnames": [],
"master_zone": "",
"zones": [
{"name": "default",
"endpoints": [],
"log_meta": "false",
"log_data": "false"}
],
"placement_targets": [
{"name": "default-placement",
"tags": [] }],
"default_placement": "default-placement"}
Set a Region
------------
Defining a region consists of creating a JSON object, specifying at least the
required settings:
#. ``name``: The name of the region. Required.
#. ``api_name``: The API name for the region. Optional.
#. ``is_master``: Determines if the region is the master region. Required.
**note:** You can only have one master region.
#. ``endpoints``: A list of all the endpoints in the region. For example,
you may use multiple domain names to refer to the same region. Remember to
escape the forward slashes (``\/``). You may also specify a
port (``fqdn:port``) for each endpoint. Optional.
#. ``hostnames``: A list of all the hostnames in the region. For example,
you may use multiple domain names to refer to the same region. Optional.
The ``rgw dns name`` setting will automatically be included in this list.
You should restart the ``radosgw`` daemon(s) after changing this setting.
#. ``master_zone``: The master zone for the region. Optional. Uses the default
zone if not specified. **note:** You can only have one master zone per
region.
#. ``zones``: A list of all zones within the region. Each zone has a
name (required), a list of endpoints (optional), and whether or not the
gateway will log metadata and data operations (false by default).
#. ``placement_targets``: A list of placement targets (optional). Each
placement target contains a name (required) for the placement target
and a list of tags (optional) so that only users with the tag can use
the placement target (i.e., the user's ``placement_tags`` field in the
user info).
#. ``default_placement``: The default placement target for the object
index and object data. Set to ``default-placement`` by default. You
may also set a per-user default placement in the user info for each
user.
To set a region, create a JSON object consisting of the required fields, save
the object to a file (e.g., ``region.json``); then, execute the following
command::
sudo radosgw-admin region set --infile region.json
Where ``region.json`` is the JSON file you created.
.. important:: The ``default`` region ``is_master`` setting is ``true`` by
default. If you create a new region and want to make it the master region,
you must either set the ``default`` region ``is_master`` setting to
``false``, or delete the ``default`` region.
Finally, update the map. ::
sudo radosgw-admin region-map update
Set a Region Map
----------------
Setting a region map consists of creating a JSON object consisting of one or more
regions, and setting the ``master_region`` for the cluster. Each region in the
region map consists of a key/value pair, where the ``key`` setting is equivalent to
the ``name`` setting for an individual region configuration, and the ``val`` is
a JSON object consisting of an individual region configuration.
You may only have one region with ``is_master`` equal to ``true``, and it must be
specified as the ``master_region`` at the end of the region map. The following
JSON object is an example of a default region map.
.. code-block:: javascript
{ "regions": [
{ "key": "default",
"val": { "name": "default",
"api_name": "",
"is_master": "true",
"endpoints": [],
"hostnames": [],
"master_zone": "",
"zones": [
{ "name": "default",
"endpoints": [],
"log_meta": "false",
"log_data": "false"}],
"placement_targets": [
{ "name": "default-placement",
"tags": []}],
"default_placement": "default-placement"
}
}
],
"master_region": "default"
}
To set a region map, execute the following::
sudo radosgw-admin region-map set --infile regionmap.json
Where ``regionmap.json`` is the JSON file you created. Ensure that you have
zones created for the ones specified in the region map. Finally, update the map.
::
sudo radosgw-admin regionmap update
Zones
=====
In Ceph v0.67 and beyond, Ceph Object Gateway supports the notion of zones. A
zone defines a logical group consisting of one or more Ceph Object Gateway
instances.
Configuring zones differs from typical configuration procedures, because not
all of the settings end up in a Ceph configuration file. In Ceph v0.67 and
beyond, you can list zones, get a zone configuration and set a zone
configuration.
List Zones
----------
To list the zones in a cluster, execute::
sudo radosgw-admin zone list
Get a Zone
----------
To get the configuration of a zone, execute::
sudo radosgw-admin zone get [--rgw-zone=<zone>]
The ``default`` zone looks like this:
.. code-block:: javascript
{ "domain_root": ".rgw",
"control_pool": ".rgw.control",
"gc_pool": ".rgw.gc",
"log_pool": ".log",
"intent_log_pool": ".intent-log",
"usage_log_pool": ".usage",
"user_keys_pool": ".users",
"user_email_pool": ".users.email",
"user_swift_pool": ".users.swift",
"user_uid_pool": ".users.uid",
"system_key": { "access_key": "", "secret_key": ""},
"placement_pools": [
{ "key": "default-placement",
"val": { "index_pool": ".rgw.buckets.index",
"data_pool": ".rgw.buckets"}
}
]
}
Set a Zone
----------
Configuring a zone involves specifying a series of Ceph Object Gateway pools.
For consistency, we recommend using a pool prefix that is
the same as the zone name. See `Pools`_ for details of configuring pools.
To set a zone, create a JSON object consisting of the pools, save
the object to a file (e.g., ``zone.json``); then, execute the following
command, replacing ``{zone-name}`` with the name of the zone::
sudo radosgw-admin zone set --rgw-zone={zone-name} --infile zone.json
Where ``zone.json`` is the JSON file you created.
Region/Zone Settings
====================
.. versionadded:: Jewel
You may include the following settings in your Ceph configuration
file under each ``[client.radosgw.{instance-name}]`` instance.
.. versionadded:: v.67
``rgw zone``
:Description: The name of the zone for the gateway instance.
:Description: The name of the zone for the gateway instance. If no zone is
set, a cluster-wide default can be configured with the command
``radosgw-admin zone default``.
:Type: String
:Default: None
.. versionadded:: v.67
``rgw zonegroup``
``rgw region``
:Description: The name of the region for the gateway instance.
:Description: The name of the zonegroup for the gateway instance. If no
zonegroup is set, a cluster-wide default can be configured with
the command ``radosgw-admin zonegroup default``.
:Type: String
:Default: None
.. versionadded:: v.67
``rgw realm``
``rgw default region info oid``
:Description: The OID for storing the default region. We do not recommend
changing this setting.
:Description: The name of the realm for the gateway instance. If no realm is
set, a cluster-wide default can be configured with the command
``radosgw-admin realm default``.
:Type: String
:Default: ``default.region``
:Default: None
``rgw run sync thread``
Pools
=====
Ceph zones map to a series of Ceph Storage Cluster pools.
.. topic:: Manually Created Pools vs. Generated Pools
If you provide write capabilities to the user key for your Ceph Object
Gateway, the gateway has the ability to create pools automatically. This
is convenient, but the Ceph Object Storage Cluster uses the default
values for the number of placement groups (which may not be ideal) or the
values you specified in your Ceph configuration file. If you allow the
Ceph Object Gateway to create pools automatically, ensure that you have
reasonable defaults for the number of placement groups. See
`Pool Configuration`_ for details. See `Cluster Pools`_ for details on
creating pools.
The default pools for the Ceph Object Gateway's default zone include:
- ``.rgw``
- ``.rgw.control``
- ``.rgw.gc``
- ``.log``
- ``.intent-log``
- ``.usage``
- ``.users``
- ``.users.email``
- ``.users.swift``
- ``.users.uid``
You have significant discretion in determining how you want a zone to access
pools. You can create pools on a per zone basis, or use the same pools for
multiple zones. As a best practice, we recommend having a separate set of pools
for your master zone and your secondary zones in each region. When creating
pools for a specific zone, consider prepending the region name and zone name to
the default pool names. For example:
- ``.region1-zone1.domain.rgw``
- ``.region1-zone1.rgw.control``
- ``.region1-zone1.rgw.gc``
- ``.region1-zone1.log``
- ``.region1-zone1.intent-log``
- ``.region1-zone1.usage``
- ``.region1-zone1.users``
- ``.region1-zone1.users.email``
- ``.region1-zone1.users.swift``
- ``.region1-zone1.users.uid``
:Description: If there are other zones in the realm to sync from, spawn threads
to handle the sync of data and metadata.
:Type: Boolean
:Default: ``true``
Ceph Object Gateways store data for the bucket index (``index_pool``) and bucket
data (``data_pool``) in placement pools. These may overlap--i.e., you may use
the same pool for the index and the data. The index pool for default
placement is ``.rgw.buckets.index`` and for the data pool for default placement
is ``.rgw.buckets``. See `Zones`_ for details on specifying pools in a zone
configuration.
``rgw data log window``
:Description: The data log entries window in seconds.
:Type: Integer
:Default: ``30``
.. deprecated:: v.67
``rgw data log changes size``
``rgw cluster root pool``
:Description: The number of in-memory entries to hold for the data changes log.
:Type: Integer
:Default: ``1000``
:Description: The Ceph Storage Cluster pool to store ``radosgw`` metadata for
this instance. Not used in Ceph version v.67 and later. Use
``rgw zone root pool`` instead.
``rgw data log obj prefix``
:Description: The object name prefix for the data log.
:Type: String
:Required: No
:Default: ``.rgw.root``
:Replaced By: ``rgw zone root pool``
:Default: ``data_log``
.. versionadded:: v.67
``rgw data log num shards``
``rgw region root pool``
:Description: The number of shards (objects) on which to keep the
data changes log.
:Description: The pool for storing all region-specific information.
Not used in Ceph version ``Jewel``.
:Type: String
:Default: ``.rgw.root``
.. versionadded:: Jewel
``rgw zonegroup root pool``
:Description: The pool for storing all zonegroup-specific information.
:Type: String
:Default: ``.rgw.root``
:Type: Integer
:Default: ``128``
.. versionadded:: v.67
``rgw md log max shards``
``rgw zone root pool``
:Description: The maximum number of shards for the metadata log.
:Type: Integer
:Default: ``64``
:Description: The pool for storing zone-specific information.
:Type: String
:Default: ``.rgw.root``
.. important:: The values of ``rgw data log num shards`` and
``rgw md log max shards`` should not be changed after sync has
started.
Swift Settings
@ -970,50 +697,6 @@ Logging Settings
:Default: ``false``
``rgw data log window``
:Description: The data log entries window in seconds.
:Type: Integer
:Default: ``30``
``rgw data log changes size``
:Description: The number of in-memory entries to hold for the data changes log.
:Type: Integer
:Default: ``1000``
``rgw data log num shards``
:Description: The number of shards (objects) on which to keep the
data changes log.
:Type: Integer
:Default: ``128``
``rgw data log obj prefix``
:Description: The object name prefix for the data log.
:Type: String
:Default: ``data_log``
``rgw replica log obj prefix``
:Description: The object name prefix for the replica log.
:Type: String
:Default: ``replica log``
``rgw md log max shards``
:Description: The maximum number of shards for the metadata log.
:Type: Integer
:Default: ``64``
Keystone Settings
=================

2
ceph/doc/radosgw/encryption.rst
View File

@ -25,7 +25,7 @@ Key Management Service
======================
This mode allows keys to be stored in a secure key management service and
retrieved on demand by the Ceph Object Gateway to service requests to encrypt
retrieved on demand by the Ceph Object Gateway to serve requests to encrypt
or decrypt data.
This is implemented in S3 according to the `Amazon SSE-KMS`_ specification.

825
ceph/doc/radosgw/federated-config.rst
View File

@ -1,825 +0,0 @@
================================
Configuring Federated Gateways
================================
.. versionadded:: 0.67 Dumpling
In Ceph version 0.67 Dumpling and beyond, you may configure each :term:`Ceph
Object Gateway` to participate in a federated architecture, with multiple
regions, and with multiple zones for a region.
- **Region**: A region represents a *logical* geographic area and contains one
or more zones. A cluster with multiple regions must specify a master region.
- **Zone**: A zone is a *logical* grouping of one or more Ceph Object Gateway
instance(s). A region has a master zone that processes client requests.
.. important:: Only write objects to the master zone in a region. You may read
objects from secondary zones. Currently, the Gateway does not prevent you
from writing to a secondary zone, but **DON'T DO IT**.
Background
==========
When you deploy a :term:`Ceph Object Store` service that spans geographical
locales, configuring Ceph Object Gateway regions and metadata synchronization
agents enables the service to maintain a global namespace, even though Ceph
Object Gateway instances run in different geographic locales and potentially on
different Ceph Storage Clusters. When you separate one or more Ceph Object
Gateway instances within a region into separate logical containers to maintain
an extra copy (or copies) of the data, configuring Ceph Object Gateway zones and
data synchronization agents enables the service to maintain one or more
copy(ies) of the master zone's data. Extra copies of the data are important for
failover, backup and disaster recovery.
You may deploy a single Ceph Storage Cluster with a federated architecture if
you have low latency network connections (this isn't recommended). You may also
deploy one Ceph Storage Cluster per region with a separate set of pools for
each zone (typical). You may also deploy a separate Ceph Storage Cluster for
each zone if your requirements and resources warrant this level of redundancy.
About this Guide
================
In the following sections, we will demonstrate how to configure a federated
cluster in two logical steps:
- **Configure a Master Region:** This section of the guide describes how to
set up a region with multiple zones, and how to synchronize data between the
master zone and the secondary zone(s) within the master region.
- **Configure a Secondary Region:** This section of the guide describes how
to repeat the section on setting up a master region and multiple zones so
that you have two regions with intra-zone synchronization in each region.
Finally, you will learn how to set up a metadata synchronization agent so
that you can maintain a global namespace for the regions in your cluster.
Configure a Master Region
=========================
This section provides an exemplary procedure for setting up a region, and two
zones within the region. The cluster will comprise two gateway daemon
instances--one per zone. This region will serve as the master region.
Naming for the Master Region
----------------------------
Before configuring the cluster, defining region, zone and instance names will
help you manage your cluster. Let's assume the region represents the United
States, and we refer to it by its standard abbreviation.
- United States: ``us``
Let's assume the zones represent the Eastern and Western United States. For
continuity, our naming convention will use ``{region name}-{zone name}`` format,
but you can use any naming convention you prefer.
- United States, East Region: ``us-east``
- United States, West Region: ``us-west``
Finally, let's assume that zones may have more than one Ceph Object Gateway
instance per zone. For continuity, our naming convention will use
``{region name}-{zone name}-{instance}`` format, but you can use any naming
convention you prefer.
- United States Region, Master Zone, Instance 1: ``us-east-1``
- United States Region, Secondary Zone, Instance 1: ``us-west-1``
Create Pools
------------
You may have a Ceph Storage Cluster for the entire region or a Ceph Storage
Cluster for each zone.
For continuity, our naming convention will use ``{region name}-{zone name}``
format prepended to the pool name, but you can use any naming convention you
prefer. For example:
- ``.us-east.rgw.root``
- ``.us-east.rgw.control``
- ``.us-east.rgw.gc``
- ``.us-east.rgw.buckets``
- ``.us-east.rgw.buckets.index``
- ``.us-east.rgw.buckets.extra``
- ``.us-east.log``
- ``.us-east.intent-log``
- ``.us-east.usage``
- ``.us-east.users``
- ``.us-east.users.email``
- ``.us-east.users.swift``
- ``.us-east.users.uid``
|
- ``.us-west.rgw.root``
- ``.us-west.rgw.control``
- ``.us-west.rgw.gc``
- ``.us-west.rgw.buckets``
- ``.us-west.rgw.buckets.index``
- ``.us-west.rgw.buckets.extra``
- ``.us-west.log``
- ``.us-west.intent-log``
- ``.us-west.usage``
- ``.us-west.users``
- ``.us-west.users.email``
- ``.us-west.users.swift``
- ``.us-west.users.uid``
See `Configuration Reference - Pools`_ for details on the default pools for
gateways. See `Pools`_ for details on creating pools. Execute the following
to create a pool::
ceph osd pool create {poolname} {pg-num} {pgp-num} {replicated | erasure} [{erasure-code-profile}] {ruleset-name} {ruleset-number}
.. tip:: When adding a large number of pools, it may take some time for your
cluster to return to a ``active + clean`` state.
.. topic:: CRUSH Maps
When deploying a Ceph Storage Cluster for the entire region, consider
using a CRUSH rule for the zone such that you do NOT have overlapping
failure domains. See `CRUSH Map`_ for details.
Ceph supports multiple CRUSH hierarchies and CRUSH rulesets, enabling
great flexibility in the way you configure your gateway. Pools such
as ``rgw.buckets.index`` may benefit from a modestly sized pool of SSDs
for fast performance. Backing storage may benefit from the increased economy
of erasure-coded storage, and/or the improved performance from cache tiering.
When you have completed this step, execute the following to ensure that
you have created all of the foregoing pools::
rados lspools
Create a Keyring
----------------
Each instance must have a user name and key to communicate with a Ceph Storage
Cluster. In the following steps, we use an admin node to create a keyring.
Then, we create a client user name and key for each instance. Next, we add the
keys to the Ceph Storage Cluster(s). Finally, we distribute the key ring to
each node containing an instance.
#. Create a keyring. ::
sudo ceph-authtool --create-keyring /etc/ceph/ceph.client.radosgw.keyring
sudo chmod +r /etc/ceph/ceph.client.radosgw.keyring
#. Generate a Ceph Object Gateway user name and key for each instance. ::
sudo ceph-authtool /etc/ceph/ceph.client.radosgw.keyring -n client.radosgw.us-east-1 --gen-key
sudo ceph-authtool /etc/ceph/ceph.client.radosgw.keyring -n client.radosgw.us-west-1 --gen-key
#. Add capabilities to each key. See `Configuration Reference - Pools`_ for details
on the effect of write permissions for the monitor and creating pools. ::
sudo ceph-authtool -n client.radosgw.us-east-1 --cap osd 'allow rwx' --cap mon 'allow rwx' /etc/ceph/ceph.client.radosgw.keyring
sudo ceph-authtool -n client.radosgw.us-west-1 --cap osd 'allow rwx' --cap mon 'allow rwx' /etc/ceph/ceph.client.radosgw.keyring
#. Once you have created a keyring and key to enable the Ceph Object Gateway
with access to the Ceph Storage Cluster, add each key as an entry to your
Ceph Storage Cluster(s). For example::
sudo ceph -k /etc/ceph/ceph.client.admin.keyring auth add client.radosgw.us-east-1 -i /etc/ceph/ceph.client.radosgw.keyring
sudo ceph -k /etc/ceph/ceph.client.admin.keyring auth add client.radosgw.us-west-1 -i /etc/ceph/ceph.client.radosgw.keyring
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four users **after**
you create the master region and the secondary region.
Install Apache/FastCGI
----------------------
For each :term:`Ceph Node` that runs a :term:`Ceph Object Gateway` daemon
instance, you must install Apache, FastCGI, the Ceph Object Gateway daemon
(``radosgw``) and the Ceph Object Gateway Sync Agent (``radosgw-agent``).
See `Install Ceph Object Gateway`_ for details.
Create Data Directories
-----------------------
Create data directories for each daemon instance on their respective
hosts. ::
ssh {us-east-1}
sudo mkdir -p /var/lib/ceph/radosgw/ceph-radosgw.us-east-1
ssh {us-west-1}
sudo mkdir -p /var/lib/ceph/radosgw/ceph-radosgw.us-west-1
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four data directories
**after** you create the master region and the secondary region.
Create a Gateway Configuration
------------------------------
For each instance, create an Ceph Object Gateway configuration file under the
``/etc/apache2/sites-available`` directory on the host(s) where you installed
the Ceph Object Gateway daemon(s). See below for an exemplary embodiment of a
gateway configuration as discussed in the following text.
.. literalinclude:: rgw.conf
:language: ini
#. Replace the ``/{path}/{socket-name}`` entry with path to the socket and
the socket name. For example,
``/var/run/ceph/client.radosgw.us-east-1.sock``. Ensure that you use the
same path and socket name in your ``ceph.conf`` entry.
#. Replace the ``{fqdn}`` entry with the fully-qualified domain name of the
server.
#. Replace the ``{email.address}`` entry with the email address for the
server administrator.
#. Add a ``ServerAlias`` if you wish to use S3-style subdomains
(of course you do).
#. Save the configuration to a file (e.g., ``rgw-us-east.conf``).
Repeat the process for the secondary zone (e.g., ``rgw-us-west.conf``).
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four gateway
configuration files on the respective nodes **after**
you create the master region and the secondary region.
Finally, if you enabled SSL, make sure that you set the port to your SSL port
(usually 443) and your configuration file includes the following::
SSLEngine on
SSLCertificateFile /etc/apache2/ssl/apache.crt
SSLCertificateKeyFile /etc/apache2/ssl/apache.key
SetEnv SERVER_PORT_SECURE 443
Enable the Configuration
------------------------
For each instance, enable the gateway configuration and disable the
default site.
#. Enable the site for the gateway configuration. ::
sudo a2ensite {rgw-conf-filename}
#. Disable the default site. ::
sudo a2dissite default
.. note:: Failure to disable the default site can lead to problems.
Add a FastCGI Script
--------------------
FastCGI requires a script for each Ceph Object Gateway instance to
enable the S3-compatible interface. To create the script, execute
the following procedures.
#. Go to the ``/var/www`` directory. ::
cd /var/www
#. Open an editor with the file name ``s3gw.fcgi``. **Note:** The configuration
file specifies this filename. ::
sudo vim s3gw.fcgi
#. Add a shell script with ``exec`` and the path to the gateway binary,
the path to the Ceph configuration file, and the user name (``-n``;
the same user name created in step 2 of `Create a Keyring`_.
Copy the following into the editor. ::
#!/bin/sh
exec /usr/bin/radosgw -c /etc/ceph/ceph.conf -n client.radosgw.{ID}
For example::
#!/bin/sh
exec /usr/bin/radosgw -c /etc/ceph/ceph.conf -n client.radosgw.us-east-1
#. Save the file.
#. Change the permissions on the file so that it is executable. ::
sudo chmod +x s3gw.fcgi
Repeat the process for the secondary zone.
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four FastCGI scripts
**after** you create the master region and the secondary region.
Add Instances to Ceph Config File
---------------------------------
On an admin node, add an entry for each instance in the Ceph configuration file
for your Ceph Storage Cluster(s). For example::
[global]
rgw region root pool = .us.rgw.root # Deprecated in Jewel
rgw zonegroup root pool = .us.rgw.root # From Jewel
[client.radosgw.us-east-1]
rgw region = us
rgw zone = us-east
rgw zone root pool = .us-east.rgw.root
keyring = /etc/ceph/ceph.client.radosgw.keyring
rgw dns name = {hostname}
rgw socket path = /var/run/ceph/$name.sock
host = {host-name}
[client.radosgw.us-west-1]
rgw region = us
rgw zone = us-west
rgw zone root pool = .us-west.rgw.root
keyring = /etc/ceph/ceph.client.radosgw.keyring
rgw dns name = {hostname}
rgw socket path = /var/run/ceph/$name.sock
host = {host-name}
Then, update each :term:`Ceph Node` with the updated Ceph configuration
file. For example::
ceph-deploy --overwrite-conf config push {node1} {node2} {nodex}
.. note:: When you use this procedure to configure the secondary region,
replace ``us`` with ``eu`` for the name, region, pools and zones.
You will have a total of four entries **after**
you create the master region and the secondary region.
Create a Region
---------------
#. Configure a region infile called ``us.json`` for the ``us`` region.
Copy the contents of the following example to a text editor. Set
``is_master`` to ``true``. Replace ``{fqdn}`` with the fully-qualified
domain name of the endpoint. It will specify a master zone as ``us-east``
and list it in the ``zones`` list along with the ``us-west`` zone.
See `Configuration Reference - Regions`_ for details.::
{ "name": "us",
"api_name": "us",
"is_master": "true",
"endpoints": [
"http:\/\/{fqdn}:80\/"],
"master_zone": "us-east",
"zones": [
{ "name": "us-east",
"endpoints": [
"http:\/\/{fqdn}:80\/"],
"log_meta": "true",
"log_data": "true"},
{ "name": "us-west",
"endpoints": [
"http:\/\/{fqdn}:80\/"],
"log_meta": "true",
"log_data": "true"}],
"placement_targets": [
{
"name": "default-placement",
"tags": []
}
],
"default_placement": "default-placement"}
#. Create the ``us`` region using the ``us.json`` infile you just
created. ::
radosgw-admin region set --infile us.json --name client.radosgw.us-east-1
#. Delete the default region (if it exists). ::
rados -p .us.rgw.root rm region_info.default
#. Set the ``us`` region as the default region. ::
radosgw-admin region default --rgw-region=us --name client.radosgw.us-east-1
Only one region can be the default region for a cluster.
#. Update the region map. ::
radosgw-admin regionmap update --name client.radosgw.us-east-1
If you use different Ceph Storage Cluster instances for regions, you should
repeat steps 2, 4 and 5 in by executing them with ``--name
client.radosgw-us-west-1``. You may also export the region map from the initial
gateway instance and import it followed by updating the region map.
.. note:: When you use this procedure to configure the secondary region,
replace ``us`` with ``eu``. You will have a total of two regions **after**
you create the master region and the secondary region.
Create Zones
------------
#. Configure a zone infile called ``us-east.json`` for the
``us-east`` zone.
Copy the contents of the following example to a text editor.
This configuration uses pool names prepended with the region name and
zone name. See `Configuration Reference - Pools`_ for additional details on
gateway pools. See `Configuration Reference - Zones`_ for additional
details on zones. ::
{ "domain_root": ".us-east.domain.rgw",
"control_pool": ".us-east.rgw.control",
"gc_pool": ".us-east.rgw.gc",
"log_pool": ".us-east.log",
"intent_log_pool": ".us-east.intent-log",
"usage_log_pool": ".us-east.usage",
"user_keys_pool": ".us-east.users",
"user_email_pool": ".us-east.users.email",
"user_swift_pool": ".us-east.users.swift",
"user_uid_pool": ".us-east.users.uid",
"system_key": { "access_key": "", "secret_key": ""},
"placement_pools": [
{ "key": "default-placement",
"val": { "index_pool": ".us-east.rgw.buckets.index",
"data_pool": ".us-east.rgw.buckets"}
}
]
}
#. Add the ``us-east`` zone using the ``us-east.json`` infile you
just created in both the east and west pools by specifying their respective
user names (i.e., ``--name``). ::
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.us-west-1
Repeat step 1 to create a zone infile for ``us-west``. Then add the zone
using the ``us-west.json`` infile in both the east and west pools by
specifying their respective user names (i.e., ``--name``). ::
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.us-west-1
#. Delete the default zone (if it exists). ::
rados -p .us-east.rgw.root rm zone_info.default
rados -p .us-west.rgw.root rm zone_info.default
#. Update the region map. ::
radosgw-admin regionmap update --name client.radosgw.us-east-1
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four zones **after**
you create the master zone and the secondary zone in each region.
Create Zone Users
-----------------
Ceph Object Gateway stores zone users in the zone pools. So you must create zone
users after configuring the zones. Copy the ``access_key`` and ``secret_key``
fields for each user so you can update your zone configuration once you complete
this step. ::
radosgw-admin user create --uid="us-east" --display-name="Region-US Zone-East" --name client.radosgw.us-east-1 --system
radosgw-admin user create --uid="us-west" --display-name="Region-US Zone-West" --name client.radosgw.us-west-1 --system
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four zone users
**after** you create the master region and the secondary region and their
zones. These users are different from the users created in `Create a
Keyring`_.
Update Zone Configurations
--------------------------
You must update the zone configuration with zone users so that
the synchronization agents can authenticate with the zones.
#. Open your ``us-east.json`` zone configuration file and paste the contents of
the ``access_key`` and ``secret_key`` fields from the step of creating
zone users into the ``system_key`` field of your zone configuration
infile. ::
{ "domain_root": ".us-east.domain.rgw",
"control_pool": ".us-east.rgw.control",
"gc_pool": ".us-east.rgw.gc",
"log_pool": ".us-east.log",
"intent_log_pool": ".us-east.intent-log",
"usage_log_pool": ".us-east.usage",
"user_keys_pool": ".us-east.users",
"user_email_pool": ".us-east.users.email",
"user_swift_pool": ".us-east.users.swift",
"user_uid_pool": ".us-east.users.uid",
"system_key": {
"access_key": "{paste-access_key-here}",
"secret_key": "{paste-secret_key-here}"
},
"placement_pools": [
{ "key": "default-placement",
"val": { "index_pool": ".us-east.rgw.buckets.index",
"data_pool": ".us-east.rgw.buckets"}
}
]
}
#. Save the ``us-east.json`` file. Then, update your zone configuration. ::
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.us-west-1
#. Repeat step 1 to update the zone infile for ``us-west``. Then, update
your zone configuration. ::
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.us-west-1
.. note:: When you use this procedure to configure the secondary region,
replace ``us-`` with ``eu-``. You will have a total of four zones **after**
you create the master zone and the secondary zone in each region.
Restart Services
----------------
Once you have redeployed your Ceph configuration files, we recommend restarting
your Ceph Storage Cluster(s) and Apache instances.
For Ubuntu, use the following on each :term:`Ceph Node`::
sudo restart ceph-all
For Red Hat/CentOS, use the following::
sudo /etc/init.d/ceph restart
To ensure that all components have reloaded their configurations, for each
gateway instance we recommend restarting the ``apache2`` service. For example::
sudo service apache2 restart
Start Gateway Instances
-----------------------
Start up the ``radosgw`` service. ::
sudo /etc/init.d/radosgw start
If you are running multiple instances on the same host, you must specify the
user name. ::
sudo /etc/init.d/radosgw start --name client.radosgw.us-east-1
Open a browser and check the endpoints for each zone. A simple HTTP request
to the domain name should return the following:
.. code-block:: xml
<ListAllMyBucketsResult xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
<Owner>
<ID>anonymous</ID>
<DisplayName/>
</Owner>
<Buckets/>
</ListAllMyBucketsResult>
Configure a Secondary Region
============================
This section provides an exemplary procedure for setting up a cluster with
multiple regions. Configuring a cluster that spans regions requires maintaining
a global namespace, so that there are no namespace clashes among object names
stored across in different regions.
This section extends the procedure in `Configure a Master Region`_, but
changes the region name and modifies a few procedures. See the following
sections for details.
Naming for the Secondary Region
-------------------------------
Before configuring the cluster, defining region, zone and instance names will
help you manage your cluster. Let's assume the region represents the European
Union, and we refer to it by its standard abbreviation.
- European Union: ``eu``
Let's assume the zones represent the Eastern and Western European Union. For
continuity, our naming convention will use ``{region name}-{zone name}``
format, but you can use any naming convention you prefer.
- European Union, East Region: ``eu-east``
- European Union, West Region: ``eu-west``
Finally, let's assume that zones may have more than one Ceph Object Gateway
instance per zone. For continuity, our naming convention will use
``{region name}-{zone name}-{instance}`` format, but you can use any naming
convention you prefer.
- European Union Region, Master Zone, Instance 1: ``eu-east-1``
- European Union Region, Secondary Zone, Instance 1: ``eu-west-1``
Configuring a Secondary Region
------------------------------
Repeat the exemplary procedure of `Configure a Master Region`_
with the following differences:
#. Use `Naming for the Secondary Region`_ in lieu of `Naming for
the Master Region`_.
#. `Create Pools`_ using ``eu`` instead of ``us``.
#. `Create a Keyring`_ and the corresponding keys using ``eu`` instead of
``us``. You may use the same keyring if you desire, but ensure that you
create the keys on the Ceph Storage Cluster for that region (or region
and zone).
#. `Install Apache/FastCGI`_.
#. `Create Data Directories`_ using ``eu`` instead of ``us``.
#. `Create a Gateway Configuration`_ using ``eu`` instead of ``us`` for
the socket names.
#. `Enable the Configuration`_.
#. `Add a FastCGI Script`_ using ``eu`` instead of ``us`` for the user names.
#. `Add Instances to Ceph Config File`_ using ``eu`` instead of ``us`` for the
pool names.
#. `Create a Region`_ using ``eu`` instead of ``us``. Set ``is_master`` to
``false``. For consistency, create the master region in the secondary region
too. ::
radosgw-admin region set --infile us.json --name client.radosgw.eu-east-1
#. `Create Zones`_ using ``eu`` instead of ``us``. Ensure that you update the
user name (i.e., ``--name``) so that you create the zones in the correct
cluster.
#. `Update Zone Configurations`_ using ``eu`` instead of ``us``.
#. Create zones from master region in the secondary region. ::
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.eu-east-1
radosgw-admin zone set --rgw-zone=us-east --infile us-east.json --name client.radosgw.eu-west-1
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.eu-east-1
radosgw-admin zone set --rgw-zone=us-west --infile us-west.json --name client.radosgw.eu-west-1
#. Create zones from secondary region in the master region. ::
radosgw-admin zone set --rgw-zone=eu-east --infile eu-east.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=eu-east --infile eu-east.json --name client.radosgw.us-west-1
radosgw-admin zone set --rgw-zone=eu-west --infile eu-west.json --name client.radosgw.us-east-1
radosgw-admin zone set --rgw-zone=eu-west --infile eu-west.json --name client.radosgw.us-west-1
#. `Restart Services`_.
#. `Start Gateway Instances`_.
Multi-Site Data Replication
===========================
The data synchronization agent replicates the data of a master zone to a
secondary zone. The master zone of a region is the source for the secondary zone
of the region and it gets selected automatically.
.. image:: ../images/zone-sync.png
To configure the synchronization agent, retrieve the access key and secret for
the source and destination, and the destination URL and port.
You may use ``radosgw-admin zone list`` to get a list of zone names. You
may use ``radosgw-admin zone get`` to identify the key and secret for the
zone. You may refer to the gateway configuration file you created under
`Create a Gateway Configuration`_ to identify the port number.
You only need the hostname and port for a single instance (assuming all
gateway instances in a region/zone access the same Ceph Storage Cluster).
Specify these values in a configuration file
(e.g., ``cluster-data-sync.conf``), and include a ``log_file`` name.
For example:
.. code-block:: ini
src_access_key: {source-access-key}
src_secret_key: {source-secret-key}
destination: https://zone-name.fqdn.com:port
dest_access_key: {destination-access-key}
dest_secret_key: {destination-secret-key}
log_file: {log.filename}
A concrete example may look like this:
.. code-block:: ini
src_access_key: DG8RE354EFPZBICHIAF0
src_secret_key: i3U0HiRP8CXaBWrcF8bbh6CbsxGYuPPwRkixfFSb
destination: https://us-west.storage.net:80
dest_access_key: U60RFI6B08F32T2PD30G
dest_secret_key: W3HuUor7Gl1Ee93pA2pq2wFk1JMQ7hTrSDecYExl
log_file: /var/log/radosgw/radosgw-sync-us-east-west.log
To activate the data synchronization agent, open a terminal and
execute the following::
radosgw-agent -c region-data-sync.conf
When the synchronization agent is running, you should see output
indicating that the agent is synchronizing shards of data. ::
INFO:radosgw_agent.sync:Starting incremental sync
INFO:radosgw_agent.worker:17910 is processing shard number 0
INFO:radosgw_agent.worker:shard 0 has 0 entries after ''
INFO:radosgw_agent.worker:finished processing shard 0
INFO:radosgw_agent.worker:17910 is processing shard number 1
INFO:radosgw_agent.sync:1/64 shards processed
INFO:radosgw_agent.worker:shard 1 has 0 entries after ''
INFO:radosgw_agent.worker:finished processing shard 1
INFO:radosgw_agent.sync:2/64 shards processed
...
.. note:: You must have an agent for each source-destination pair.
Inter-Region Metadata Replication
=================================
The data synchronization agent replicates the metadata of master zone in the
master region to a master zone in a secondary region. Metadata consists of
gateway users and buckets, but not the objects within the buckets--ensuring a
unified namespace across the cluster. The master zone of the master region is
the source for the master zone of the secondary region and it gets selected
automatically.
.. image:: ../images/region-sync.png
:align: center
Follow the same steps in `Multi-Site Data Replication`_ by specifying the master
zone of the master region as the source zone and the master zone of the
secondary region as the secondary zone. When activating the ``radosgw-agent``,
specify ``--metadata-only`` so that it only copies metadata. For example::
radosgw-agent -c inter-region-data-sync.conf --metadata-only
Once you have completed the foregoing procedure, you should have a cluster
consisting of a master region (``us``) and a secondary region (``eu``) where
there is a unified namespace between the two regions.
.. _CRUSH Map: ../../rados/operations/crush-map
.. _Install Ceph Object Gateway: ../../install/install-ceph-gateway
.. _Ceph configuration file: ../../rados/configuration/ceph-conf
.. _Configuration Reference - Pools: ../config-ref#pools
.. _Configuration Reference - Regions: ../config-ref#regions
.. _Configuration Reference - Zones: ../config-ref#zones
.. _Pools: ../../rados/operations/pools
.. _Simple Configuration: ../config-fcgi

7
ceph/doc/radosgw/index.rst
View File

@ -13,7 +13,7 @@ Ceph Storage Clusters. :term:`Ceph Object Storage` supports two interfaces:
that is compatible with a large subset of the OpenStack Swift API.
Ceph Object Storage uses the Ceph Object Gateway daemon (``radosgw``), which is
a FastCGI module for interacting with a Ceph Storage Cluster. Since it
an HTTP server for interacting with a Ceph Storage Cluster. Since it
provides interfaces compatible with OpenStack Swift and Amazon S3, the Ceph
Object Gateway has its own user management. Ceph Object Gateway can store data
in the same Ceph Storage Cluster used to store data from Ceph Filesystem clients
@ -37,9 +37,8 @@ you may write data with one API and retrieve it with the other.
:maxdepth: 1
Manual Install w/Civetweb <../../install/install-ceph-gateway>
Simple Configuration w/Apache/FastCGI <config-fcgi>
Federated Configuration (Deprecated) <federated-config>
Multisite Configuration <multisite>
Configuring Pools <pools>
Config Reference <config-ref>
Admin Guide <admin>
S3 API <s3>
@ -51,8 +50,8 @@ you may write data with one API and retrieve it with the other.
Multi-tenancy <multitenancy>
Compression <compression>
Server-Side Encryption <encryption>
Bucket Policy <bucketpolicy>
Data Layout in RADOS <layout>
Upgrade to Older Versions of Jewel <upgrade_to_jewel>
troubleshooting
Manpage radosgw <../../man/8/radosgw>
Manpage radosgw-admin <../../man/8/radosgw-admin>

87
ceph/doc/radosgw/layout.rst
View File

@ -82,12 +82,14 @@ The user ID in RGW is a string, typically the actual user name from the user
credentials and not a hashed or mapped identifier.
When accessing a user's data, the user record is loaded from an object
"<user_id>" in pool ".users.uid".
"<user_id>" in pool "default.rgw.meta" with namespace "users.uid".
Bucket names are represented directly in the pool ".rgw". Bucket record is
Bucket names are represented in the pool "default.rgw.meta" with namespace
"root". Bucket record is
loaded in order to obtain so-called marker, which serves as a bucket ID.
The object is located in pool ".rgw.buckets". Object name is "<marker>_<key>",
The object is located in pool "default.rgw.buckets.data".
Object name is "<marker>_<key>",
for example "default.7593.4_image.png", where the marker is "default.7593.4"
and the key is "image.png". Since these concatenated names are not parsed,
only passed down to RADOS, the choice of the separator is not important and
@ -110,7 +112,8 @@ Bucket and Object Listing
-------------------------
Buckets that belong to a given user are listed in an omap of an object named
"<user_id>.buckets" (for example, "foo.buckets") in pool ".users.uid".
"<user_id>.buckets" (for example, "foo.buckets") in pool "default.rgw.meta"
with namespace "users.uid".
These objects are accessed when listing buckets, when updating bucket
contents, and updating and retrieving bucket statistics (e.g. for quota).
@ -121,7 +124,7 @@ These listings are kept consistent with buckets in pool ".rgw".
Objects that belong to a given bucket are listed in a bucket index,
as discussed in sub-section 'Bucket Index' above. The default naming
for index objects is ".dir.<marker>" in pool ".rgw.buckets.index".
for index objects is ".dir.<marker>" in pool "default.rgw.buckets.index".
Footnotes
---------
@ -136,64 +139,66 @@ In Hammer, one LevelDB is used to store omap in each OSD.
exist and the 'bucket' metadata contained its information. It is possible
to encounter such buckets in old installations.
[3] In Infernalis, a pending commit exists that removes the need of prefixing
all the rgw system pools with a period, and also renames all of these pools.
See Github pull request #4944 "rgw noperiod".
[3] The pool names have been changed starting with the Infernalis release.
If you are looking at an older setup, some details may be different. In
particular there was a different pool for each of the namespaces that are
now being used inside the default.root.meta pool.
Appendix: Compendum
-------------------
Appendix: Compendium
--------------------
Known pools:
.rgw.root
Unspecified region, zone, and global information records, one per object.
.rgw.control
<zone>.rgw.control
notify.<N>
.rgw
<bucket>
.bucket.meta.<bucket>:<marker> # see put_bucket_instance_info()
<zone>.rgw.meta
Multiple namespaces with different kinds of metadata:
The tenant is used to disambiguate buckets, but not bucket instances.
Example:
namespace: root
<bucket>
.bucket.meta.<bucket>:<marker> # see put_bucket_instance_info()
.bucket.meta.prodtx:test%25star:default.84099.6
.bucket.meta.testcont:default.4126.1
.bucket.meta.prodtx:testcont:default.84099.4
prodtx/testcont
prodtx/test%25star
testcont
The tenant is used to disambiguate buckets, but not bucket instances.
Example::
.rgw.gc
gc.<N>
.bucket.meta.prodtx:test%25star:default.84099.6
.bucket.meta.testcont:default.4126.1
.bucket.meta.prodtx:testcont:default.84099.4
prodtx/testcont
prodtx/test%25star
testcont
.users.uid
Contains _both_ per-user information (RGWUserInfo) in "<user>" objects
and per-user lists of buckets in omaps of "<user>.buckets" objects.
The "<user>" may contain the tenant if non-empty, for example:
namespace: users.uid
Contains _both_ per-user information (RGWUserInfo) in "<user>" objects
and per-user lists of buckets in omaps of "<user>.buckets" objects.
The "<user>" may contain the tenant if non-empty, for example::
prodtx$prodt
test2.buckets
prodtx$prodt.buckets
test2
prodtx$prodt
test2.buckets
prodtx$prodt.buckets
test2
.users.email
Unimportant
namespace: users.email
Unimportant
.users
47UA98JSTJZ9YAN3OS3O
It's unclear why user ID is not used to name objects in this pool.
namespace: users.keys
47UA98JSTJZ9YAN3OS3O
.users.swift
test:tester
This allows radosgw to look up users by their access keys during authentication.
.rgw.buckets.index
namespace: users.swift
test:tester
<zone>.rgw.buckets.index
Objects are named ".dir.<marker>", each contains a bucket index.
If the index is sharded, each shard appends the shard index after
the marker.
.rgw.buckets
<zone>.rgw.buckets.data
default.7593.4__shadow_.488urDFerTYXavx4yAd-Op8mxehnvTI_1
<marker>_<key>

63
ceph/doc/radosgw/multisite.rst
View File

@ -63,7 +63,7 @@ gateway instances, one for each Ceph storage cluster.
This guide assumes at least two Ceph storage clusters in geographically
separate locations; however, the configuration can work on the same
site. This guide also assumes four Ceph object gateway servers named
site. This guide also assumes two Ceph object gateway servers named
``rgw1`` and ``rgw2``.
A multi-site configuration requires a master zone group and a master
@ -74,58 +74,8 @@ In this guide, the ``rgw1`` host will serve as the master zone of the
master zone group; and, the ``rgw2`` host will serve as the secondary zone
of the master zone group.
Pools
=====
We recommend using the `Ceph Placement Group’s per Pool
Calculator <http://ceph.com/pgcalc/>`__ to calculate a
suitable number of placement groups for the pools the ``ceph-radosgw``
daemon will create. Set the calculated values as defaults in your Ceph
configuration file. For example:
::
osd pool default pg num = 50
osd pool default pgp num = 50
.. note:: Make this change to the Ceph configuration file on your
storage cluster; then, either make a runtime change to the
configuration so that it will use those defaults when the gateway
instance creates the pools.
Alternatively, create the pools manually. See
`Pools <http://docs.ceph.com/docs/master/rados/operations/pools/#pools>`__
for details on creating pools.
Pool names particular to a zone follow the naming convention
``{zone-name}.pool-name``. For example, a zone named ``us-east`` will
have the following pools:
- ``.rgw.root``
- ``us-east.rgw.control``
- ``us-east.rgw.data.root``
- ``us-east.rgw.gc``
- ``us-east.rgw.log``
- ``us-east.rgw.intent-log``
- ``us-east.rgw.usage``
- ``us-east.rgw.users.keys``
- ``us-east.rgw.users.email``
- ``us-east.rgw.users.swift``
- ``us-east.rgw.users.uid``
- ``us-east.rgw.buckets.index``
- ``us-east.rgw.buckets.data``
See `Pools`_ for instructions on creating and tuning pools for Ceph
Object Storage.
Configuring a Master Zone
@ -814,8 +764,8 @@ realm. Alternatively, to change which realm is the default, execute:
# radosgw-admin realm default --rgw-realm=movies
..note:: When the realm is default, the command line assumes
``--rgw-realm=<realm-name>`` as an argument.
.. note:: When the realm is default, the command line assumes
``--rgw-realm=<realm-name>`` as an argument.
Delete a Realm
~~~~~~~~~~~~~~
@ -1504,3 +1454,6 @@ instance.
| | keeping inter-zone group | | |
| | synchronization progress. | | |
+-------------------------------------+-----------------------------------+---------+-----------------------+
.. _`Pools`: ../pools

366
ceph/doc/radosgw/nfs.rst Normal file
View File

@ -0,0 +1,366 @@
===
NFS
===
.. versionadded:: Jewel
Ceph Object Gateway namespaces can now be exported over file-based
access protocols such as NFSv3 and NFSv4, alongside traditional HTTP access
protocols (S3 and Swift).
In particular, the Ceph Object Gateway can now be configured to
provide file-based access when embedded in the NFS-Ganesha NFS server.
librgw
======
The librgw.so shared library (Unix) provides a loadable interface to
Ceph Object Gateway services, and instantiates a full Ceph Object Gateway
instance on initialization.
In turn, librgw.so exports rgw_file, a stateful API for file-oriented
access to RGW buckets and objects. The API is general, but its design
is strongly influenced by the File System Abstraction Layer (FSAL) API
of NFS-Ganesha, for which it has been primarily designed.
A set of Python bindings is also provided.
Namespace Conventions
=====================
The implementation conforms to Amazon Web Services (AWS) hierarchical
namespace conventions which map UNIX-style path names onto S3 buckets
and objects.
The top level of the attached namespace consists of S3 buckets,
represented as NFS directories. Files and directories subordinate to
buckets are each represented as objects, following S3 prefix and
delimiter conventions, with '/' being the only supported path
delimiter [#]_.
For example, if an NFS client has mounted an RGW namespace at "/nfs",
then a file "/nfs/mybucket/www/index.html" in the NFS namespace
corresponds to an RGW object "www/index.html" in a bucket/container
"mybucket."
Although it is generally invisible to clients, the NFS namespace is
assembled through concatenation of the corresponding paths implied by
the objects in the namespace. Leaf objects, whether files or
directories, will always be materialized in an RGW object of the
corresponding key name, "<name>" if a file, "<name>/" if a directory.
Non-leaf directories (e.g., "www" above) might only be implied by
their appearance in the names of one or more leaf objects. Directories
created within NFS or directly operated on by an NFS client (e.g., via
an attribute-setting operation such as chown or chmod) always have a
leaf object representation used to store materialized attributes such
as Unix ownership and permissions.
Supported Operations
====================
The RGW NFS interface supports most operations on files and
directories, with the following restrictions:
- Links, including symlinks, are not supported
- NFS ACLs are not supported
+ Unix user and group ownership and permissions *are* supported
- Directories may not be moved/renamed
+ files may be moved between directories
- Only full, sequential *write* i/o is supported
+ i.e., write operations are constrained to be **uploads**
+ many typical i/o operations such as editing files in place will necessarily fail as they perform non-sequential stores
+ some file utilities *apparently* writing sequentially (e.g., some versions of GNU tar) may fail due to infrequent non-sequential stores
+ When mounting via NFS, sequential application i/o can generally be constrained to be written sequentially to the NFS server via a synchronous mount option (e.g. -osync in Linux)
+ NFS clients which cannot mount synchronously (e.g., MS Windows) will not be able to upload files
Security
========
The RGW NFS interface provides a hybrid security model with the
following characteristics:
- NFS protocol security is provided by the NFS-Ganesha server, as negotiated by the NFS server and clients
+ e.g., clients can by trusted (AUTH_SYS), or required to present Kerberos user credentials (RPCSEC_GSS)
+ RPCSEC_GSS wire security can be integrity only (krb5i) or integrity and privacy (encryption, krb5p)
+ various NFS-specific security and permission rules are available
* e.g., root-squashing
- a set of RGW/S3 security credentials (unknown to NFS) is associated with each RGW NFS mount (i.e., NFS-Ganesha EXPORT)
+ all RGW object operations performed via the NFS server will be performed by the RGW user associated with the credentials stored in the export being accessed (currently only RGW and RGW LDAP credentials are supported)
* additional RGW authentication types such as Keystone are not currently supported
Configuring an NFS-Ganesha Instance
===================================
Each NFS RGW instance is an NFS-Ganesha server instance *embeddding*
a full Ceph RGW instance.
Therefore, the RGW NFS configuration includes Ceph and Ceph Object
Gateway-specific configuration in a local ceph.conf, as well as
NFS-Ganesha-specific configuration in the NFS-Ganesha config file,
ganesha.conf.
ceph.conf
---------
Required ceph.conf configuration for RGW NFS includes:
* valid [client.radosgw.{instance-name}] section
* valid values for minimal instance configuration, in particular, an installed and correct ``keyring``
Other config variables are optional, front-end-specific and front-end
selection variables (e.g., ``rgw data`` and ``rgw frontends``) are
optional and in some cases ignored.
A small number of config variables (e.g., ``rgw_namespace_expire_secs``)
are unique to RGW NFS.
ganesha.conf
------------
A strictly minimal ganesha.conf for use with RGW NFS includes one
EXPORT block with embedded FSAL block of type RGW::
EXPORT
{
Export_ID={numeric-id};
Path = "/";
Pseudo = "/";
Access_Type = RW;
SecType = "sys";
NFS_Protocols = 4;
Transport_Protocols = TCP;
# optional, permit unsquashed access by client "root" user
#Squash = No_Root_Squash;
FSAL {
Name = RGW;
User_Id = {s3-user-id};
Access_Key_Id ="{s3-access-key}";
Secret_Access_Key = "{s3-secret}";
}
}
``Export_ID`` must have an integer value, e.g., "77"
``Path`` (for RGW) should be "/"
``Pseudo`` defines an NFSv4 pseudo root name (NFSv4 only)
``SecType = sys;`` allows clients to attach without Kerberos
authentication
``Squash = No_Root_Squash;`` enables the client root user to override
permissions (Unix convention). When root-squashing is enabled,
operations attempted by the root user are performed as if by the local
"nobody" (and "nogroup") user on the NFS-Ganesha server
The RGW FSAL additionally supports RGW-specific configuration
variables in the RGW config section::
RGW {
cluster = "{cluster name, default 'ceph'}";
name = "client.rgw.{instance-name}";
ceph_conf = "/opt/ceph-rgw/etc/ceph/ceph.conf";
init_args = "-d --debug-rgw=16";
}
``cluster`` sets a Ceph cluster name (must match the cluster being exported)
``name`` sets an RGW instance name (must match the cluster being exported)
``ceph_conf`` gives a path to a non-default ceph.conf file to use
Other useful NFS-Ganesha configuration:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Any EXPORT block which should support NFSv3 should include version 3
in the NFS_Protocols setting. Additionally, NFSv3 is the last major
version to support the UDP transport. To enable UDP, include it in the
Transport_Protocols setting. For example::
EXPORT {
...
NFS_Protocols = 3,4;
Transport_Protocols = UDP,TCP;
...
}
One important family of options pertains to interaction with the Linux
idmapping service, which is used to normalize user and group names
across systems. Details of idmapper integration are not provided here.
With Linux NFS clients, NFS-Ganesha can be configured
to accept client-supplied numeric user and group identifiers with
NFSv4, which by default stringifies these--this may be useful in small
setups and for experimentation::
NFSV4 {
Allow_Numeric_Owners = true;
Only_Numeric_Owners = true;
}
Troubleshooting
~~~~~~~~~~~~~~~
NFS-Ganesha configuration problems are usually debugged by running the
server with debugging options, controlled by the LOG config section.
NFS-Ganesha log messages are grouped into various components, logging
can be enabled separately for each component. Valid values for
component logging include::
*FATAL* critical errors only
*WARN* unusual condition
*DEBUG* mildly verbose trace output
*FULL_DEBUG* verbose trace output
Example::
LOG {
Components {
MEMLEAKS = FATAL;
FSAL = FATAL;
NFSPROTO = FATAL;
NFS_V4 = FATAL;
EXPORT = FATAL;
FILEHANDLE = FATAL;
DISPATCH = FATAL;
CACHE_INODE = FATAL;
CACHE_INODE_LRU = FATAL;
HASHTABLE = FATAL;
HASHTABLE_CACHE = FATAL;
DUPREQ = FATAL;
INIT = DEBUG;
MAIN = DEBUG;
IDMAPPER = FATAL;
NFS_READDIR = FATAL;
NFS_V4_LOCK = FATAL;
CONFIG = FATAL;
CLIENTID = FATAL;
SESSIONS = FATAL;
PNFS = FATAL;
RW_LOCK = FATAL;
NLM = FATAL;
RPC = FATAL;
NFS_CB = FATAL;
THREAD = FATAL;
NFS_V4_ACL = FATAL;
STATE = FATAL;
FSAL_UP = FATAL;
DBUS = FATAL;
}
# optional: redirect log output
# Facility {
# name = FILE;
# destination = "/tmp/ganesha-rgw.log";
# enable = active;
}
}
Running Multiple NFS Gateways
=============================
Each NFS-Ganesha instance acts as a full gateway endpoint, with the
limitation that currently an NFS-Ganesha instance cannot be configured
to export HTTP services. As with ordinary gateway instances, any
number of NFS-Ganesha instances can be started, exporting the same or
different resources from the cluster. This enables the clustering of
NFS-Ganesha instances. However, this does not imply high availability.
When regular gateway instances and NFS-Ganesha instances overlap the
same data resources, they will be accessible from both the standard S3
API and through the NFS-Ganesha instance as exported. You can
co-locate the NFS-Ganesha instance with a Ceph Object Gateway instance
on the same host.
RGW vs RGW NFS
==============
Exporting an NFS namespace and other RGW namespaces (e.g., S3 or Swift
via the Civetweb HTTP front-end) from the same program instance is
currently not supported.
When adding objects and buckets outside of NFS, those objects will
appear in the NFS namespace in the time set by
``rgw_nfs_namespace_expire_secs``, which defaults to 300 seconds (5 minutes).
Override the default value for ``rgw_nfs_namespace_expire_secs`` in the
Ceph configuration file to change the refresh rate.
If exporting Swift containers that do not conform to valid S3 bucket
naming requirements, set ``rgw_relaxed_s3_bucket_names`` to true in the
[client.radosgw] section of the Ceph configuration file. For example,
if a Swift container name contains underscores, it is not a valid S3
bucket name and will be rejected unless ``rgw_relaxed_s3_bucket_names``
is set to true.
Configuring NFSv4 clients
=========================
To access the namespace, mount the configured NFS-Ganesha export(s)
into desired locations in the local POSIX namespace. As noted, this
implementation has a few unique restrictions:
- NFS 4.1 and higher protocol flavors are preferred
+ NFSv4 OPEN and CLOSE operations are used to track upload transactions
- To upload data successfully, clients must preserve write ordering
+ on Linux and many Unix NFS clients, use the -osync mount option
Conventions for mounting NFS resources are platform-specific. The
following conventions work on Linux and some Unix platforms:
From the command line::
mount -t nfs -o nfsvers=4,noauto,soft,proto=tcp <ganesha-host-name>:/ <mount-point>
In /etc/fstab::
<ganesha-host-name>:/ <mount-point> nfs noauto,soft,nfsvers=4.1,sync,proto=tcp 0 0
Specify the NFS-Ganesha host name and the path to the mount point on
the client.
Configuring NFSv3 Clients
=========================
Linux clients can be configured to mount with NFSv3 by supplying
``nfsvers=3`` and ``noacl`` as mount options. To use UDP as the
transport, add ``proto=udp`` to the mount options. However, TCP is the
preferred transport::
<ganesha-host-name>:/ <mount-point> nfs noauto,noacl,soft,nfsvers=3,sync,proto=tcp 0 0
Configure the NFS Ganesha EXPORT block Protocols setting with version
3 and the Transports setting with UDP if the mount will use version 3 with UDP.
NFSv3 Semantics
---------------
Since NFSv3 does not communicate client OPEN and CLOSE operations to
file servers, RGW NFS cannot use these operations to mark the
beginning and ending of file upload transactions. Instead, RGW NFS
starts a new upload when the first write is sent to a file at offset
0, and finalizes the upload when no new writes to the file have been
seen for a period of time, by default, 10 seconds. To change this
timeout, set an alternate value for ``rgw_nfs_write_completion_interval_s``
in the RGW section(s) of the Ceph configuration file.
References
==========
.. [#] http://docs.aws.amazon.com/AmazonS3/latest/dev/ListingKeysHierarchy.html

55
ceph/doc/radosgw/pools.rst Normal file
View File

@ -0,0 +1,55 @@
=====
Pools
=====
The Ceph Object Gateway uses several pools for its various storage needs,
which are listed in the Zone object (see ``radosgw-admin zone get``). A
single zone named ``default`` is created automatically with pool names
starting with ``default.rgw.``, but a `Multisite Configuration`_ will have
multiple zones.
Tuning
======
When ``radosgw`` first tries to operate on a zone pool that does not
exist, it will create that pool with the default values from
``osd pool default pg num`` and ``osd pool default pgp num``. These defaults
are sufficient for some pools, but others (especially those listed in
``placement_pools`` for the bucket index and data) will require additional
tuning. We recommend using the `Ceph Placement Group’s per Pool
Calculator <http://ceph.com/pgcalc/>`__ to calculate a suitable number of
placement groups for these pools. See
`Pools <http://docs.ceph.com/docs/master/rados/operations/pools/#pools>`__
for details on pool creation.
Pool Namespaces
===============
.. versionadded:: Luminous
Pool names particular to a zone follow the naming convention
``{zone-name}.pool-name``. For example, a zone named ``us-east`` will
have the following pools:
- ``.rgw.root``
- ``us-east.rgw.control``
- ``us-east.rgw.meta``
- ``us-east.rgw.log``
- ``us-east.rgw.buckets.index``
- ``us-east.rgw.buckets.data``
The zone definitions list several more pools than that, but many of those
are consolidated through the use of rados namespaces. For example, all of
the following pool entries use namespaces of the ``us-east.rgw.meta`` pool::
"user_keys_pool": "us-east.rgw.meta:users.keys",
"user_email_pool": "us-east.rgw.meta:users.email",
"user_swift_pool": "us-east.rgw.meta:users.swift",
"user_uid_pool": "us-east.rgw.meta:users.uid",
.. _`Multisite Configuration`: ../multisite

30
ceph/doc/radosgw/rgw-centos.conf
View File

@ -1,30 +0,0 @@
FastCgiExternalServer /var/www/html/s3gw.fcgi -socket /var/run/ceph/ceph.radosgw.gateway.fastcgi.sock
<VirtualHost *:80>
ServerName {fqdn}
<!--Remove the comment. Add a server alias with *.{fqdn} for S3 subdomains-->
<!--ServerAlias *.{fqdn}-->
ServerAdmin {email.address}
DocumentRoot /var/www/html
RewriteEngine On
RewriteRule ^/(.*) /s3gw.fcgi?%{QUERY_STRING} [E=HTTP_AUTHORIZATION:%{HTTP:Authorization},L]
<IfModule mod_fastcgi.c>
<Directory /var/www/html>
Options +ExecCGI
AllowOverride All
SetHandler fastcgi-script
Order allow,deny
Allow from all
AuthBasicAuthoritative Off
</Directory>
</IfModule>
AllowEncodedSlashes On
ErrorLog /var/log/httpd/error.log
CustomLog /var/log/httpd/access.log combined
ServerSignature Off
</VirtualHost>

29
ceph/doc/radosgw/rgw-debian.conf
View File

@ -1,29 +0,0 @@
FastCgiExternalServer /var/www/s3gw.fcgi -socket /var/run/ceph/ceph.radosgw.gateway.fastcgi.sock
<VirtualHost *:80>
ServerName {fqdn}
<!--Remove the comment. Add a server alias with *.{fqdn} for S3 subdomains-->
<!--ServerAlias *.{fqdn}-->
ServerAdmin {email.address}
DocumentRoot /var/www
RewriteEngine On
RewriteRule ^/(.*) /s3gw.fcgi?%{QUERY_STRING} [E=HTTP_AUTHORIZATION:%{HTTP:Authorization},L]
<IfModule mod_fastcgi.c>
<Directory /var/www>
Options +ExecCGI
AllowOverride All
SetHandler fastcgi-script
Order allow,deny
Allow from all
AuthBasicAuthoritative Off
</Directory>
</IfModule>
AllowEncodedSlashes On
ErrorLog /var/log/apache2/error.log
CustomLog /var/log/apache2/access.log combined
ServerSignature Off
</VirtualHost>

30
ceph/doc/radosgw/rgw.conf
View File

@ -1,30 +0,0 @@
FastCgiExternalServer /var/www/s3gw.fcgi -socket /{path}/{socket-name}.sock
<VirtualHost *:80>
ServerName {fqdn}
<!--Remove the comment. Add a server alias with *.{fqdn} for S3 subdomains-->
<!--ServerAlias *.{fqdn}-->
ServerAdmin {email.address}
DocumentRoot /var/www
RewriteEngine On
RewriteRule ^/(.*) /s3gw.fcgi?%{QUERY_STRING} [E=HTTP_AUTHORIZATION:%{HTTP:Authorization},L]
<IfModule mod_fastcgi.c>
<Directory /var/www>
Options +ExecCGI
AllowOverride All
SetHandler fastcgi-script
Order allow,deny
Allow from all
AuthBasicAuthoritative Off
</Directory>
</IfModule>
AllowEncodedSlashes On
ErrorLog /var/log/apache2/error.log
CustomLog /var/log/apache2/access.log combined
ServerSignature Off
</VirtualHost>

43
ceph/doc/radosgw/upgrade_to_jewel.rst
View File

@ -1,43 +0,0 @@
=================================================================
RGW upgrading to Jewel versions 10.2.0, 10.2.1, 10.2.2 and 10.2.3
=================================================================
.. versionadded:: Jewel
Upgrade of :term:`Ceph Object Gateway` to older versions of jewel (up to 10.2.3 included) caused issues. This document describes the needed recovery procedure.
Mixed version of :term:`Ceph Object Gateway` is not supported
Backup of old configuration
===========================
rados mkpool .rgw.root.backup
rados cppool .rgw.root .rgw.root.backup
Non default setting for `rgw region root pool`
==============================================
If an existing multisite configuration uses a non-default setting for
`rgw region root pool`, the new pool settings `rgw zonegroup root pool`,
`rgw period root pool` and `rgw realm root pool` should be set to match.
Fix confgiuration after upgrade
===============================
Stop all :term:`Ceph Object Gateway` running in the cluster.
Run the following commands:::
$ rados rmpool .rgw.root
$ radosgw-admin zonegroup get --rgw-zonegroup=default | sed 's/"id":.*/"id": "default",/g' | sed 's/"master_zone.*/"master_zone":"default",/g' > default-zg.json
$ raodsgw-admin zone get --zone-id=default > default-zone.json
$ radosgw-admin realm create --rgw-realm=myrealm
$ radosgw-admin zonegroup set --rgw-zonegroup=default --default < default-zg.json
$ radosgw-admin zone set --rgw-zone=default --default < default-zone.json
$ radosgw-admin period update --commit
Start all :term:`Ceph Object Gateway` in the cluster.

898
ceph/doc/release-notes.rst
View File

@ -2,6 +2,823 @@
Release Notes
===============
v12.1.0 Luminous (RC)
====================
This is the first release candidate for Luminous, the next long term stable release.
Major Changes from Kraken
-------------------------
* When assigning a network to the public network and not to
the cluster network the network specification of the public
network will be used for the cluster network as well.
In older versions this would lead to cluster services
being bound to 0.0.0.0:<port>, thus making the
cluster service even more publicly available than the
public services. When only specifying a cluster network it
will still result in the public services binding to 0.0.0.0.
* In previous versions, if a client sent an op to the wrong OSD, the OSD
would reply with ENXIO. The rationale here is that the client or OSD is
clearly buggy and we want to surface the error as clearly as possible.
We now only send the ENXIO reply if the osd_enxio_on_misdirected_op option
is enabled (it's off by default). This means that a VM using librbd that
previously would have gotten an EIO and gone read-only will now see a
blocked/hung IO instead.
* The "journaler allow split entries" config setting has been removed.
- *librados*:
* Some variants of the omap_get_keys and omap_get_vals librados
functions have been deprecated in favor of omap_get_vals2 and
omap_get_keys2. The new methods include an output argument
indicating whether there are additional keys left to fetch.
Previously this had to be inferred from the requested key count vs
the number of keys returned, but this breaks with new OSD-side
limits on the number of keys or bytes that can be returned by a
single omap request. These limits were introduced by kraken but
are effectively disabled by default (by setting a very large limit
of 1 GB) because users of the newly deprecated interface cannot
tell whether they should fetch more keys or not. In the case of
the standalone calls in the C++ interface
(IoCtx::get_omap_{keys,vals}), librados has been updated to loop on
the client side to provide a correct result via multiple calls to
the OSD. In the case of the methods used for building
multi-operation transactions, however, client-side looping is not
practical, and the methods have been deprecated. Note that use of
either the IoCtx methods on older librados versions or the
deprecated methods on any version of librados will lead to
incomplete results if/when the new OSD limits are enabled.
* The original librados rados_objects_list_open (C) and objects_begin
(C++) object listing API, deprecated in Hammer, has finally been
removed. Users of this interface must update their software to use
either the rados_nobjects_list_open (C) and nobjects_begin (C++) API or
the new rados_object_list_begin (C) and object_list_begin (C++) API
before updating the client-side librados library to Luminous.
Object enumeration (via any API) with the latest librados version
and pre-Hammer OSDs is no longer supported. Note that no in-tree
Ceph services rely on object enumeration via the deprecated APIs, so
only external librados users might be affected.
The newest (and recommended) rados_object_list_begin (C) and
object_list_begin (C++) API is only usable on clusters with the
SORTBITWISE flag enabled (Jewel and later). (Note that this flag is
required to be set before upgrading beyond Jewel.)
- *CephFS*:
* When configuring ceph-fuse mounts in /etc/fstab, a new syntax is
available that uses "ceph.<arg>=<val>" in the options column, instead
of putting configuration in the device column. The old style syntax
still works. See the documentation page "Mount CephFS in your
file systems table" for details.
* CephFS clients without the 'p' flag in their authentication capability
string will no longer be able to set quotas or any layout fields. This
flag previously only restricted modification of the pool and namespace
fields in layouts.
* CephFS directory fragmentation (large directory support) is enabled
by default on new filesystems. To enable it on existing filesystems
use "ceph fs set <fs_name> allow_dirfrags".
* CephFS will generate a health warning if you have fewer standby daemons
than it thinks you wanted. By default this will be 1 if you ever had
a standby, and 0 if you did not. You can customize this using
``ceph fs set <fs> standby_count_wanted <number>``. Setting it
to zero will effectively disable the health check.
* The "ceph mds tell ..." command has been removed. It is superceded
by "ceph tell mds.<id> ..."
- *MGR*
* ceph-mgr supports a default dashboard
* ceph-mgr introduces a new pecan based rest API
- *RGW*:
* RGW introduces server side encryption of uploaded objects with 3 options for
the management of encryption keys, automatic encryption (only recommended for
test setups), customer provided keys similar to Amazon SSE-C specification and
using a key management service (Openstack Barbician) similar to Amazon SSE-KMS
specification.
* RGW's metadata search with ElasticSearch now supports end user requests
serviced via RGW itself and now supports custom metadata fields
* RGW has consolidated the several metadata index pools via the use of rados
namespaces
* RGW now supports dynamic bucket index sharding
Notable Changes
---------------
* bluestore: ceph-disk: add --filestore argument, default to --bluestore (`pr#15437 <https://github.com/ceph/ceph/pull/15437>`_, Loic Dachary, Sage Weil)
* bluestore,core: os/bluestore: fix warning (`pr#15435 <https://github.com/ceph/ceph/pull/15435>`_, Sage Weil)
* bluestore,core: os/bluestore: improve mempool usage (`pr#15402 <https://github.com/ceph/ceph/pull/15402>`_, Sage Weil)
* bluestore,core: os/bluestore: write "mkfs_done" into disk only if we pass fsck() tests (`pr#15238 <https://github.com/ceph/ceph/pull/15238>`_, xie xingguo)
* bluestore,core: os: remove experimental status for BlueStore (`pr#15177 <https://github.com/ceph/ceph/pull/15177>`_, Sage Weil)
* bluestore: os/bluestore/BlockDevice: support pmem device as bluestore backend (`pr#15102 <https://github.com/ceph/ceph/pull/15102>`_, Jianpeng Ma)
* bluestore: os/bluestore: fix a typo about bleustore (`pr#15357 <https://github.com/ceph/ceph/pull/15357>`_, Dongsheng Yang)
* bluestore: os/bluestore: fix BitMapAllocator assert on out-of-bound hint value (`pr#15289 <https://github.com/ceph/ceph/pull/15289>`_, Igor Fedotov)
* bluestore: os/bluestore: fix buffers pinned by indefinitely deferred writes (`pr#15398 <https://github.com/ceph/ceph/pull/15398>`_, Sage Weil)
* bluestore: os/bluestore: fix false assert in IOContext::aio_wake (`pr#15268 <https://github.com/ceph/ceph/pull/15268>`_, Igor Fedotov)
* bluestore: os/bluestore: fix false asserts in Cache::trim_all() (`pr#15470 <https://github.com/ceph/ceph/pull/15470>`_, xie xingguo)
* bluestore: os/bluestore: fix fsck deferred_replay (`pr#15295 <https://github.com/ceph/ceph/pull/15295>`_, Sage Weil)
* bluestore: os/bluestore/KernelDevice: fix comments (`pr#15264 <https://github.com/ceph/ceph/pull/15264>`_, xie xingguo)
* bluestore: os/bluestore/KernelDevice: helpful warning when aio limit exhausted (`pr#15116 <https://github.com/ceph/ceph/pull/15116>`_, Sage Weil)
* bluestore,performance: os/bluestore: avoid overloading extents during reshard; atomic deferred_batch_ops (`pr#15502 <https://github.com/ceph/ceph/pull/15502>`_, xie xingguo)
* bluestore,performance: os/bluestore: batch throttle (`pr#15284 <https://github.com/ceph/ceph/pull/15284>`_, Jianpeng Ma)
* bluestore,performance: os/bluestore: keep statfs replica in RAM to avoid expensive KV retrieval (`pr#15309 <https://github.com/ceph/ceph/pull/15309>`_, Igor Fedotov)
* bluestore,performance: os/bluestore/KernelDevice: fix sync write vs flush (`pr#15034 <https://github.com/ceph/ceph/pull/15034>`_, Sage Weil)
* bluestore,performance: os/bluestore: move cache_trim into MempoolThread (`pr#15380 <https://github.com/ceph/ceph/pull/15380>`_, xie xingguo)
* bluestore,performance: os/bluestore: put bluefs in the middle of the shared device (`pr#14873 <https://github.com/ceph/ceph/pull/14873>`_, Sage Weil)
* bluestore,performance: os/bluestore: separate kv_sync_thread into two parts (`pr#14035 <https://github.com/ceph/ceph/pull/14035>`_, Jianpeng Ma, Igor Fedotov, Sage Weil)
* bluestore,performance: os/bluestore: try to unshare blobs for EC overwrite workload (`pr#14239 <https://github.com/ceph/ceph/pull/14239>`_, Sage Weil)
* build/ops: 12.0.3 (`pr#15600 <https://github.com/ceph/ceph/pull/15600>`_, Jenkins Build Slave User)
* build/ops: build: move bash_completion.d/ceph to ceph-common (`pr#15148 <https://github.com/ceph/ceph/pull/15148>`_, Leo Zhang)
* build/ops: build: remove ceph-disk-udev entirely (`pr#15259 <https://github.com/ceph/ceph/pull/15259>`_, Leo Zhang)
* build/ops: build: revert -Wvla from #15342 (`pr#15469 <https://github.com/ceph/ceph/pull/15469>`_, Willem Jan Withagen)
* build/ops: build: Use .S suffix for ppc64le assembly files (`issue#20106 <http://tracker.ceph.com/issues/20106>`_, `pr#15373 <https://github.com/ceph/ceph/pull/15373>`_, Andrew Solomon)
* build/ops: ceph-detect-init: detect init system by poking the system (`issue#19884 <http://tracker.ceph.com/issues/19884>`_, `pr#15043 <https://github.com/ceph/ceph/pull/15043>`_, Kefu Chai)
* build/ops,common: build: Adds C++ warning flag for C Variable-Length Arrays. (`pr#15342 <https://github.com/ceph/ceph/pull/15342>`_, Jesse Williamson)
* build/ops,common: common/blkdev.cc: propagate get_device_by_fd to different OSes (`pr#15547 <https://github.com/ceph/ceph/pull/15547>`_, Willem Jan Withagen)
* build/ops: conditionalize rgw Beast frontend so it isn't built on s390x architecture (`issue#20048 <http://tracker.ceph.com/issues/20048>`_, `pr#15225 <https://github.com/ceph/ceph/pull/15225>`_, Willem Jan Withagen, Nathan Cutler, Kefu Chai, Tim Serong, Casey Bodley)
* build/ops,core,tests: osd/dmclock/testing: reorganize testing, building now optional (`pr#15375 <https://github.com/ceph/ceph/pull/15375>`_, J. Eric Ivancich)
* build/ops: debian: ceph-mgr: fix package description (`pr#15513 <https://github.com/ceph/ceph/pull/15513>`_, Fabian Grünbichler)
* build/ops: debian: sync logrotate packaging with downstream (`issue#19938 <http://tracker.ceph.com/issues/19938>`_, `pr#15567 <https://github.com/ceph/ceph/pull/15567>`_, Fabian Grünbichler)
* build/ops: do_cmake.sh: enable ccache if installed (`pr#15274 <https://github.com/ceph/ceph/pull/15274>`_, Sage Weil)
* build/ops: drop libfcgi build dependency (`pr#15285 <https://github.com/ceph/ceph/pull/15285>`_, Nathan Cutler)
* build/ops: install-deps.sh: workaround setuptools' dependency on six (`pr#15406 <https://github.com/ceph/ceph/pull/15406>`_, Kefu Chai)
* build/ops: rpm: apply epoch only if %epoch macro is defined (`pr#15286 <https://github.com/ceph/ceph/pull/15286>`_, Nathan Cutler)
* build/ops: rpm: make librbd1 %post scriptlet depend on coreutils (`issue#20052 <http://tracker.ceph.com/issues/20052>`_, `pr#15231 <https://github.com/ceph/ceph/pull/15231>`_, Giacomo Comes, Nathan Cutler)
* build/ops: rpm: move _epoch_prefix below Epoch definition (`pr#15417 <https://github.com/ceph/ceph/pull/15417>`_, Nathan Cutler)
* build/ops: rpm: move RDMA and python-prettytables build dependencies to distro-conditional section (`pr#15200 <https://github.com/ceph/ceph/pull/15200>`_, Nathan Cutler)
* build/ops: selinux: Allow read on var_run_t (`issue#16674 <http://tracker.ceph.com/issues/16674>`_, `pr#15523 <https://github.com/ceph/ceph/pull/15523>`_, Boris Ranto)
* build/ops: selinux: Do parallel relabel on package install (`issue#20077 <http://tracker.ceph.com/issues/20077>`_, `pr#14871 <https://github.com/ceph/ceph/pull/14871>`_, Boris Ranto)
* build/ops: selinux: Install ceph-base before ceph-selinux (`issue#20184 <http://tracker.ceph.com/issues/20184>`_, `pr#15490 <https://github.com/ceph/ceph/pull/15490>`_, Boris Ranto)
* build/ops: Set subman cron attributes in spec file (`issue#20074 <http://tracker.ceph.com/issues/20074>`_, `pr#15270 <https://github.com/ceph/ceph/pull/15270>`_, Thomas Serlin)
* build/ops: The Clangtastic Mr. Clocks (`pr#15186 <https://github.com/ceph/ceph/pull/15186>`_, Adam C. Emerson)
* build/ops: yasm-wrapper: filter -pthread (`pr#15249 <https://github.com/ceph/ceph/pull/15249>`_, Alessandro Barbieri)
* cephfs: #17980: MDS client blacklisting and blacklist on eviction (`issue#17980 <http://tracker.ceph.com/issues/17980>`_, `issue#9754 <http://tracker.ceph.com/issues/9754>`_, `pr#14610 <https://github.com/ceph/ceph/pull/14610>`_, John Spray)
* cephfs: ceph: simplify CInode::maybe_export_pin() (`pr#15106 <https://github.com/ceph/ceph/pull/15106>`_, "Yan, Zheng")
* cephfs: client: fix display ino in the ldout (`pr#15314 <https://github.com/ceph/ceph/pull/15314>`_, huanwen ren)
* cephfs: client/inode: fix the dump type of Inode::dump() (`pr#15198 <https://github.com/ceph/ceph/pull/15198>`_, huanwen ren)
* cephfs,common,rbd: blkin: librbd trace hooks (`pr#15053 <https://github.com/ceph/ceph/pull/15053>`_, Victor Araujo, Jason Dillaman)
* cephfs: mon/FSCommand: fix indentation (`pr#15423 <https://github.com/ceph/ceph/pull/15423>`_, Sage Weil)
* cephfs: mon/MDSMonitor: respect mds_standby_for_rank config (`pr#15129 <https://github.com/ceph/ceph/pull/15129>`_, "Yan, Zheng")
* cephfs: osdc/Journaler: avoid executing on_safe contexts prematurely (`issue#20055 <http://tracker.ceph.com/issues/20055>`_, `pr#15240 <https://github.com/ceph/ceph/pull/15240>`_, "Yan, Zheng")
* cephfs: qa/cephfs: disable mds_bal_frag for TestStrays.test_purge_queue_op_rate (`issue#19892 <http://tracker.ceph.com/issues/19892>`_, `pr#15105 <https://github.com/ceph/ceph/pull/15105>`_, "Yan, Zheng")
* cephfs: qa/tasks/cephfs: use getattr to guarantee inode is in client cache (`issue#19912 <http://tracker.ceph.com/issues/19912>`_, `pr#15062 <https://github.com/ceph/ceph/pull/15062>`_, "Yan, Zheng")
* cephfs: qa: update log whitelists for kcephfs suite (`pr#14922 <https://github.com/ceph/ceph/pull/14922>`_, "Yan, Zheng")
* cephfs,tests: qa: fix float parse error in test_fragment (`pr#15122 <https://github.com/ceph/ceph/pull/15122>`_, Patrick Donnelly)
* cephfs,tests: qa: silence upgrade test failure (`issue#19934 <http://tracker.ceph.com/issues/19934>`_, `pr#15126 <https://github.com/ceph/ceph/pull/15126>`_, Patrick Donnelly)
* cephfs,tests: qa: simplify TestJournalRepair (`pr#15096 <https://github.com/ceph/ceph/pull/15096>`_, John Spray)
* cleanup: src: put-to operator function - const input cleanup (`issue#3977 <http://tracker.ceph.com/issues/3977>`_, `pr#15364 <https://github.com/ceph/ceph/pull/15364>`_, Jos Collin)
* cmake: Add -finstrument-functions flag to OSD code (`pr#15055 <https://github.com/ceph/ceph/pull/15055>`_, Mohamad Gebai)
* cmake: check the existence of gperf before using it (`pr#15164 <https://github.com/ceph/ceph/pull/15164>`_, Kefu Chai)
* cmake: do not link libcommon against some libs (`pr#15340 <https://github.com/ceph/ceph/pull/15340>`_, Willem Jan Withagen)
* cmake: fix boost components for WITH_SYSTEM_BOOST (`pr#15160 <https://github.com/ceph/ceph/pull/15160>`_, Bassam Tabbara)
* cmake: improved build speed by 5x when using ccache (`pr#15147 <https://github.com/ceph/ceph/pull/15147>`_, Bassam Tabbara)
* cmake: link against fcgi only if enabled (`pr#15425 <https://github.com/ceph/ceph/pull/15425>`_, Yao Zongyou)
* cmake: misc fixes for build on i386 (`pr#15516 <https://github.com/ceph/ceph/pull/15516>`_, James Page)
* cmake: rgw: do not link against boost in a wholesale (`pr#15347 <https://github.com/ceph/ceph/pull/15347>`_, Nathan Cutler, Kefu Chai)
* cmake: workaound ccache issue with .S assembly files (`pr#15142 <https://github.com/ceph/ceph/pull/15142>`_, Bassam Tabbara)
* common: add ceph::size() (`pr#15181 <https://github.com/ceph/ceph/pull/15181>`_, Kefu Chai)
* common: ceph_osd: remove client message cap limit (`pr#14944 <https://github.com/ceph/ceph/pull/14944>`_, Haomai Wang)
* common: cls: optimize header file dependency (`pr#15165 <https://github.com/ceph/ceph/pull/15165>`_, Brad Hubbard, Xiaowei Chen)
* common: cmdparse: more constness (`pr#15023 <https://github.com/ceph/ceph/pull/15023>`_, Kefu Chai)
* common: common/ceph_context: 'config diff get' option added (`pr#10736 <https://github.com/ceph/ceph/pull/10736>`_, Daniel Oliveira)
* common: common/ceph_context: fix leak of registered commands on exit (`pr#15302 <https://github.com/ceph/ceph/pull/15302>`_, xie xingguo)
* common: common/iso_8601.cc: Make return expression Clang compatible (`pr#15336 <https://github.com/ceph/ceph/pull/15336>`_, Willem Jan Withagen)
* common: common/LogEntry: include EntityName in log entries (`pr#15395 <https://github.com/ceph/ceph/pull/15395>`_, Sage Weil)
* common: common,osdc: remove atomic_t completely (`pr#15562 <https://github.com/ceph/ceph/pull/15562>`_, Kefu Chai)
* common: common/perf_counters: add average time for PERFCOUNTER_TIME (`pr#15478 <https://github.com/ceph/ceph/pull/15478>`_, xie xingguo)
* common: common/perf_counters: make schema more friendly and update docs (`pr#14933 <https://github.com/ceph/ceph/pull/14933>`_, Sage Weil)
* common: common,test: migrate atomic_t to std::atomic (`pr#14866 <https://github.com/ceph/ceph/pull/14866>`_, Jesse Williamson)
* common,core: ceph_test_rados_api_misc: fix LibRadosMiscConnectFailure.ConnectFailure retry (`issue#19901 <http://tracker.ceph.com/issues/19901>`_, `pr#15522 <https://github.com/ceph/ceph/pull/15522>`_, Sage Weil)
* common,core: osd/OSDMap: make osd_state 32 bits wide (`pr#15390 <https://github.com/ceph/ceph/pull/15390>`_, Sage Weil)
* common,core: osd/OSDMap: replace require_*_osds flags with a single require_osd_release field (`pr#15068 <https://github.com/ceph/ceph/pull/15068>`_, Sage Weil)
* common,core: osd/OSDMap: replace string-based min_compat_client with a CEPH_RELEASE_* uint8_t (`pr#15351 <https://github.com/ceph/ceph/pull/15351>`_, Sage Weil)
* common: crc32c: include acconfig.h to fix ceph_crc32c_aarch64() (`pr#15515 <https://github.com/ceph/ceph/pull/15515>`_, Kefu Chai)
* common: crush/CrushWrapper: fix has_incompat_choose_args (`pr#15218 <https://github.com/ceph/ceph/pull/15218>`_, Sage Weil)
* common: crush/CrushWrapper: fix has_incompat_choose_args() (`pr#15244 <https://github.com/ceph/ceph/pull/15244>`_, Sage Weil)
* common: denc: add encode/decode for basic_sstring (`pr#15135 <https://github.com/ceph/ceph/pull/15135>`_, Kefu Chai, Casey Bodley)
* common: get_process_name: use getprogname on bsd systems (`pr#15338 <https://github.com/ceph/ceph/pull/15338>`_, Mykola Golub)
* common: Improved CRC calculation for zero buffers (`pr#11966 <https://github.com/ceph/ceph/pull/11966>`_, Adam Kupczyk)
* common: include/lru.h: add const to member functions (`pr#15408 <https://github.com/ceph/ceph/pull/15408>`_, yonghengdexin735)
* common: include/rados: Fix typo in rados_ioctx_cct() doc (`pr#15220 <https://github.com/ceph/ceph/pull/15220>`_, Jos Collin)
* common: include: Redo some includes for FreeBSD (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15337 <https://github.com/ceph/ceph/pull/15337>`_, Willem Jan Withagen)
* common: int_types.h: remove hacks to workaround old systems (`pr#15069 <https://github.com/ceph/ceph/pull/15069>`_, Kefu Chai)
* common: librados,libradosstriper,test: migrate atomic_t to std::atomic (baragon) (`pr#14658 <https://github.com/ceph/ceph/pull/14658>`_, Jesse Williamson)
* common: libradosstriper: Add example code (`pr#15350 <https://github.com/ceph/ceph/pull/15350>`_, Logan Blyth)
* common: mempool: improve dump; fix buffer accounting bugs (`pr#15403 <https://github.com/ceph/ceph/pull/15403>`_, Sage Weil)
* common,mon: messenger,client,compressor: migrate atomic_t to std::atomic (`pr#14657 <https://github.com/ceph/ceph/pull/14657>`_, Jesse Williamson)
* common,mon: mon,crush: add 'osd crush swap-bucket' command (`pr#15072 <https://github.com/ceph/ceph/pull/15072>`_, Sage Weil)
* common,performance: buffer: allow buffers to be accounted in arbitrary mempools (`pr#15352 <https://github.com/ceph/ceph/pull/15352>`_, Sage Weil)
* common,performance: crc32c: Add ppc64le fast zero optimized assembly. (`pr#15100 <https://github.com/ceph/ceph/pull/15100>`_, Andrew Solomon)
* common,performance: inline_memory: optimized mem_is_zero for non-x64 (`pr#15307 <https://github.com/ceph/ceph/pull/15307>`_, Piotr Dałek)
* common,performance: kv/rocksdb: supports SliceParts interface (`pr#15058 <https://github.com/ceph/ceph/pull/15058>`_, Haomai Wang)
* common,performance: osd/OSDMap: make pg_temp more efficient (`pr#15291 <https://github.com/ceph/ceph/pull/15291>`_, Sage Weil)
* common: Remove redundant includes - 2 (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15169 <https://github.com/ceph/ceph/pull/15169>`_, Jos Collin)
* common: Remove redundant includes - 3 (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15204 <https://github.com/ceph/ceph/pull/15204>`_, Jos Collin)
* common: Remove redundant includes - 5 (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15267 <https://github.com/ceph/ceph/pull/15267>`_, Jos Collin)
* common: Remove redundant includes - 6 (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15299 <https://github.com/ceph/ceph/pull/15299>`_, Jos Collin)
* common: Remove redundant includes (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15042 <https://github.com/ceph/ceph/pull/15042>`_, Brad Hubbard)
* common: Remove redundant includes (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15086 <https://github.com/ceph/ceph/pull/15086>`_, Jos Collin)
* common,tests: ceph_test_rados_api_list: more fix LibRadosListNP.ListObjectsError (`issue#19963 <http://tracker.ceph.com/issues/19963>`_, `pr#15138 <https://github.com/ceph/ceph/pull/15138>`_, Sage Weil)
* common: xio: migrate atomic_t to std::atomic<> (`pr#15230 <https://github.com/ceph/ceph/pull/15230>`_, Jesse Williamson)
* core: ceph-disk: do not setup_statedir on trigger (`issue#19941 <http://tracker.ceph.com/issues/19941>`_, `pr#15410 <https://github.com/ceph/ceph/pull/15410>`_, Loic Dachary)
* core: compressor: add LZ4 support (`pr#15434 <https://github.com/ceph/ceph/pull/15434>`_, Haomai Wang)
* core: compressor: optimize header file dependency (`pr#15187 <https://github.com/ceph/ceph/pull/15187>`_, Brad Hubbard, Xiaowei Chen)
* core: crush, mon: make jewel the lower bound for client/crush compat for new clusters (`pr#15370 <https://github.com/ceph/ceph/pull/15370>`_, Sage Weil)
* core: erasure-code: optimize header file dependency (`pr#15172 <https://github.com/ceph/ceph/pull/15172>`_, Brad Hubbard, Xiaowei Chen)
* core: erasure-code: Remove duplicate of isa-l files (`pr#15372 <https://github.com/ceph/ceph/pull/15372>`_, Ganesh Mahalingam)
* core: filestore: migrate atomic_t to std::atomic<> (`pr#15228 <https://github.com/ceph/ceph/pull/15228>`_, Jesse Williamson)
* core: include/types.h, introduce host_to_ceph_errno (`pr#15496 <https://github.com/ceph/ceph/pull/15496>`_, Willem Jan Withagen)
* core: Install Pecan for FreeBSD (`pr#15610 <https://github.com/ceph/ceph/pull/15610>`_, Willem Jan Withagen)
* core,mgr: mgr/DaemonServer: stop spamming log with pg stats (`pr#15487 <https://github.com/ceph/ceph/pull/15487>`_, Sage Weil)
* core,mgr,mon: mon/PGMap: fix osd_epoch update when removing osd_stat (`issue#20208 <http://tracker.ceph.com/issues/20208>`_, `pr#15573 <https://github.com/ceph/ceph/pull/15573>`_, Sage Weil)
* core,mon: mon/LogMonitor: 'log last' command (`pr#15497 <https://github.com/ceph/ceph/pull/15497>`_, Sage Weil)
* core,mon: mon/OSDMonitor: cancel mapping job from update_from_paxos (`issue#20067 <http://tracker.ceph.com/issues/20067>`_, `pr#15320 <https://github.com/ceph/ceph/pull/15320>`_, Sage Weil)
* core,mon: mon/OSDMonitor: use up set instead of acting set in reweight_by_utilization (`pr#13802 <https://github.com/ceph/ceph/pull/13802>`_, Mingxin Liu)
* core,mon: mon/PGMap: call blocked requests ERR not WARN (`pr#15501 <https://github.com/ceph/ceph/pull/15501>`_, Sage Weil)
* core: mon/OSDMonitor: change info in 'osd failed' messages (`pr#15321 <https://github.com/ceph/ceph/pull/15321>`_, Sage Weil)
* core: mon,osd/OSDMap: a couple pg-upmap fixes (`pr#15319 <https://github.com/ceph/ceph/pull/15319>`_, Sage Weil)
* core: msg/async: avoid requeue racing with handle_write (`issue#20093 <http://tracker.ceph.com/issues/20093>`_, `pr#15324 <https://github.com/ceph/ceph/pull/15324>`_, Haomai Wang)
* core: osd,librados: add manifest, redirect (`pr#15325 <https://github.com/ceph/ceph/pull/15325>`_, Sage Weil)
* core: osd/OSDMap: improve upmap calculation (`issue#19818 <http://tracker.ceph.com/issues/19818>`_, `pr#14902 <https://github.com/ceph/ceph/pull/14902>`_, Sage Weil)
* core: osd/PG: drop pre-firefly compat_mode for choose_*_acting (`pr#15057 <https://github.com/ceph/ceph/pull/15057>`_, Sage Weil)
* core: osd/pglog: remove loop through empty collection (`pr#15121 <https://github.com/ceph/ceph/pull/15121>`_, J. Eric Ivancich)
* core,performance: msg/async: reduce write_lock contention (`pr#15092 <https://github.com/ceph/ceph/pull/15092>`_, Haomai Wang)
* core,rgw: qa: Removed all 'default_idle_timeout' due to chnage in rwg task (`pr#15420 <https://github.com/ceph/ceph/pull/15420>`_, Yuri Weinstein)
* core,rgw,tests: qa/rgw_snaps: move default_idle_timeout config under the client (`issue#20128 <http://tracker.ceph.com/issues/20128>`_, `pr#15400 <https://github.com/ceph/ceph/pull/15400>`_, Yehuda Sadeh)
* core: src/ceph.in: Use env(CEPH_DEV) to suppress noise from ceph (`pr#14746 <https://github.com/ceph/ceph/pull/14746>`_, Willem Jan Withagen)
* core,tests: ceph-disk: sensible default for block.db (`pr#15576 <https://github.com/ceph/ceph/pull/15576>`_, Loic Dachary)
* core,tests: qa/suites/rados/*/at-end: wait for healthy before scrubbing (`pr#15245 <https://github.com/ceph/ceph/pull/15245>`_, Sage Weil)
* core,tests: qa/suites/rados/singleton-nomsg/health-warnings: behave on ext4 (`issue#20043 <http://tracker.ceph.com/issues/20043>`_, `pr#15207 <https://github.com/ceph/ceph/pull/15207>`_, Sage Weil)
* core,tests: qa/suites/rados: temporarily remove scrub_test from basic/ until post-luminous (`issue#19935 <http://tracker.ceph.com/issues/19935>`_, `pr#15202 <https://github.com/ceph/ceph/pull/15202>`_, Sage Weil)
* core,tests: qa/suites/upgrade/kraken-x: enable experimental for bluestore (`pr#15359 <https://github.com/ceph/ceph/pull/15359>`_, Sage Weil)
* core,tests: qa/workunits/cephtool/test.sh: fix osd full health detail grep (`issue#20187 <http://tracker.ceph.com/issues/20187>`_, `pr#15494 <https://github.com/ceph/ceph/pull/15494>`_, Sage Weil)
* core,tests: qa/workunits/rados/test_health_warning: misc fixes (`issue#19990 <http://tracker.ceph.com/issues/19990>`_, `pr#15201 <https://github.com/ceph/ceph/pull/15201>`_, Sage Weil)
* core,tests: test/osd/TestRados.cc: run set-redirect test after finishing setup (`issue#20114 <http://tracker.ceph.com/issues/20114>`_, `pr#15385 <https://github.com/ceph/ceph/pull/15385>`_, Myoungwon Oh)
* core,tools: osdmaptool: require --upmap-save before modifying input osdmap (`pr#15247 <https://github.com/ceph/ceph/pull/15247>`_, Sage Weil)
* crush: add missing tunable in tests (`pr#15412 <https://github.com/ceph/ceph/pull/15412>`_, Loic Dachary)
* crush: encode can override weights with weight set (`issue#19836 <http://tracker.ceph.com/issues/19836>`_, `pr#15002 <https://github.com/ceph/ceph/pull/15002>`_, Loic Dachary)
* crush: optimize header file dependency (`pr#9307 <https://github.com/ceph/ceph/pull/9307>`_, Xiaowei Chen)
* crush: update choose_args when items are added/removed (`pr#15311 <https://github.com/ceph/ceph/pull/15311>`_, Loic Dachary)
* doc: add descriptions for mon/mgr options (`pr#15032 <https://github.com/ceph/ceph/pull/15032>`_, Kefu Chai)
* doc: add FreeBSD manual install (`pr#14941 <https://github.com/ceph/ceph/pull/14941>`_, Willem Jan Withagen)
* doc: add new cn ceph mirror to doc and mirroring (`pr#15089 <https://github.com/ceph/ceph/pull/15089>`_, Shengjing Zhu)
* doc: add rados xattr commands to manpage (`pr#15362 <https://github.com/ceph/ceph/pull/15362>`_, Andreas Gerstmayr)
* doc: add README to dmclock subdir to inform developers it's a git subtree (`pr#15386 <https://github.com/ceph/ceph/pull/15386>`_, J. Eric Ivancich)
* doc: AUTHORS: update with release manager, backport team (`pr#15391 <https://github.com/ceph/ceph/pull/15391>`_, Sage Weil)
* doc: Change the default values of some OSD options (`issue#20199 <http://tracker.ceph.com/issues/20199>`_, `pr#15566 <https://github.com/ceph/ceph/pull/15566>`_, Bara Ancincova)
* doc: describe CephFS max_file_size (`pr#15287 <https://github.com/ceph/ceph/pull/15287>`_, Ken Dreyer)
* doc: dev improve the s3tests doc to reflect current scripts (`pr#15180 <https://github.com/ceph/ceph/pull/15180>`_, Abhishek Lekshmanan)
* doc: doc/cephfs: mention RADOS object size limit (`pr#15550 <https://github.com/ceph/ceph/pull/15550>`_, John Spray)
* doc: doc/release-notes: update which jewel version does sortbitwise warning (`pr#15209 <https://github.com/ceph/ceph/pull/15209>`_, Sage Weil)
* doc: doc/rgw: remove fastcgi page and sample configs (`pr#15133 <https://github.com/ceph/ceph/pull/15133>`_, Casey Bodley)
* doc: Documentation Fixes for http://tracker.ceph.com/issues/19879 (`issue#20057 <http://tracker.ceph.com/issues/20057>`_, `issue#19879 <http://tracker.ceph.com/issues/19879>`_, `pr#15606 <https://github.com/ceph/ceph/pull/15606>`_, Sameer Tiwari)
* doc: document perf historgrams (`pr#15150 <https://github.com/ceph/ceph/pull/15150>`_, Piotr Dałek)
* doc: Document RGW quota cache options (`issue#18747 <http://tracker.ceph.com/issues/18747>`_, `pr#13395 <https://github.com/ceph/ceph/pull/13395>`_, Daniel Gryniewicz)
* doc: fix broken link in erasure-code.rst (`issue#19972 <http://tracker.ceph.com/issues/19972>`_, `pr#15143 <https://github.com/ceph/ceph/pull/15143>`_, MinSheng Lin)
* doc: fix factual inaccuracy in doc/architecture.rst (`pr#15235 <https://github.com/ceph/ceph/pull/15235>`_, Nathan Cutler, Sage Weil)
* doc: fixing an error in 12.0.3 release notes (`pr#15195 <https://github.com/ceph/ceph/pull/15195>`_, Abhishek Lekshmanan)
* doc: fix syntax on code snippets in cephfs/multimds (`pr#15499 <https://github.com/ceph/ceph/pull/15499>`_, John Spray)
* doc: kill some broken links (`pr#15203 <https://github.com/ceph/ceph/pull/15203>`_, liuchang0812)
* doc: mailmap: Leo Zhang infomation and affiliation (`pr#15145 <https://github.com/ceph/ceph/pull/15145>`_, Leo Zhang)
* doc: mention certain conf vars should be in global (`pr#15119 <https://github.com/ceph/ceph/pull/15119>`_, Ali Maredia)
* doc: Merge pull request from stiwari/wip-19879 (`issue#19879 <http://tracker.ceph.com/issues/19879>`_, `pr#15609 <https://github.com/ceph/ceph/pull/15609>`_, Sameer Tiwari)
* doc: minor fixes in radosgw/ (`pr#15103 <https://github.com/ceph/ceph/pull/15103>`_, Drunkard Zhang)
* doc: PendingReleaseNotes: notes on whiteouts vs pgnls (`pr#15575 <https://github.com/ceph/ceph/pull/15575>`_, Sage Weil)
* doc: PendingReleaseNotes: warning about 'osd rm ...' and #19119 (`issue#19119 <http://tracker.ceph.com/issues/19119>`_, `pr#13731 <https://github.com/ceph/ceph/pull/13731>`_, Sage Weil)
* doc: release-notes clarify about rgw encryption (`pr#14800 <https://github.com/ceph/ceph/pull/14800>`_, Abhishek Lekshmanan)
* doc: release notes for v12.0.3 (dev) (`pr#15090 <https://github.com/ceph/ceph/pull/15090>`_, Abhishek Lekshmanan)
* docs document "osd recovery max single start" setting (`issue#17396 <http://tracker.ceph.com/issues/17396>`_, `pr#15275 <https://github.com/ceph/ceph/pull/15275>`_, Ken Dreyer)
* doc: typo fixes on hyperlink/words (`pr#15144 <https://github.com/ceph/ceph/pull/15144>`_, Drunkard Zhang)
* doc: update sample explaning "%" operator in test suites (`pr#15511 <https://github.com/ceph/ceph/pull/15511>`_, Kefu Chai)
* doc: Update some RGW documentation (`pr#15175 <https://github.com/ceph/ceph/pull/15175>`_, Jens Rosenboom)
* doc: update the usage of 'ceph-deploy purge' (`pr#15080 <https://github.com/ceph/ceph/pull/15080>`_, Yu Shengzuo)
* doc: use do_cmake.sh instead of `cmake ..` (`pr#15110 <https://github.com/ceph/ceph/pull/15110>`_, Kefu Chai)
* librbd: discard related IO should skip op if object non-existent (`issue#19962 <http://tracker.ceph.com/issues/19962>`_, `pr#15239 <https://github.com/ceph/ceph/pull/15239>`_, Mykola Golub)
* librbd: do not raise an error if trash list returns -ENOENT (`pr#15085 <https://github.com/ceph/ceph/pull/15085>`_, runsisi)
* librbd: filter expected error codes from is_exclusive_lock_owner (`issue#20182 <http://tracker.ceph.com/issues/20182>`_, `pr#15483 <https://github.com/ceph/ceph/pull/15483>`_, Jason Dillaman)
* librbd: fix valgrind errors and ensure tests detect future leaks (`pr#15415 <https://github.com/ceph/ceph/pull/15415>`_, Jason Dillaman)
* librbd: optimize copy-up to add hints only once to object op (`issue#19875 <http://tracker.ceph.com/issues/19875>`_, `pr#15037 <https://github.com/ceph/ceph/pull/15037>`_, Mykola Golub)
* librbd: potential read IO hang when image is flattened (`issue#19832 <http://tracker.ceph.com/issues/19832>`_, `pr#15234 <https://github.com/ceph/ceph/pull/15234>`_, Jason Dillaman)
* librbd: reacquire lock should update lock owner client id (`issue#19929 <http://tracker.ceph.com/issues/19929>`_, `pr#15093 <https://github.com/ceph/ceph/pull/15093>`_, Jason Dillaman)
* librbd: reduce potential of erroneous blacklisting on image close (`issue#19970 <http://tracker.ceph.com/issues/19970>`_, `pr#15162 <https://github.com/ceph/ceph/pull/15162>`_, Jason Dillaman)
* librbd: remove unused rbd_image_options_t ostream operator (`pr#15443 <https://github.com/ceph/ceph/pull/15443>`_, Mykola Golub)
* mds: change the type of data_pools (`pr#15278 <https://github.com/ceph/ceph/pull/15278>`_, Vicente Cheng)
* mds: check export pin during replay (`issue#20039 <http://tracker.ceph.com/issues/20039>`_, `pr#15205 <https://github.com/ceph/ceph/pull/15205>`_, Patrick Donnelly)
* mds: fix CDir::merge() for mds_debug_auth_pins (`issue#19946 <http://tracker.ceph.com/issues/19946>`_, `pr#15130 <https://github.com/ceph/ceph/pull/15130>`_, "Yan, Zheng")
* mds: fix client ID truncation (`pr#15258 <https://github.com/ceph/ceph/pull/15258>`_, Henry Chang)
* mds: limit client writable range increment (`issue#19955 <http://tracker.ceph.com/issues/19955>`_, `pr#15131 <https://github.com/ceph/ceph/pull/15131>`_, "Yan, Zheng")
* mds: miscellaneous multimds fixes (`pr#14550 <https://github.com/ceph/ceph/pull/14550>`_, "Yan, Zheng")
* mds: Pass empty string to clear mantle balancer (`issue#20076 <http://tracker.ceph.com/issues/20076>`_, `pr#15282 <https://github.com/ceph/ceph/pull/15282>`_, Zhi Zhang)
* mds: properly create aux subtrees for pinned directory (`issue#20083 <http://tracker.ceph.com/issues/20083>`_, `pr#15300 <https://github.com/ceph/ceph/pull/15300>`_, "Yan, Zheng")
* mgr: ceph-create-keys: update client.admin if it already exists (`issue#19940 <http://tracker.ceph.com/issues/19940>`_, `pr#15112 <https://github.com/ceph/ceph/pull/15112>`_, John Spray)
* mgr: ceph: introduce "tell x help" subcommand (`issue#19885 <http://tracker.ceph.com/issues/19885>`_, `pr#15111 <https://github.com/ceph/ceph/pull/15111>`_, liuchang0812)
* mgr: ceph-mgr: Implement new pecan-based rest api (`pr#14457 <https://github.com/ceph/ceph/pull/14457>`_, Boris Ranto)
* mgr: cleanup, stop clients sending in perf counters (`pr#15578 <https://github.com/ceph/ceph/pull/15578>`_, John Spray)
* mgr: dashboard code cleanup (`pr#15577 <https://github.com/ceph/ceph/pull/15577>`_, John Spray)
* mgr: dashboard GUI module (`pr#14946 <https://github.com/ceph/ceph/pull/14946>`_, John Spray, Dan Mick)
* mgr: load modules in separate python sub-interpreters (`pr#14971 <https://github.com/ceph/ceph/pull/14971>`_, Tim Serong)
* mgr: Mark session connections down on shutdown (`issue#19900 <http://tracker.ceph.com/issues/19900>`_, `pr#15192 <https://github.com/ceph/ceph/pull/15192>`_, Brad Hubbard)
* mgr: mgr/DaemonServer.cc: log daemon type string as well as id (`pr#15560 <https://github.com/ceph/ceph/pull/15560>`_, Dan Mick)
* mgr: mgr/MgrStandby: prevent use-after-free on just-shut-down Mgr (`issue#19595 <http://tracker.ceph.com/issues/19595>`_, `pr#15297 <https://github.com/ceph/ceph/pull/15297>`_, Sage Weil)
* mgr: mgr/MgrStandby: respawn when deactivated (`issue#19595 <http://tracker.ceph.com/issues/19595>`_, `issue#19549 <http://tracker.ceph.com/issues/19549>`_, `pr#15557 <https://github.com/ceph/ceph/pull/15557>`_, Sage Weil)
* mgr: mgr,osd: ceph-mgr --help, unify usage text of other daemons (`pr#15176 <https://github.com/ceph/ceph/pull/15176>`_, Tim Serong)
* mgr,mon: mon,mgr: extricate PGmap from monitor (`issue#20067 <http://tracker.ceph.com/issues/20067>`_, `issue#20174 <http://tracker.ceph.com/issues/20174>`_, `issue#20050 <http://tracker.ceph.com/issues/20050>`_, `pr#15073 <https://github.com/ceph/ceph/pull/15073>`_, Kefu Chai, Sage Weil, Greg Farnum)
* mgr,mon: mon/MgrMonitor: add 'mgr dump [epoch]' command (`pr#15158 <https://github.com/ceph/ceph/pull/15158>`_, Sage Weil)
* mgr: optimize DaemonStateIndex::cull() a little bit (`pr#14967 <https://github.com/ceph/ceph/pull/14967>`_, Kefu Chai)
* mgr: pybind/mgr/dashboard: monkeypatch os.exit to stop cherrypy from taking down mgr (`issue#20216 <http://tracker.ceph.com/issues/20216>`_, `pr#15588 <https://github.com/ceph/ceph/pull/15588>`_, Sage Weil)
* mgr: pybind/mgr: Delete `rest` module (`pr#15429 <https://github.com/ceph/ceph/pull/15429>`_, John Spray)
* mgr: pybind/mgr/restful: improve cert handling; work with vstart (`pr#15405 <https://github.com/ceph/ceph/pull/15405>`_, Sage Weil)
* mon: add crush type down health warnings (`pr#14914 <https://github.com/ceph/ceph/pull/14914>`_, Neha Ojha)
* mon: Add override for FsNewHandler::handle() (`pr#15331 <https://github.com/ceph/ceph/pull/15331>`_, yonghengdexin735)
* mon: cleanups (`pr#15272 <https://github.com/ceph/ceph/pull/15272>`_, Kefu Chai)
* mon: delete useless function definition (`pr#15188 <https://github.com/ceph/ceph/pull/15188>`_, shiqi)
* mon: don't prefix mgr summary with epoch number (`pr#15512 <https://github.com/ceph/ceph/pull/15512>`_, John Spray)
* mon: fix accesing pending_fsmap from peon (`issue#20040 <http://tracker.ceph.com/issues/20040>`_, `pr#15213 <https://github.com/ceph/ceph/pull/15213>`_, John Spray)
* mon: fix a few bugs with the osd health reporting (`pr#15179 <https://github.com/ceph/ceph/pull/15179>`_, Sage Weil)
* mon: Fixed typo in function comment blocks and in other comments (`pr#15304 <https://github.com/ceph/ceph/pull/15304>`_, linbing)
* mon: Fixed typo in @post of _active() (`pr#15191 <https://github.com/ceph/ceph/pull/15191>`_, Linbing)
* mon: fix mon_keyvaluedb application (`pr#15059 <https://github.com/ceph/ceph/pull/15059>`_, Sage Weil)
* mon: it's no need to get pg action_primary osd twice in pg scrub (`pr#15313 <https://github.com/ceph/ceph/pull/15313>`_, linbing)
* mon: mon/MgrMonitor: send digests only if is_active() (`pr#15109 <https://github.com/ceph/ceph/pull/15109>`_, Kefu Chai)
* mon: mon/MonClient: cancel pending commands on shutdown (`issue#20051 <http://tracker.ceph.com/issues/20051>`_, `pr#15227 <https://github.com/ceph/ceph/pull/15227>`_, Kefu Chai, Sage Weil)
* mon: {mon,osd,mds} {versions,count-metadata} (`pr#15436 <https://github.com/ceph/ceph/pull/15436>`_, Sage Weil)
* mon: mon/PGMap: show %used in formatted output (`issue#20123 <http://tracker.ceph.com/issues/20123>`_, `pr#15387 <https://github.com/ceph/ceph/pull/15387>`_, Joao Eduardo Luis)
* mon: Removed unnecessary function declaration in MDSMonitor.h (`pr#15374 <https://github.com/ceph/ceph/pull/15374>`_, yonghengdexin735)
* mon: replace osds with `osd destroy` and `osd new` (`pr#14074 <https://github.com/ceph/ceph/pull/14074>`_, Joao Eduardo Luis, Sage Weil)
* mon: revise "ceph status" output (`pr#15396 <https://github.com/ceph/ceph/pull/15396>`_, John Spray)
* mon: show io status quickly if no update in a long period (`pr#14176 <https://github.com/ceph/ceph/pull/14176>`_, Mingxin Liu)
* mon: track features from connect clients, and use it to gate set-require-min-compat-client (`pr#15371 <https://github.com/ceph/ceph/pull/15371>`_, Sage Weil)
* mon: trim the creating_pgs after updating it with pgmap (`issue#20067 <http://tracker.ceph.com/issues/20067>`_, `pr#15318 <https://github.com/ceph/ceph/pull/15318>`_, Kefu Chai)
* msg: do not enable client-side binding by default (`issue#20049 <http://tracker.ceph.com/issues/20049>`_, `pr#15392 <https://github.com/ceph/ceph/pull/15392>`_, Jason Dillaman)
* msg: don't set msgr addr when disabing client bind (`pr#15243 <https://github.com/ceph/ceph/pull/15243>`_, Haomai Wang)
* msgr: msg/async: Lower down the AsyncMessenger's standby warning from debug (`pr#15242 <https://github.com/ceph/ceph/pull/15242>`_, Pan Liu)
* msgr: msg/async: remove false alert "assert" (`pr#15288 <https://github.com/ceph/ceph/pull/15288>`_, Haomai Wang)
* osd: don't leak pgrefs or reservations in SnapTrimmer (`issue#19931 <http://tracker.ceph.com/issues/19931>`_, `pr#15214 <https://github.com/ceph/ceph/pull/15214>`_, Greg Farnum)
* osd: fix argument-dependent lookup of swap() (`pr#15124 <https://github.com/ceph/ceph/pull/15124>`_, Casey Bodley)
* osd: fix past_intervals base case by adding epoch_pool_created to pg_history_t (`issue#19877 <http://tracker.ceph.com/issues/19877>`_, `pr#14989 <https://github.com/ceph/ceph/pull/14989>`_, Sage Weil)
* osd: hdd vs ssd defaults for osd op thread pool (`pr#15422 <https://github.com/ceph/ceph/pull/15422>`_, Sage Weil)
* osd: Implement asynchronous recovery sleep (`pr#15212 <https://github.com/ceph/ceph/pull/15212>`_, Neha Ojha)
* osd: Move scrub sleep timer to osdservice (`issue#19986 <http://tracker.ceph.com/issues/19986>`_, `pr#15217 <https://github.com/ceph/ceph/pull/15217>`_, Brad Hubbard)
* osd: Object level shard errors are tracked and used if no auth available (`issue#20089 <http://tracker.ceph.com/issues/20089>`_, `pr#15397 <https://github.com/ceph/ceph/pull/15397>`_, David Zafman)
* osd: osd/OSDMap.cc: check if osd is out in subtree_type_is_down (`issue#19989 <http://tracker.ceph.com/issues/19989>`_, `pr#15250 <https://github.com/ceph/ceph/pull/15250>`_, Neha Ojha)
* osd: 'osd tree in|out|up|down' to filter tree results (`pr#15294 <https://github.com/ceph/ceph/pull/15294>`_, Sage Weil)
* osd: reduce buffer pinning from EC entries (`pr#15120 <https://github.com/ceph/ceph/pull/15120>`_, Sage Weil)
* osd: reduce map cache size (`pr#15292 <https://github.com/ceph/ceph/pull/15292>`_, Sage Weil)
* osd: reduce rados_max_object_size from 100 GB -> 128 MB (`pr#15520 <https://github.com/ceph/ceph/pull/15520>`_, Sage Weil)
* osd: rename osd -> osd_pglog; include pglog-related bufferlists (`pr#15531 <https://github.com/ceph/ceph/pull/15531>`_, Sage Weil)
* osd: Return early on shutdown (`issue#19900 <http://tracker.ceph.com/issues/19900>`_, `pr#15345 <https://github.com/ceph/ceph/pull/15345>`_, Brad Hubbard)
* osd: take PGRef for recovery sleep wakeup event (`issue#20226 <http://tracker.ceph.com/issues/20226>`_, `pr#15582 <https://github.com/ceph/ceph/pull/15582>`_, Sage Weil)
* osd: when osd in not in failure_pending, we don't need to get osd inst from osdmap. (`pr#15558 <https://github.com/ceph/ceph/pull/15558>`_, linbing)
* osd: When scrub finds an attr error mark shard inconsistent (`issue#20089 <http://tracker.ceph.com/issues/20089>`_, `pr#15368 <https://github.com/ceph/ceph/pull/15368>`_, David Zafman)
* performance: common/config_opts.h: Lower HDD throttle cost. (`pr#15485 <https://github.com/ceph/ceph/pull/15485>`_, Mark Nelson)
* performance: denc: add need_contiguous to denc_traits (`pr#15224 <https://github.com/ceph/ceph/pull/15224>`_, Kefu Chai)
* pybind: pybind/ceph_argparse: fix empty string check (`issue#20135 <http://tracker.ceph.com/issues/20135>`_, `pr#15500 <https://github.com/ceph/ceph/pull/15500>`_, Sage Weil)
* pybind: pybind/ceph_daemon.py: fix Termsize.update (`pr#15253 <https://github.com/ceph/ceph/pull/15253>`_, Kefu Chai)
* pybind: pybind/rados: avoid call free() on invalid pointer (`pr#15159 <https://github.com/ceph/ceph/pull/15159>`_, Mingxin Liu)
* pybind,rbd: pybind/rbd: OSError should be picklable (`issue#20223 <http://tracker.ceph.com/issues/20223>`_, `pr#15574 <https://github.com/ceph/ceph/pull/15574>`_, Jason Dillaman)
* pybind: support mon target in pybind (`pr#15409 <https://github.com/ceph/ceph/pull/15409>`_, liuchang0812)
* rbd-mirror: coordinate image syncs with leader (`issue#18789 <http://tracker.ceph.com/issues/18789>`_, `pr#14745 <https://github.com/ceph/ceph/pull/14745>`_, Mykola Golub)
* rbd-mirror: lock loss during sync should wait for in-flight copies (`pr#15532 <https://github.com/ceph/ceph/pull/15532>`_, Jason Dillaman)
* rbd-mirror: permit release of local image exclusive lock after force promotion (`issue#18963 <http://tracker.ceph.com/issues/18963>`_, `pr#15140 <https://github.com/ceph/ceph/pull/15140>`_, Jason Dillaman)
* rbd: properly decode features when using image name optional (`issue#20185 <http://tracker.ceph.com/issues/20185>`_, `pr#15492 <https://github.com/ceph/ceph/pull/15492>`_, Jason Dillaman)
* rbd: pybind/rbd: fix crash if more than 1024 images in trash bin (`pr#15134 <https://github.com/ceph/ceph/pull/15134>`_, runsisi)
* rbd: rbd/bench: fix write gaps when doing sequential writes with io-threads > 1 (`pr#15206 <https://github.com/ceph/ceph/pull/15206>`_, Igor Fedotov)
* rbd: removed hardcoded default pool (`pr#15518 <https://github.com/ceph/ceph/pull/15518>`_, Jason Dillaman)
* rbd,tests: qa: krbd discard/zeroout tests (`pr#15388 <https://github.com/ceph/ceph/pull/15388>`_, Ilya Dryomov)
* rbd,tests: qa/suites/krbd: unmap subsuite needs straw buckets (`pr#15290 <https://github.com/ceph/ceph/pull/15290>`_, Ilya Dryomov)
* rbd,tests: qa: update krbd_data_pool.sh to match the new rados ls behavior (`pr#15594 <https://github.com/ceph/ceph/pull/15594>`_, Ilya Dryomov)
* rbd,tests: test/librbd: unit tests cleanup (`pr#15113 <https://github.com/ceph/ceph/pull/15113>`_, Mykola Golub)
* rdma: msg/async/rdma: Add DSCP support (`pr#15484 <https://github.com/ceph/ceph/pull/15484>`_, Sarit Zubakov)
* rdma: msg/async: Revert RDMA-CM (`pr#15262 <https://github.com/ceph/ceph/pull/15262>`_, Amir Vadai)
* rgw: Adding code to create tenanted user for s3 bucket policy tests. (`pr#15028 <https://github.com/ceph/ceph/pull/15028>`_, Pritha Srivastava)
* rgw: add "rgw_verify_ssl" config (`pr#15301 <https://github.com/ceph/ceph/pull/15301>`_, Shasha Lu)
* rgw: add the Vim's modeline into rgw_orphan.cc. (`pr#15431 <https://github.com/ceph/ceph/pull/15431>`_, Radoslaw Zarzynski)
* rgw: bucket index check in radosgw-admin removes valid index. (`issue#18470 <http://tracker.ceph.com/issues/18470>`_, `pr#12851 <https://github.com/ceph/ceph/pull/12851>`_, Zhang Shaowen)
* rgw: datalog trim and mdlog trim handles the result returned by osd incorrectly. (`issue#20190 <http://tracker.ceph.com/issues/20190>`_, `pr#15507 <https://github.com/ceph/ceph/pull/15507>`_, Zhang Shaowen)
* rgw: display more info when using radosgw-admin bucket stats (`pr#15256 <https://github.com/ceph/ceph/pull/15256>`_, fang.yuxiang)
* rgw: drop asio/{yield,coroutine}.hpp replacements (`pr#15413 <https://github.com/ceph/ceph/pull/15413>`_, Kefu Chai)
* rgw: drop using std ns in header files and other cleanups (`pr#15137 <https://github.com/ceph/ceph/pull/15137>`_, Abhishek Lekshmanan)
* rgw: dynamic resharding (`pr#15493 <https://github.com/ceph/ceph/pull/15493>`_, Yehuda Sadeh, Orit Wasserman)
* rgw: fix 'gc list --include-all' command infinite loop the first items (`issue#19978 <http://tracker.ceph.com/issues/19978>`_, `pr#12774 <https://github.com/ceph/ceph/pull/12774>`_, Shasha Lu, fang yuxiang)
* rgw: fix lc list failure when shards not be all created (`issue#19898 <http://tracker.ceph.com/issues/19898>`_, `pr#15025 <https://github.com/ceph/ceph/pull/15025>`_, Jiaying Ren)
* rgw: fix radosgw-admin retcode (`pr#15257 <https://github.com/ceph/ceph/pull/15257>`_, Shasha Lu)
* rgw: fix test_multi.py default config file path (`pr#15306 <https://github.com/ceph/ceph/pull/15306>`_, Jiaying Ren)
* rgw: fix X-Object-Meta-Static-Large-Object in SLO download (`issue#19951 <http://tracker.ceph.com/issues/19951>`_, `pr#15045 <https://github.com/ceph/ceph/pull/15045>`_, Shasha Lu)
* rgw: metadata search part 2 (`pr#14351 <https://github.com/ceph/ceph/pull/14351>`_, Yehuda Sadeh)
* rgw: migrate atomic_t to std::atomic<> (`pr#15001 <https://github.com/ceph/ceph/pull/15001>`_, Jesse Williamson)
* rgw: optimize data sync. Add zones_trace in log to avoid needless sync. (`issue#19219 <http://tracker.ceph.com/issues/19219>`_, `pr#13851 <https://github.com/ceph/ceph/pull/13851>`_, Zhang Shaowen)
* rgw: optimize generating torrent file. Object data won't stay in memory now. (`pr#15153 <https://github.com/ceph/ceph/pull/15153>`_, Zhang Shaowen)
* rgw: pass authentication domain to civetweb (`issue#17657 <http://tracker.ceph.com/issues/17657>`_, `pr#12861 <https://github.com/ceph/ceph/pull/12861>`_, Abhishek Lekshmanan)
* rgw: polymorphic error codes (`pr#10690 <https://github.com/ceph/ceph/pull/10690>`_, Pritha Srivastava, Marcus Watts)
* rgw: remove fastcgi from default rgw frontends (`pr#15098 <https://github.com/ceph/ceph/pull/15098>`_, Casey Bodley)
* rgw: rename s3_code to err_code for swift (`pr#12300 <https://github.com/ceph/ceph/pull/12300>`_, Guo Zhandong)
* rgw: return the version id in get object and object metadata request. (`issue#19370 <http://tracker.ceph.com/issues/19370>`_, `pr#14117 <https://github.com/ceph/ceph/pull/14117>`_, Zhang Shaowen)
* rgw: rgw-admin: fix bucket limit check argparse, div(0) (`pr#15316 <https://github.com/ceph/ceph/pull/15316>`_, Matt Benjamin)
* rgw: rgw_common: use string::npos for the results of str.find (`pr#14341 <https://github.com/ceph/ceph/pull/14341>`_, Abhishek Lekshmanan)
* rgw: rgw_file: add lock protection for readdir against gc (`issue#20121 <http://tracker.ceph.com/issues/20121>`_, `pr#15329 <https://github.com/ceph/ceph/pull/15329>`_, Gui Hecheng)
* rgw: rgw_file cleanup names (`pr#15568 <https://github.com/ceph/ceph/pull/15568>`_, Gui Hecheng)
* rgw: rgw_file: fix flags set on unsuccessful unlink (`pr#15222 <https://github.com/ceph/ceph/pull/15222>`_, Gui Hecheng)
* rgw: rgw_file: release rgw_fh lock and ref on ENOTEMPTY (`issue#20061 <http://tracker.ceph.com/issues/20061>`_, `pr#15246 <https://github.com/ceph/ceph/pull/15246>`_, Matt Benjamin)
* rgw: rgw_file: removed extra rele() on fs in rgw_umount() (`pr#15152 <https://github.com/ceph/ceph/pull/15152>`_, Gui Hecheng)
* rgw: rgw_file: remove hidden uxattr objects from buckets on delete (`issue#20045 <http://tracker.ceph.com/issues/20045>`_, `pr#15210 <https://github.com/ceph/ceph/pull/15210>`_, Matt Benjamin)
* rgw: rgw_file: remove post-unlink lookup check (`issue#20047 <http://tracker.ceph.com/issues/20047>`_, `pr#15216 <https://github.com/ceph/ceph/pull/15216>`_, Matt Benjamin)
* rgw: rgw_file: replace raw fs->fh_lru.unref with predefined fs->unref (`pr#15541 <https://github.com/ceph/ceph/pull/15541>`_, Gui Hecheng)
* rgw: rgw_file: store bucket uxattrs on the bucket (`issue#20082 <http://tracker.ceph.com/issues/20082>`_, `pr#15293 <https://github.com/ceph/ceph/pull/15293>`_, Matt Benjamin)
* rgw: rgw_file: v3: fix write-timer action (`issue#19932 <http://tracker.ceph.com/issues/19932>`_, `pr#15097 <https://github.com/ceph/ceph/pull/15097>`_, Matt Benjamin)
* rgw: rgw_rados: create sync module instances only if run_sync_thread is set (`issue#19830 <http://tracker.ceph.com/issues/19830>`_, `pr#14994 <https://github.com/ceph/ceph/pull/14994>`_, Abhishek Lekshmanan)
* rgw: rgw/rgw_swift_auth.cc: using string::back() instead as the C++11 recommend (`pr#14827 <https://github.com/ceph/ceph/pull/14827>`_, liuyuhong)
* rgw: segment fault when shard id out of range (`issue#19732 <http://tracker.ceph.com/issues/19732>`_, `pr#14389 <https://github.com/ceph/ceph/pull/14389>`_, redickwang)
* rgw: set object accounted size correctly (`issue#20071 <http://tracker.ceph.com/issues/20071>`_, `pr#14950 <https://github.com/ceph/ceph/pull/14950>`_, fang yuxiang)
* rgw: set placement rule properly (`pr#15221 <https://github.com/ceph/ceph/pull/15221>`_, fang.yuxiang)
* rgw: support certain archaic and antiquated distributions (`pr#15498 <https://github.com/ceph/ceph/pull/15498>`_, Adam C. Emerson)
* rgw,tests: qa/rgw: add multisite suite to configure and run multisite tests (`pr#14688 <https://github.com/ceph/ceph/pull/14688>`_, Casey Bodley)
* rgw,tests: qa/rgw: remove apache/fastcgi and radosgw-agent tests (`pr#15184 <https://github.com/ceph/ceph/pull/15184>`_, Casey Bodley)
* rgw: Turn off fcgi as a frontend (`issue#16784 <http://tracker.ceph.com/issues/16784>`_, `pr#15070 <https://github.com/ceph/ceph/pull/15070>`_, Thomas Serlin)
* rgw: use get_data_extra_pool() when get extra pool (`issue#20064 <http://tracker.ceph.com/issues/20064>`_, `pr#15219 <https://github.com/ceph/ceph/pull/15219>`_, fang yuxiang)
* rgw: use pre-defined calls to replace raw flag operation (`pr#15107 <https://github.com/ceph/ceph/pull/15107>`_, Gui Hecheng)
* tests: Add integration tests for admin socket output (`pr#15223 <https://github.com/ceph/ceph/pull/15223>`_, Brad Hubbard)
* tests: ceph-disk: add setting for external py-modules for tox-testing (`pr#15433 <https://github.com/ceph/ceph/pull/15433>`_, Willem Jan Withagen)
* tests: Check make_writeable() return value (`pr#15266 <https://github.com/ceph/ceph/pull/15266>`_, zhanglei)
* tests: config_opts: drop unused opts (`pr#15031 <https://github.com/ceph/ceph/pull/15031>`_, Kefu Chai)
* tests: qa: add task for dnsmasq configuration (`pr#15071 <https://github.com/ceph/ceph/pull/15071>`_, Casey Bodley)
* tests: qa: split test_tiering into smaller pieces (`pr#15146 <https://github.com/ceph/ceph/pull/15146>`_, Kefu Chai)
* tests: qa/suites/rados: fix ec thrashing (`pr#15087 <https://github.com/ceph/ceph/pull/15087>`_, Sage Weil)
* tests: qa/suites/rados/singleton-nomsgr: fix syntax (`pr#15276 <https://github.com/ceph/ceph/pull/15276>`_, Sage Weil)
* tests: qa/suites/rados/thrash: make sure osds have map before legacy scrub (`pr#15117 <https://github.com/ceph/ceph/pull/15117>`_, Sage Weil)
* tests: qa/suites/rados/upgrade: restart mds (`pr#15517 <https://github.com/ceph/ceph/pull/15517>`_, Sage Weil)
* tests: qa/tasks/ceph_manager: 'ceph $service tell ...' is obsolete (`pr#15252 <https://github.com/ceph/ceph/pull/15252>`_, Sage Weil)
* tests: qa/tasks/rebuild_mondb: grant "mgr:allow *" to client.admin (`issue#19439 <http://tracker.ceph.com/issues/19439>`_, `pr#14284 <https://github.com/ceph/ceph/pull/14284>`_, Kefu Chai)
* tests: qa/tasks/repair_test: unset flags we set (`pr#15296 <https://github.com/ceph/ceph/pull/15296>`_, Sage Weil)
* tests: qa/workunits/ceph-helpers.sh: use syntax understood by jq 1.3 (`pr#15530 <https://github.com/ceph/ceph/pull/15530>`_, Kefu Chai)
* tests: Rename FileJournal object to distinguish (`pr#15279 <https://github.com/ceph/ceph/pull/15279>`_, Jos Collin)
* tests: test/crush: silence warnings from -Walloc-size-larger-than= and -Wstringop-overflow= (`pr#15173 <https://github.com/ceph/ceph/pull/15173>`_, Jos Collin)
* tests: test: migrate atomic_t to std::atomic (`pr#14655 <https://github.com/ceph/ceph/pull/14655>`_, Jesse Williamson)
* tests: test/msgr: silence warnings from -Wsign-compare (`pr#15356 <https://github.com/ceph/ceph/pull/15356>`_, Jos Collin)
* tests: test/msgr: silence warnings from -Wsign-compare (`pr#15570 <https://github.com/ceph/ceph/pull/15570>`_, Jos Collin)
* tests: test/objectstore: Check apply_transaction() return values (`pr#15171 <https://github.com/ceph/ceph/pull/15171>`_, zhanglei)
* tests: test/old: Removed commented code (`pr#15366 <https://github.com/ceph/ceph/pull/15366>`_, Jos Collin)
* tests: test/osdc: fix comparison error and silence warning from -Wunused-value (`pr#15353 <https://github.com/ceph/ceph/pull/15353>`_, Willem Jan Withagen)
* tests: test: osd/TestOSDMap.cc: fix Clang complain about promotion (`pr#15525 <https://github.com/ceph/ceph/pull/15525>`_, Willem Jan Withagen)
* tests: test: test_denc.cc: silence warning from -Wsign-compare (`pr#15355 <https://github.com/ceph/ceph/pull/15355>`_, Jos Collin)
* tests: test: Test fix for SnapSet change (`pr#15161 <https://github.com/ceph/ceph/pull/15161>`_, David Zafman)
* tests: test/unittest_bluefs: check whether mounted success (`pr#14988 <https://github.com/ceph/ceph/pull/14988>`_, shiqi)
* tools: ceph.in: adjust usage width according to user's tty (`pr#15190 <https://github.com/ceph/ceph/pull/15190>`_, Kefu Chai)
* tools: ceph.in: assert(state==connected) before help_for_target() (`pr#15156 <https://github.com/ceph/ceph/pull/15156>`_, Kefu Chai)
* tools: ceph.in: drop the compatiiblity to handle non json commands (`pr#15508 <https://github.com/ceph/ceph/pull/15508>`_, Kefu Chai)
* tools: ceph.in: print return code when json_command failed (`pr#15378 <https://github.com/ceph/ceph/pull/15378>`_, liuchang0812)
* tools: tools/ceph_kvstore_tool: add "bluestore-kv" to usage (`pr#15326 <https://github.com/ceph/ceph/pull/15326>`_, xie xingguo)
* tools: tools/crushtool: replicated-rule API support (`pr#15011 <https://github.com/ceph/ceph/pull/15011>`_, xie xingguo)
* tools: vstart: "debug_ms=1" for mgr by default (`pr#15127 <https://github.com/ceph/ceph/pull/15127>`_, Kefu Chai)
* tools: vstart: print "start osd.$id" instead of "start osd$id" (`pr#15427 <https://github.com/ceph/ceph/pull/15427>`_, Kefu Chai)
v12.0.3 Luminous (dev)
======================
This is the fourth development checkpoint release of Luminous, the next long
term stable release. This release introduces several improvements in bluestore,
monitor, rbd & rgw.
Major changes from v12.0.2
--------------------------
Notable Changes
---------------
* bluestore,core: osd/OSDMap: should update input param if osd dne (`pr#14863 <https://github.com/ceph/ceph/pull/14863>`_, Kefu Chai)
* bluestore: include/intarith: templatize ctz/clz/cbits helpers (`pr#14862 <https://github.com/ceph/ceph/pull/14862>`_, Kefu Chai)
* bluestore: os/bluestore: align reclaim size to bluefs_alloc_size (`pr#14744 <https://github.com/ceph/ceph/pull/14744>`_, Haomai Wang)
* bluestore: os/bluestore: assert blob map returns success (`pr#14473 <https://github.com/ceph/ceph/pull/14473>`_, shiqi)
* bluestore: os/bluestore: fix deferred write race (`issue#19880 <http://tracker.ceph.com/issues/19880>`_, `pr#15004 <https://github.com/ceph/ceph/pull/15004>`_, Sage Weil)
* bluestore: os/bluestore: fix typo(s/trasnaction/transaction/) (`pr#14890 <https://github.com/ceph/ceph/pull/14890>`_, xie xingguo)
* bluestore: os/bluestore: fix use after free race with aio_wait (`pr#14956 <https://github.com/ceph/ceph/pull/14956>`_, Sage Weil)
* bluestore: os/bluestore: pre-calculate number of ghost buffers to evict (`pr#15029 <https://github.com/ceph/ceph/pull/15029>`_, xie xingguo)
* bluestore: os/bluestore: Record l_bluestore_state_kv_queued_lat for sync_submit_… (`pr#14448 <https://github.com/ceph/ceph/pull/14448>`_, Jianpeng Ma)
* bluestore: os/bluestore: Remove ExtentFreeListManager. (`pr#14772 <https://github.com/ceph/ceph/pull/14772>`_, Jianpeng Ma)
* bluestore: os/bluestore: remove unused condition variable (`pr#14973 <https://github.com/ceph/ceph/pull/14973>`_, Igor Fedotov)
* bluestore: os/bluestore: rename/fix throttle options (`pr#14717 <https://github.com/ceph/ceph/pull/14717>`_, Sage Weil)
* bluestore: os/bluestore: roundoff bluefs allocs to bluefs_alloc_size (`pr#14876 <https://github.com/ceph/ceph/pull/14876>`_, Ramesh Chander)
* bluestore: os/bluestore: use correct bound encode size for unused (`pr#14731 <https://github.com/ceph/ceph/pull/14731>`_, Haomai Wang)
* bluestore,performance: common/config_opts.h: compaction readahead for bluestore/rocksdb (`pr#14932 <https://github.com/ceph/ceph/pull/14932>`_, Mark Nelson)
* bluestore,performance: os/bluestore/BlueFS: add bluefs_sync_write option (`pr#14510 <https://github.com/ceph/ceph/pull/14510>`_, Sage Weil)
* bluestore,performance: os/bluestore: do not balance bluefs on every kv_sync_thread iteration (`pr#14557 <https://github.com/ceph/ceph/pull/14557>`_, Sage Weil)
* bluestore,performance: os/bluestore: eliminate some excessive stuff (`pr#14675 <https://github.com/ceph/ceph/pull/14675>`_, Igor Fedotov)
* bluestore,performance: os/bluestore: get rid off excessive lock at BitMapAllocator (`pr#14749 <https://github.com/ceph/ceph/pull/14749>`_, Igor Fedotov)
* bluestore,performance: os/blueStore: In osd_tp_thread, call _txc_finalize_kv. (`pr#14709 <https://github.com/ceph/ceph/pull/14709>`_, Jianpeng Ma)
* bluestore,performance: os/bluestore: make bluestore_max_blob_size parameter hdd/ssd case dependant (`pr#14434 <https://github.com/ceph/ceph/pull/14434>`_, Igor Fedotov)
* bluestore,performance: os/bluestore: rewrite deferred write handling (`issue#16644 <http://tracker.ceph.com/issues/16644>`_, `pr#14491 <https://github.com/ceph/ceph/pull/14491>`_, Sage Weil)
* bluestore,performance: os/bluestore: use denc for varint encoding (`pr#14911 <https://github.com/ceph/ceph/pull/14911>`_, Piotr Dałek)
* bluestore,performance: os/fs/aio: use small_vector for aio_t; clean up header location (`pr#14853 <https://github.com/ceph/ceph/pull/14853>`_, Sage Weil)
* bluestore,tests: unittest_alloc: add test_alloc_big (`issue#16662 <http://tracker.ceph.com/issues/16662>`_, `pr#14844 <https://github.com/ceph/ceph/pull/14844>`_, Sage Weil)
* bluestore,tools: ceph-kvstore-tool: allow 'bluestore-kv' as kvdb type; add escaping, compaction (`pr#14718 <https://github.com/ceph/ceph/pull/14718>`_, Sage Weil)
* build/ops: alpine: add alpine linux dev support (`pr#9853 <https://github.com/ceph/ceph/pull/9853>`_, John Coyle)
* build/ops: arch: use __get_cpuid instead of do_cpuid (`issue#7869 <http://tracker.ceph.com/issues/7869>`_, `pr#14857 <https://github.com/ceph/ceph/pull/14857>`_, Jos Collin)
* build/ops: CMakeLists.txt: don't do crypto/isa-l if not Intel (`pr#14721 <https://github.com/ceph/ceph/pull/14721>`_, Dan Mick)
* build/ops: compressor/zlib: fix plugin for non-Intel arches (`pr#14947 <https://github.com/ceph/ceph/pull/14947>`_, Dan Mick)
* build/ops: debian/rpm: move radosgw-admin to ceph-common (`issue#19577 <http://tracker.ceph.com/issues/19577>`_, `pr#14940 <https://github.com/ceph/ceph/pull/14940>`_, Ali Maredia)
* build/ops: dmclock: error: ‘function’ in namespace ‘std’ does not name a template type (`pr#14909 <https://github.com/ceph/ceph/pull/14909>`_, Jos Collin)
* build/ops: dmclock: initial commit of dmclock QoS library (`pr#14330 <https://github.com/ceph/ceph/pull/14330>`_, J. Eric Ivancich)
* build/ops: init-ceph: add ceph libraries path to environment (`pr#14693 <https://github.com/ceph/ceph/pull/14693>`_, Mohamad Gebai)
* build/ops: init-ceph: should have a space before "]" (`pr#14796 <https://github.com/ceph/ceph/pull/14796>`_, Kefu Chai)
* build/ops: merge v12.0.2 release tag (`pr#15091 <https://github.com/ceph/ceph/pull/15091>`_, Jenkins Build Slave User)
* build/ops,mgr: debian/ceph-base.dirs: create bootstrap-mgr dirs (`pr#14838 <https://github.com/ceph/ceph/pull/14838>`_, Sage Weil)
* build/ops,mon: mon/ConfigKeyService: add 'config-key dump' to show keys and vals (`pr#14858 <https://github.com/ceph/ceph/pull/14858>`_, Dan Mick)
* build/ops,performance,rbd: byteorder: use gcc intrinsics for byteswap (`pr#15012 <https://github.com/ceph/ceph/pull/15012>`_, Kefu Chai)
* build/ops: rocksdb: sync with upstream (`pr#14818 <https://github.com/ceph/ceph/pull/14818>`_, Nathan Cutler, Kefu Chai)
* build/ops: rpm: fix python-Sphinx package name for SUSE (`pr#15015 <https://github.com/ceph/ceph/pull/15015>`_, Nathan Cutler, Jan Matejek)
* build/ops: rpm: gperftools-devel >= 2.4 (`issue#13522 <http://tracker.ceph.com/issues/13522>`_, `pr#14870 <https://github.com/ceph/ceph/pull/14870>`_, Nathan Cutler)
* build/ops: rpm: package crypto on x86_64 only (`pr#14779 <https://github.com/ceph/ceph/pull/14779>`_, Nathan Cutler)
* build/ops: debian: package crypto plugin only on amd64 (`pr#14820 <https://github.com/ceph/ceph/pull/14820>`_, Kefu Chai)
* build/ops: src/init-ceph.in: allow one((re)?start|stop) as commands (`pr#14560 <https://github.com/ceph/ceph/pull/14560>`_, Willem Jan Withagen)
* build/ops: yasm-wrapper: strip -E (stops ccache trashing source files) (`pr#14633 <https://github.com/ceph/ceph/pull/14633>`_, Tim Serong)
* cephfs: ceph-fuse: use user space permission check by default (`issue#19820 <http://tracker.ceph.com/issues/19820>`_, `pr#14907 <https://github.com/ceph/ceph/pull/14907>`_, "Yan, Zheng")
* cephfs: client: client_quota no longer optional (`pr#14978 <https://github.com/ceph/ceph/pull/14978>`_, Dan van der Ster)
* cephfs: client: fix UserPerm::gid_in_group() (`issue#19903 <http://tracker.ceph.com/issues/19903>`_, `pr#15039 <https://github.com/ceph/ceph/pull/15039>`_, "Yan, Zheng")
* cephfs: client: getattr before returning quota/layout xattrs (`issue#17939 <http://tracker.ceph.com/issues/17939>`_, `pr#14018 <https://github.com/ceph/ceph/pull/14018>`_, John Spray)
* cephfs: fs/ceph-fuse: normalize file open flags on the wire (`pr#14822 <https://github.com/ceph/ceph/pull/14822>`_, Jan Fajerski)
* cephfs: mds/Server.cc: Don't evict a slow client if... (`issue#17855 <http://tracker.ceph.com/issues/17855>`_, `pr#12935 <https://github.com/ceph/ceph/pull/12935>`_, Michal Jarzabek)
* cephfs: osdc/Filer: truncate large file party by party (`issue#19755 <http://tracker.ceph.com/issues/19755>`_, `pr#14769 <https://github.com/ceph/ceph/pull/14769>`_, "Yan, Zheng")
* cephfs: osdc: remove journaler_allow_split_entries option (`issue#19691 <http://tracker.ceph.com/issues/19691>`_, `pr#14636 <https://github.com/ceph/ceph/pull/14636>`_, John Spray)
* cephfs,performance: client: make seeky readdir more efficiency (`issue#19306 <http://tracker.ceph.com/issues/19306>`_, `pr#14317 <https://github.com/ceph/ceph/pull/14317>`_, "Yan, Zheng")
* cephfs: qa/cephfs: Fix for test_data_scan (`issue#19893 <http://tracker.ceph.com/issues/19893>`_, `pr#15094 <https://github.com/ceph/ceph/pull/15094>`_, Douglas Fuller)
* cephfs: qa/suites/fs: reserve more space for mds in full tests (`issue#19891 <http://tracker.ceph.com/issues/19891>`_, `pr#15026 <https://github.com/ceph/ceph/pull/15026>`_, "Yan, Zheng")
* cephfs,tests: qa: silence spurious insufficient standby health warnings (`pr#15035 <https://github.com/ceph/ceph/pull/15035>`_, Patrick Donnelly)
* cephfs,tests: qa: Tidy up fs/ suite (`pr#14575 <https://github.com/ceph/ceph/pull/14575>`_, John Spray)
* cleanup: dmclock: include missing <functional> header. (`pr#14923 <https://github.com/ceph/ceph/pull/14923>`_, Jos Collin)
* cleanup: kill clang warnings (`pr#14549 <https://github.com/ceph/ceph/pull/14549>`_, Kefu Chai)
* cleanup: test: c_write_operations.cc: silence warning from -Wsign-compare (`pr#14889 <https://github.com/ceph/ceph/pull/14889>`_, Jos Collin)
* cmake: add "container" to required boost components (`pr#14850 <https://github.com/ceph/ceph/pull/14850>`_, Kefu Chai)
* cmake: align cmake names of library packages (`issue#19853 <http://tracker.ceph.com/issues/19853>`_, `pr#14951 <https://github.com/ceph/ceph/pull/14951>`_, Nathan Cutler)
* cmake: Allow tests to build without NSS (`pr#13315 <https://github.com/ceph/ceph/pull/13315>`_, Daniel Gryniewicz)
* cmake: do not compile crush twice (`pr#14725 <https://github.com/ceph/ceph/pull/14725>`_, Kefu Chai)
* cmake: enable cross-compilation of boost (`issue#18938 <http://tracker.ceph.com/issues/18938>`_, `pr#14881 <https://github.com/ceph/ceph/pull/14881>`_, Kefu Chai)
* cmake: fix the linked lib reference of unittest_rgw_crypto (`pr#14869 <https://github.com/ceph/ceph/pull/14869>`_, Willem Jan Withagen)
* cmake: kill duplicated cmake commands (`pr#14948 <https://github.com/ceph/ceph/pull/14948>`_, liuchang0812)
* cmake: pass -d0 to b2 if not CMAKE_VERBOSE_MAKEFILE (`pr#14651 <https://github.com/ceph/ceph/pull/14651>`_, Kefu Chai)
* cmake: should not compile crc32c_ppc.c on intel arch. (`pr#14423 <https://github.com/ceph/ceph/pull/14423>`_, Kefu Chai)
* common: Better handling for missing/inaccessible ceph.conf files (`issue#19658 <http://tracker.ceph.com/issues/19658>`_, `pr#14757 <https://github.com/ceph/ceph/pull/14757>`_, Dan Mick)
* common: cls/log/cls_log.cc: reduce logging noise (`issue#19835 <http://tracker.ceph.com/issues/19835>`_, `pr#14879 <https://github.com/ceph/ceph/pull/14879>`_, Willem Jan Withagen)
* common: common/admin_socket: add config for admin socket permission bits (`pr#11684 <https://github.com/ceph/ceph/pull/11684>`_, runsisi)
* common: common/ceph_context.cc: Use CEPH_DEV to reduce logfile noise (`pr#10384 <https://github.com/ceph/ceph/pull/10384>`_, Willem Jan Withagen)
* common: common/config: Add /usr/local/etc/ceph to default paths (`pr#14797 <https://github.com/ceph/ceph/pull/14797>`_, Willem Jan Withagen)
* common: common/config_opts: Set the HDD throttle cost to 1.5M (`pr#14808 <https://github.com/ceph/ceph/pull/14808>`_, Mark Nelson)
* common: common/Finisher: fix uninitialized variable warning (`pr#14958 <https://github.com/ceph/ceph/pull/14958>`_, Piotr Dałek)
* common: common/interval_set: return int64_t for size() (`pr#12898 <https://github.com/ceph/ceph/pull/12898>`_, Xinze Chi)
* common: crypto: cleanup NSPR in main thread (`pr#14801 <https://github.com/ceph/ceph/pull/14801>`_, Kefu Chai)
* common: fix building against libcryptopp (`pr#14949 <https://github.com/ceph/ceph/pull/14949>`_, Shengjing Zhu)
* common: Fix unused variable references warnings (`pr#14790 <https://github.com/ceph/ceph/pull/14790>`_, Willem Jan Withagen)
* common: msg/async: return right away in NetHandler::set_priority() if not supported (`pr#14795 <https://github.com/ceph/ceph/pull/14795>`_, Kefu Chai)
* common: osdc/Objecter: fix pool dne corner case (`issue#19552 <http://tracker.ceph.com/issues/19552>`_, `pr#14901 <https://github.com/ceph/ceph/pull/14901>`_, Sage Weil)
* common: osdc/Objecter: handle command target that goes down (`issue#19452 <http://tracker.ceph.com/issues/19452>`_, `pr#14302 <https://github.com/ceph/ceph/pull/14302>`_, Sage Weil)
* common: osd/OSDMap: fix feature commit comment (`pr#15056 <https://github.com/ceph/ceph/pull/15056>`_, Sage Weil)
* common,performance: common/Finisher: batch handle perfcounter && only send signal when waiter existed. (`pr#14363 <https://github.com/ceph/ceph/pull/14363>`_, Jianpeng Ma)
* common: Remove redundant includes (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15003 <https://github.com/ceph/ceph/pull/15003>`_, Brad Hubbard)
* common: Remove redundant includes (`issue#19883 <http://tracker.ceph.com/issues/19883>`_, `pr#15019 <https://github.com/ceph/ceph/pull/15019>`_, Brad Hubbard)
* common: src/common: change last_work_queue to next_work_queue. (`pr#14738 <https://github.com/ceph/ceph/pull/14738>`_, Pan Liu)
* core: ceph-disk: ceph-disk on FreeBSD should not use mpath-code (`pr#14837 <https://github.com/ceph/ceph/pull/14837>`_, Willem Jan Withagen)
* core: ceph-disk: implement prepare --no-locking (`pr#14728 <https://github.com/ceph/ceph/pull/14728>`_, Dan van der Ster, Loic Dachary)
* core: ceph-disk: separate ceph-osd --check-needs-* logs (`issue#19888 <http://tracker.ceph.com/issues/19888>`_, `pr#15016 <https://github.com/ceph/ceph/pull/15016>`_, Loic Dachary)
* core: erasure-code: sync jerasure/gf-complete submodules (`pr#14424 <https://github.com/ceph/ceph/pull/14424>`_, Loic Dachary)
* core: introduce DirectMessenger (`pr#14755 <https://github.com/ceph/ceph/pull/14755>`_, Casey Bodley, Matt Benjamin)
* core: messages: remove compat cruft (`pr#14475 <https://github.com/ceph/ceph/pull/14475>`_, Sage Weil)
* core,mon: common/Finisher: fix uninitialized variable warning (`issue#19874 <http://tracker.ceph.com/issues/19874>`_, `pr#14979 <https://github.com/ceph/ceph/pull/14979>`_, Sage Weil)
* core: mon,osd: add require_min_compat_client setting to enforce and clarify client compatibility (`pr#14959 <https://github.com/ceph/ceph/pull/14959>`_, Sage Weil)
* core: mon/OSDMonitor: skip prime_pg_temp if mapping is prior to osdmap (`pr#14826 <https://github.com/ceph/ceph/pull/14826>`_, Kefu Chai)
* core: mon/PGMonitor: fix wrongly report "pg stuck in inactive" (`pr#14391 <https://github.com/ceph/ceph/pull/14391>`_, Mingxin Liu)
* core: osd,librados: cmpext support (`pr#14715 <https://github.com/ceph/ceph/pull/14715>`_, Zhengyong Wang, David Disseldorp, Mike Christie)
* core: osd/OSDMap: bump encoding version for require_min_compat_client (`pr#15046 <https://github.com/ceph/ceph/pull/15046>`_, "Yan, Zheng")
* core: osd/PG.cc: Optimistic estimation on PG.last_active (`pr#14799 <https://github.com/ceph/ceph/pull/14799>`_, Xiaoxi Chen)
* core: osd/PG.cc: unify the call of checking whether lock is held (`pr#15013 <https://github.com/ceph/ceph/pull/15013>`_, Jin Cai)
* core: osd/PG: fix possible overflow on unfound objects (`pr#12669 <https://github.com/ceph/ceph/pull/12669>`_, huangjun)
* core: osd/PrimaryLogPG: do not call on_shutdown() if (pg.deleting) (`issue#19902 <http://tracker.ceph.com/issues/19902>`_, `pr#15040 <https://github.com/ceph/ceph/pull/15040>`_, Kefu Chai)
* core: osd/PrimayLogPG: update modified range to include the whole object size for write_full op (`pr#15021 <https://github.com/ceph/ceph/pull/15021>`_, runsisi)
* core: osd/ReplicatedBackend: remove MOSDSubOp cruft from repop_applied (`pr#14358 <https://github.com/ceph/ceph/pull/14358>`_, Jianpeng Ma)
* core: os/filestore/FileJournal: Fix typo in the comment (`pr#14493 <https://github.com/ceph/ceph/pull/14493>`_, Zhou Zhengping)
* core: os/filestore: fix infinit loops in fiemap() (`pr#14367 <https://github.com/ceph/ceph/pull/14367>`_, Ning Yao)
* core,performance: osd, os: reduce fiemap burden (`pr#14640 <https://github.com/ceph/ceph/pull/14640>`_, Piotr Dałek)
* core,performance: os/filestore: use new sleep strategy when io_submit gets EAGAIN. (`pr#14860 <https://github.com/ceph/ceph/pull/14860>`_, Pan Liu)
* core,performance: os/kstore: Added rocksdb bloom filter settings (`pr#13053 <https://github.com/ceph/ceph/pull/13053>`_, Ted-Chang)
* core,tests: ceph_test_rados_api_watch_notify: make LibRadosWatchNotify.Watch3Timeout tolerate thrashing (`issue#19433 <http://tracker.ceph.com/issues/19433>`_, `pr#14899 <https://github.com/ceph/ceph/pull/14899>`_, Sage Weil)
* core,tools: ceph: perfcounter priorities and daemonperf updates to use them (`pr#14793 <https://github.com/ceph/ceph/pull/14793>`_, Sage Weil, Dan Mick)
* core,tools: kv: move 'bluestore-kv' hackery out of KeyValueDB into ceph-kvstore-tool (`issue#19778 <http://tracker.ceph.com/issues/19778>`_, `pr#14895 <https://github.com/ceph/ceph/pull/14895>`_, Sage Weil)
* crush: builder: legacy has chooseleaf_stable = 0 (`pr#14695 <https://github.com/ceph/ceph/pull/14695>`_, Loic Dachary)
* crush: crush_init_workspace starts with struct crush_work (`pr#14696 <https://github.com/ceph/ceph/pull/14696>`_, Loic Dachary)
* crush: update documentation for negative choose step (`pr#14970 <https://github.com/ceph/ceph/pull/14970>`_, Loic Dachary)
* doc: AUTHORS: update tech leads (`pr#14350 <https://github.com/ceph/ceph/pull/14350>`_, Patrick Donnelly)
* doc: correct and improve add user capability section (`pr#14055 <https://github.com/ceph/ceph/pull/14055>`_, Chu, Hua-Rong)
* doc: Correcting the remove bucket example and adding bucket link/unlink examples (`pr#12460 <https://github.com/ceph/ceph/pull/12460>`_, Uday Mullangi)
* doc: dev add a note about ccache (`pr#14478 <https://github.com/ceph/ceph/pull/14478>`_, Abhishek Lekshmanan)
* doc: doc/dev: add some info about FreeBSD (`pr#14503 <https://github.com/ceph/ceph/pull/14503>`_, Willem Jan Withagen)
* doc: fio: update README.md so only the fio ceph engine is built (`pr#15081 <https://github.com/ceph/ceph/pull/15081>`_, Kefu Chai)
* doc: fix link that pointed to a nonexistent file (`pr#14740 <https://github.com/ceph/ceph/pull/14740>`_, Peter Maloney)
* doc: Indicate how to add multiple admin capbabilies (`pr#13956 <https://github.com/ceph/ceph/pull/13956>`_, Chu, Hua-Rong)
* doc: mailmap: add Alibaba into organization map (`pr#14900 <https://github.com/ceph/ceph/pull/14900>`_, James Liu)
* doc: mailmap: update organization info (`pr#14747 <https://github.com/ceph/ceph/pull/14747>`_, liuchang0812)
* doc: mailmap: V12.0.1 credits (`pr#14479 <https://github.com/ceph/ceph/pull/14479>`_, M Ranga Swami Reddy)
* doc: mailmap: Weibing Zhang mailmap affiliation (`pr#15076 <https://github.com/ceph/ceph/pull/15076>`_, Weibing Zhang)
* doc: mailmap: ztczll affiliation (`pr#15079 <https://github.com/ceph/ceph/pull/15079>`_, zhanglei)
* doc: mention teuthology-worker security group (`pr#14748 <https://github.com/ceph/ceph/pull/14748>`_, Nathan Cutler)
* doc: peoplemap: add pdonnell alias (`pr#14352 <https://github.com/ceph/ceph/pull/14352>`_, Patrick Donnelly)
* doc: remove deprecated subcommand in man/8/ceph.rst (`pr#14928 <https://github.com/ceph/ceph/pull/14928>`_, Drunkard Zhang)
* doc: Re-word the warnings about using git subtrees. (`pr#14999 <https://github.com/ceph/ceph/pull/14999>`_, J. Eric Ivancich)
* doc: rgw: Clean up create subuser parameters (`pr#14335 <https://github.com/ceph/ceph/pull/14335>`_, hrchu)
* doc: rgw: correct get usage parameter default value (`pr#14372 <https://github.com/ceph/ceph/pull/14372>`_, hrchu)
* doc: rgw: Get user usage needs to specify user (`pr#14804 <https://github.com/ceph/ceph/pull/14804>`_, hrchu)
* doc: rgw: note rgw_enable_usage_log option in adminops guide (`pr#14803 <https://github.com/ceph/ceph/pull/14803>`_, hrchu)
* doc: rgw: Rewrite Java swift examples (`pr#14268 <https://github.com/ceph/ceph/pull/14268>`_, Chu, Hua-Rong)
* doc: rgw: Rewrite the key management (`pr#14384 <https://github.com/ceph/ceph/pull/14384>`_, hrchu)
* doc: style fix for doc/cephfs/client-config-ref.rst (`pr#14840 <https://github.com/ceph/ceph/pull/14840>`_, Drunkard Zhang)
* doc: two minor fixes (`pr#14494 <https://github.com/ceph/ceph/pull/14494>`_, Drunkard Zhang)
* doc: update the support status of swift static website (`pr#13824 <https://github.com/ceph/ceph/pull/13824>`_, Jing Wenjun)
* doc: v12.0.2 (dev) release notes (`pr#14625 <https://github.com/ceph/ceph/pull/14625>`_, Abhishek Lekshmanan)
* librados: fix rados_pool_list when buf is null (`pr#14859 <https://github.com/ceph/ceph/pull/14859>`_, Sage Weil)
* librbd: cleanup logging code under librbd/io (`pr#14975 <https://github.com/ceph/ceph/pull/14975>`_, runsisi)
* librbd: do not instantiate templates while building tests (`issue#18938 <http://tracker.ceph.com/issues/18938>`_, `pr#14891 <https://github.com/ceph/ceph/pull/14891>`_, Kefu Chai)
* librbd: minor fixes for image trash move (`pr#14834 <https://github.com/ceph/ceph/pull/14834>`_, runsisi)
* librbd: remove redundant check for image id emptiness (`pr#14830 <https://github.com/ceph/ceph/pull/14830>`_, runsisi)
* librbd: silence -Wunused-variable warning (`pr#14953 <https://github.com/ceph/ceph/pull/14953>`_, Kefu Chai)
* mds: add perf counters for file system operations (`pr#14938 <https://github.com/ceph/ceph/pull/14938>`_, Michael Sevilla)
* mds: change_attr++ and set ctime for set_vxattr (`issue#19583 <http://tracker.ceph.com/issues/19583>`_, `pr#14726 <https://github.com/ceph/ceph/pull/14726>`_, Patrick Donnelly)
* mds: fix mantle script to not fail for last rank (`issue#19589 <http://tracker.ceph.com/issues/19589>`_, `pr#14704 <https://github.com/ceph/ceph/pull/14704>`_, Patrick Donnelly)
* mds: fix use-after-free in Locker::file_update_finish() (`issue#19828 <http://tracker.ceph.com/issues/19828>`_, `pr#14991 <https://github.com/ceph/ceph/pull/14991>`_, "Yan, Zheng")
* mds: issue new caps when sending reply to client (`issue#19635 <http://tracker.ceph.com/issues/19635>`_, `pr#14743 <https://github.com/ceph/ceph/pull/14743>`_, "Yan, Zheng")
* mds: relocate PTRWAITER put near get (`pr#14921 <https://github.com/ceph/ceph/pull/14921>`_, Patrick Donnelly)
* mds: remove boost::pool usage and use tcmalloc directly (`issue#18425 <http://tracker.ceph.com/issues/18425>`_, `pr#12792 <https://github.com/ceph/ceph/pull/12792>`_, Zhi Zhang)
* mds: remove "mds log" config option (`issue#18816 <http://tracker.ceph.com/issues/18816>`_, `pr#14652 <https://github.com/ceph/ceph/pull/14652>`_, John Spray)
* mds: support export pinning on directories (`issue#17834 <http://tracker.ceph.com/issues/17834>`_, `pr#14598 <https://github.com/ceph/ceph/pull/14598>`_, "Yan, Zheng", Patrick Donnelly)
* mds: use debug_mds for most subsys (`issue#19734 <http://tracker.ceph.com/issues/19734>`_, `pr#15052 <https://github.com/ceph/ceph/pull/15052>`_, Patrick Donnelly)
* mgr: add machinery for python modules to send MCommands to daemons (`pr#14920 <https://github.com/ceph/ceph/pull/14920>`_, John Spray)
* mgr: add mgr allow * to client.admin (`pr#14864 <https://github.com/ceph/ceph/pull/14864>`_, huanwen ren)
* mgr: do shutdown using finisher so we can do it in the right order (`issue#19743 <http://tracker.ceph.com/issues/19743>`_, `pr#14835 <https://github.com/ceph/ceph/pull/14835>`_, Kefu Chai)
* mgr: do the shutdown in the right order (`issue#19813 <http://tracker.ceph.com/issues/19813>`_, `pr#14952 <https://github.com/ceph/ceph/pull/14952>`_, Kefu Chai)
* mgr: fix crash on set_config from python module with insufficient caps (`issue#19629 <http://tracker.ceph.com/issues/19629>`_, `pr#14706 <https://github.com/ceph/ceph/pull/14706>`_, Tim Serong)
* mgr: fix metadata handling from old MDS daemons (`pr#14161 <https://github.com/ceph/ceph/pull/14161>`_, John Spray)
* mgr: fix python module teardown & add tests (`issue#19407 <http://tracker.ceph.com/issues/19407>`_, `issue#19412 <http://tracker.ceph.com/issues/19412>`_, `issue#19258 <http://tracker.ceph.com/issues/19258>`_, `pr#14232 <https://github.com/ceph/ceph/pull/14232>`_, John Spray)
* mgr: fix session leak (`issue#19591 <http://tracker.ceph.com/issues/19591>`_, `pr#14720 <https://github.com/ceph/ceph/pull/14720>`_, Sage Weil)
* mgr: Misc. bug fixes (`issue#18994 <http://tracker.ceph.com/issues/18994>`_, `pr#14883 <https://github.com/ceph/ceph/pull/14883>`_, John Spray)
* mgr: mkdir bootstrap-mgr (`pr#14824 <https://github.com/ceph/ceph/pull/14824>`_, huanwen ren)
* mgr,mon: mon/MgrMonitor: only propose if we updated (`pr#14645 <https://github.com/ceph/ceph/pull/14645>`_, Sage Weil)
* mgr,mon: mon,mgr: trim osdmap without the help of pgmap (`pr#14504 <https://github.com/ceph/ceph/pull/14504>`_, Kefu Chai)
* mgr: pybind/mgr/rest: completely terminate cherrypy in shutdown (`pr#14995 <https://github.com/ceph/ceph/pull/14995>`_, Tim Serong)
* mgr: redirect python stdout,stderr to ceph log (`pr#14189 <https://github.com/ceph/ceph/pull/14189>`_, Kefu Chai, Tim Serong, Dan Mick)
* mgr: release allocated PyString (`pr#14716 <https://github.com/ceph/ceph/pull/14716>`_, Kefu Chai)
* mgr: remove non-existent MDS daemons from FSMap (`issue#17453 <http://tracker.ceph.com/issues/17453>`_, `pr#14937 <https://github.com/ceph/ceph/pull/14937>`_, Spandan Kumar Sahu)
* mgr,tests: qa/suites: move mgr tests into rados suite (`pr#14687 <https://github.com/ceph/ceph/pull/14687>`_, John Spray)
* mgr: various cleanups (`pr#14802 <https://github.com/ceph/ceph/pull/14802>`_, Kefu Chai)
* mon: check is_shutdown() in timer callbacks (`issue#19825 <http://tracker.ceph.com/issues/19825>`_, `pr#14919 <https://github.com/ceph/ceph/pull/14919>`_, Kefu Chai)
* mon: do not prime_pg_temp creating pgs; clean up pg create conditions (`issue#19826 <http://tracker.ceph.com/issues/19826>`_, `pr#14913 <https://github.com/ceph/ceph/pull/14913>`_, Sage Weil)
* mon: don't call propose_pending in prepare_update() (`issue#19738 <http://tracker.ceph.com/issues/19738>`_, `pr#14711 <https://github.com/ceph/ceph/pull/14711>`_, John Spray)
* mon: logclient: use the seq id of the 1st log entry when resetting session (`issue#19427 <http://tracker.ceph.com/issues/19427>`_, `pr#14927 <https://github.com/ceph/ceph/pull/14927>`_, Kefu Chai)
* mon: mon/OSDMonitor: check get()'s return value instead of bl's length (`pr#14805 <https://github.com/ceph/ceph/pull/14805>`_, Kefu Chai)
* mon: mon/OSDMonitor: cleanup pending_created_pgs after done with it (`pr#14898 <https://github.com/ceph/ceph/pull/14898>`_, Kefu Chai)
* mon: mon/OSDMonitor: do not alter the "created" epoch of a pg (`issue#19787 <http://tracker.ceph.com/issues/19787>`_, `pr#14849 <https://github.com/ceph/ceph/pull/14849>`_, Kefu Chai)
* mon: mon/OSDMonitor: fix output func name in can_mark_out (`pr#14758 <https://github.com/ceph/ceph/pull/14758>`_, xie xingguo)
* mon: mon/OSDMonitor: increase last_epoch_clean's lower bound if possible (`pr#14855 <https://github.com/ceph/ceph/pull/14855>`_, Kefu Chai)
* mon: mon/OSDMonitor: tolerate upgrade from post-kraken dev cluster (`pr#14442 <https://github.com/ceph/ceph/pull/14442>`_, Sage Weil)
* mon: mon/OSDMonitor: update creating_pgs using pending_creatings (`issue#19814 <http://tracker.ceph.com/issues/19814>`_, `pr#14897 <https://github.com/ceph/ceph/pull/14897>`_, Kefu Chai)
* mon: mon/OSDMonitor: update pg_creatings even the new acting set is empty (`issue#19744 <http://tracker.ceph.com/issues/19744>`_, `pr#14730 <https://github.com/ceph/ceph/pull/14730>`_, Kefu Chai)
* mon: mon/PGMap: add up_primary pg number field for pg-dump cmd (`pr#13451 <https://github.com/ceph/ceph/pull/13451>`_, xie xingguo)
* mon: mon/PGMap.cc: fix "osd_epochs" section of dump_basic (`pr#14996 <https://github.com/ceph/ceph/pull/14996>`_, xie xingguo)
* mon: mon/PGMonitor: clean up min/max span warning (`pr#14611 <https://github.com/ceph/ceph/pull/14611>`_, Sage Weil)
* mon: move 'pg map' to OSDMonitor (`pr#14559 <https://github.com/ceph/ceph/pull/14559>`_, Sage Weil)
* mon: osd/PGMonitor: always update pgmap with latest osdmap (`issue#19398 <http://tracker.ceph.com/issues/19398>`_, `pr#14777 <https://github.com/ceph/ceph/pull/14777>`_, Kefu Chai)
* mon: show inactive % in ceph status (`pr#14810 <https://github.com/ceph/ceph/pull/14810>`_, Sage Weil)
* msg: Increase loglevels on some messages (`pr#14707 <https://github.com/ceph/ceph/pull/14707>`_, Willem Jan Withagen)
* msg: msg/async/net_handler: errno should be stored before calling next function (`pr#14985 <https://github.com/ceph/ceph/pull/14985>`_, Zhou Zhengping)
* msg: msg/simple: use my addr when setting sock priority (`issue#19801 <http://tracker.ceph.com/issues/19801>`_, `pr#14878 <https://github.com/ceph/ceph/pull/14878>`_, Kefu Chai)
* msg: src/msg/async/AsyncConnect.cc: Use of sizeof() on a Pointer Type (`pr#14773 <https://github.com/ceph/ceph/pull/14773>`_, Svyatoslav)
* msg: src/msg/simple/Pipe.cc: Fix the inclusion of '}' (`pr#14843 <https://github.com/ceph/ceph/pull/14843>`_, Willem Jan Withagen)
* osd: check fsid is normal before osd mkfs (`pr#13898 <https://github.com/ceph/ceph/pull/13898>`_, song baisen)
* osd: clean nonused work queue (`pr#14990 <https://github.com/ceph/ceph/pull/14990>`_, Wei Jin)
* osd: eliminate snapdir objects and move clone snaps vector into SnapSet (`pr#13610 <https://github.com/ceph/ceph/pull/13610>`_, Sage Weil)
* osd: fix occasional MOSDMap leak (`issue#18293 <http://tracker.ceph.com/issues/18293>`_, `pr#14558 <https://github.com/ceph/ceph/pull/14558>`_, Sage Weil)
* osd: fix typo in comment (`pr#13061 <https://github.com/ceph/ceph/pull/13061>`_, Gu Zhongyan)
* osd: Implement asynchronous scrub sleep (`issue#19497 <http://tracker.ceph.com/issues/19497>`_, `pr#14886 <https://github.com/ceph/ceph/pull/14886>`_, Brad Hubbard)
* osd: Implement peering state timing (`pr#14627 <https://github.com/ceph/ceph/pull/14627>`_, Brad Hubbard)
* osd: objclass sdk (`pr#14723 <https://github.com/ceph/ceph/pull/14723>`_, Neha Ojha)
* osd: osdc/Objecter: more constness (`pr#14819 <https://github.com/ceph/ceph/pull/14819>`_, Kefu Chai)
* osd: osdc: silence warning from `-Wsign-compare` (`pr#14729 <https://github.com/ceph/ceph/pull/14729>`_, Jos Collin)
* osd: pglog trimming fixes (`pr#12882 <https://github.com/ceph/ceph/pull/12882>`_, Zhiqiang Wang)
* osd: pglog: with config, don't assert in the presence of stale diverg… (`issue#17916 <http://tracker.ceph.com/issues/17916>`_, `pr#14648 <https://github.com/ceph/ceph/pull/14648>`_, Greg Farnum)
* osd: put osdmap in mempool (`pr#14780 <https://github.com/ceph/ceph/pull/14780>`_, Sage Weil)
* osd: renamed the new vector name in OSDMap::build_simple_crush_map_from_conf (`pr#14583 <https://github.com/ceph/ceph/pull/14583>`_, Jos Collin)
* osd: ReplicatedBackend::prep_push() remove redundant variable assignments (`pr#14817 <https://github.com/ceph/ceph/pull/14817>`_, Jin Cai)
* osd: sched_scrub() lock pg only if all scrubbing conditions are fulfilled (`pr#14968 <https://github.com/ceph/ceph/pull/14968>`_, Jin Cai)
* osd: simplify past_intervals representation (`pr#14444 <https://github.com/ceph/ceph/pull/14444>`_, Samuel Just, Sage Weil)
* osd: stop mgrc earlier in shutdown() (`issue#19638 <http://tracker.ceph.com/issues/19638>`_, `pr#14904 <https://github.com/ceph/ceph/pull/14904>`_, Kefu Chai)
* osd: stop MgrClient callbacks on shutdown (`issue#19638 <http://tracker.ceph.com/issues/19638>`_, `pr#14896 <https://github.com/ceph/ceph/pull/14896>`_, Sage Weil)
* osd: strip pglog op name (`pr#14764 <https://github.com/ceph/ceph/pull/14764>`_, liuchang0812)
* osd: support dumping long ops (`pr#13019 <https://github.com/ceph/ceph/pull/13019>`_, Zhiqiang Wang)
* osd: switch filestore to default to rocksdb (`pr#14814 <https://github.com/ceph/ceph/pull/14814>`_, Neha Ojha)
* osd: tag fast dispatch messages with min_epoch (`pr#13681 <https://github.com/ceph/ceph/pull/13681>`_, Sage Weil)
* osd: use append(bufferlist &) to avoid unnecessary copy (`pr#12272 <https://github.com/ceph/ceph/pull/12272>`_, Yunchuan Wen)
* osd: zipkin tracing (`pr#14305 <https://github.com/ceph/ceph/pull/14305>`_, Sage Weil, Marios-Evaggelos Kogias, Victor Araujo, Casey Bodley, Andrew Shewmaker, Chendi.Xue)
* performance: buffer, osd: add missing crc cache miss perf counter (`pr#14957 <https://github.com/ceph/ceph/pull/14957>`_, Piotr Dałek)
* performance: osd/PG.cc: loop invariant code motion (`pr#12720 <https://github.com/ceph/ceph/pull/12720>`_, Li Wang)
* pybind: better error msg (`pr#14497 <https://github.com/ceph/ceph/pull/14497>`_, Kefu Chai)
* pybind: fix open flags calculation (`issue#19890 <http://tracker.ceph.com/issues/19890>`_, `pr#15018 <https://github.com/ceph/ceph/pull/15018>`_, "Yan, Zheng")
* qa: qa/added overrides (`pr#14917 <https://github.com/ceph/ceph/pull/14917>`_, Yuri Weinstein)
* qa: qa/suite: replace reference to fs/xfs.yaml (`pr#14756 <https://github.com/ceph/ceph/pull/14756>`_, Yehuda Sadeh)
* qa: qa/suites/rados/singleton-bluestore: concat settings (`pr#14884 <https://github.com/ceph/ceph/pull/14884>`_, Kefu Chai)
* rbd: cls_rbd: default initialize snapshot namespace for legacy clients (`issue#19413 <http://tracker.ceph.com/issues/19413>`_, `pr#14903 <https://github.com/ceph/ceph/pull/14903>`_, Jason Dillaman)
* rbd: common/bit_vector: utilize deep-copy during data decode (`issue#19863 <http://tracker.ceph.com/issues/19863>`_, `pr#15017 <https://github.com/ceph/ceph/pull/15017>`_, Jason Dillaman)
* rbd: import needs to sanity check auto-generated image name (`issue#19128 <http://tracker.ceph.com/issues/19128>`_, `pr#14754 <https://github.com/ceph/ceph/pull/14754>`_, Mykola Golub)
* rbd: info command should indicate if parent is in trash (`pr#14875 <https://github.com/ceph/ceph/pull/14875>`_, Jason Dillaman)
* rbd-mirror: ensure missing images are re-synced when detected (`issue#19811 <http://tracker.ceph.com/issues/19811>`_, `pr#14945 <https://github.com/ceph/ceph/pull/14945>`_, Jason Dillaman)
* rbd-mirror: failover and failback of unmodified image results in split-brain (`issue#19858 <http://tracker.ceph.com/issues/19858>`_, `pr#14963 <https://github.com/ceph/ceph/pull/14963>`_, Jason Dillaman)
* rbd-mirror: image deletions should be handled by assigned instance (`pr#14832 <https://github.com/ceph/ceph/pull/14832>`_, Jason Dillaman)
* rbd-mirror: remove tracking of image names from pool watcher (`pr#14712 <https://github.com/ceph/ceph/pull/14712>`_, Jason Dillaman)
* rbd-mirror: resolve admin socket path names collision (`issue#19907 <http://tracker.ceph.com/issues/19907>`_, `pr#15048 <https://github.com/ceph/ceph/pull/15048>`_, Mykola Golub)
* rbd-nbd: relax size check for newer kernel versions (`issue#19871 <http://tracker.ceph.com/issues/19871>`_, `pr#14976 <https://github.com/ceph/ceph/pull/14976>`_, Mykola Golub)
* rbd: rbd/bench: add notes of default values, it's easy to use (`pr#14762 <https://github.com/ceph/ceph/pull/14762>`_, Zheng Yin)
* rbd: rbd, librbd: migrate atomic_t to std::atomic (`pr#14656 <https://github.com/ceph/ceph/pull/14656>`_, Jesse Williamson)
* rbd: rbd-mirror A/A: proxy InstanceReplayer APIs via InstanceWatcher RPC (`issue#18787 <http://tracker.ceph.com/issues/18787>`_, `pr#13978 <https://github.com/ceph/ceph/pull/13978>`_, Mykola Golub)
* rbd: recognize exclusive option (`pr#14785 <https://github.com/ceph/ceph/pull/14785>`_, Ilya Dryomov)
* rbd: removed spurious error message from mirror pool commands (`pr#14935 <https://github.com/ceph/ceph/pull/14935>`_, Jason Dillaman)
* rbd: stop indefinite thread waiting in krbd udev handling (`issue#17195 <http://tracker.ceph.com/issues/17195>`_, `pr#14051 <https://github.com/ceph/ceph/pull/14051>`_, Spandan Kumar Sahu)
* rbd,tests: qa: krbd write-after-checksum tests (`pr#14836 <https://github.com/ceph/ceph/pull/14836>`_, Ilya Dryomov)
* rbd,tests: qa/workunits/rbd: increased trash deferment period (`pr#14846 <https://github.com/ceph/ceph/pull/14846>`_, Jason Dillaman)
* rbd,tests: qa/workunits: switch to OpenStack Ocata release for RBD testing (`pr#14465 <https://github.com/ceph/ceph/pull/14465>`_, Jason Dillaman)
* rbd,tests: test/librbd/test_librbd.cc: set *features even if RBD_FEATURES is unset (`issue#19865 <http://tracker.ceph.com/issues/19865>`_, `pr#14965 <https://github.com/ceph/ceph/pull/14965>`_, Dan Mick)
* rbd,tests: test/librbd/test_notify.py: don't disable feature in slave (`issue#19716 <http://tracker.ceph.com/issues/19716>`_, `pr#14751 <https://github.com/ceph/ceph/pull/14751>`_, Mykola Golub)
* rbd,tests: test/rbd_mirror: race in TestMockLeaderWatcher.AcquireError (`issue#19405 <http://tracker.ceph.com/issues/19405>`_, `pr#14741 <https://github.com/ceph/ceph/pull/14741>`_, Mykola Golub)
* rbd,tests: test: remove hard-coded image name from RBD metadata test (`issue#19798 <http://tracker.ceph.com/issues/19798>`_, `pr#14848 <https://github.com/ceph/ceph/pull/14848>`_, Jason Dillaman)
* rdma: msg/async/rdma: add inqueue rx chunks perf counter (`pr#14782 <https://github.com/ceph/ceph/pull/14782>`_, Haomai Wang)
* rgw: add bucket size limit check to radosgw-admin (`issue#17925 <http://tracker.ceph.com/issues/17925>`_, `pr#11796 <https://github.com/ceph/ceph/pull/11796>`_, Matt Benjamin)
* rgw: add 'state==SyncState::IncrementalSync' condition when add item … (`pr#14552 <https://github.com/ceph/ceph/pull/14552>`_, Shasha Lu)
* rgw: add support container and object levels of swift bulkupload (`pr#14775 <https://github.com/ceph/ceph/pull/14775>`_, Jing Wenjun)
* rgw: add support for FormPost of Swift API (`issue#17273 <http://tracker.ceph.com/issues/17273>`_, `pr#11179 <https://github.com/ceph/ceph/pull/11179>`_, Radoslaw Zarzynski, Orit Wasserman)
* rgw: civetweb don't go past the array index while calling mg_start (`issue#19749 <http://tracker.ceph.com/issues/19749>`_, `pr#14750 <https://github.com/ceph/ceph/pull/14750>`_, Abhishek Lekshmanan, Jesse Williamson)
* rgw: clean unuse code in cls_statelog_check_state (`pr#10260 <https://github.com/ceph/ceph/pull/10260>`_, weiqiaomiao)
* rgw: cleanup: fix variable name in RGWRados::create_pool() declaration (`pr#14547 <https://github.com/ceph/ceph/pull/14547>`_, Nathan Cutler)
* rgw: cleanup lc continuation (`pr#14906 <https://github.com/ceph/ceph/pull/14906>`_, Jiaying Ren)
* rgw: cls/rgw: list_plain_entries() stops before bi_log entries (`issue#19876 <http://tracker.ceph.com/issues/19876>`_, `pr#14981 <https://github.com/ceph/ceph/pull/14981>`_, Casey Bodley)
* rgw: custom user data header (`issue#19644 <http://tracker.ceph.com/issues/19644>`_, `pr#14592 <https://github.com/ceph/ceph/pull/14592>`_, Pavan Rallabhandi)
* rgw: deduplicate variants of rgw_make_bucket_entry_name(). (`pr#14299 <https://github.com/ceph/ceph/pull/14299>`_, Radoslaw Zarzynski)
* rgw: don't do unneccesary write if buffer with zero length (`pr#14925 <https://github.com/ceph/ceph/pull/14925>`_, fang yuxiang)
* rgw: dont spawn error_repo until lease is acquired (`issue#19446 <http://tracker.ceph.com/issues/19446>`_, `pr#14714 <https://github.com/ceph/ceph/pull/14714>`_, Casey Bodley)
* rgw: drop unused param "bucket" from select_bucket_placement (`pr#14390 <https://github.com/ceph/ceph/pull/14390>`_, Shasha Lu)
* rgw: drop unused port var (`pr#14412 <https://github.com/ceph/ceph/pull/14412>`_, Jiaying Ren)
* rgw: fix broken /crossdomain.xml, /info and /healthcheck of Swift API. (`issue#19520 <http://tracker.ceph.com/issues/19520>`_, `pr#14373 <https://github.com/ceph/ceph/pull/14373>`_, Radoslaw Zarzynski)
* rgw: fix forward request for bulkupload to be applied in multisite (`issue#19645 <http://tracker.ceph.com/issues/19645>`_, `pr#14601 <https://github.com/ceph/ceph/pull/14601>`_, Jing Wenjun)
* rgw: fix handling of --remote in radosgw-admin period commands (`issue#19554 <http://tracker.ceph.com/issues/19554>`_, `pr#14407 <https://github.com/ceph/ceph/pull/14407>`_, Casey Bodley)
* rgw: fix RadosGW hang during multi-chunk upload of AWSv4. (`issue#19754 <http://tracker.ceph.com/issues/19754>`_, `pr#14770 <https://github.com/ceph/ceph/pull/14770>`_, Radoslaw Zarzynski)
* rgw: migrate atomic_t to std::atomic<> (ebirah) (`pr#14839 <https://github.com/ceph/ceph/pull/14839>`_, Jesse Williamson)
* rgw: radosgw-admin: warn that 'realm rename' does not update other clusters (`issue#19746 <http://tracker.ceph.com/issues/19746>`_, `pr#14722 <https://github.com/ceph/ceph/pull/14722>`_, Casey Bodley)
* rgw: rgw_file: cleanup virtual keyword on derived functions (`pr#14908 <https://github.com/ceph/ceph/pull/14908>`_, Gui Hecheng)
* rgw: rgw_ldap: log the ldap err in case of bind failure (`pr#14781 <https://github.com/ceph/ceph/pull/14781>`_, Abhishek Lekshmanan)
* rgw: rgw multisite: automated mdlog trimming (`pr#13111 <https://github.com/ceph/ceph/pull/13111>`_, Casey Bodley)
* rgw: rgw multisite: fixes for meta sync across periods (`issue#18639 <http://tracker.ceph.com/issues/18639>`_, `pr#13070 <https://github.com/ceph/ceph/pull/13070>`_, Casey Bodley)
* rgw: rgw multisite: remove the redundant post in OPT_ZONEGROUP_MODIFY (`pr#14359 <https://github.com/ceph/ceph/pull/14359>`_, Jing Wenjun)
* rgw: RGWPeriodPusher spawns http thread before cr thread (`issue#19834 <http://tracker.ceph.com/issues/19834>`_, `pr#14936 <https://github.com/ceph/ceph/pull/14936>`_, Casey Bodley)
* rgw: rgw_rados drop deprecated global var (`pr#14411 <https://github.com/ceph/ceph/pull/14411>`_, Jiaying Ren)
* rgw: should delete in_stream_req if conn->get_obj(...) return not zero value (`pr#9950 <https://github.com/ceph/ceph/pull/9950>`_, weiqiaomiao)
* rgw: swift: ability to update swift read and write acls separately. (`issue#19289 <http://tracker.ceph.com/issues/19289>`_, `pr#14499 <https://github.com/ceph/ceph/pull/14499>`_, Marcus Watts)
* rgw: swift: disable revocation thread if sleep == 0 (`issue#19499 <http://tracker.ceph.com/issues/19499>`_, `issue#9493 <http://tracker.ceph.com/issues/9493>`_, `pr#14501 <https://github.com/ceph/ceph/pull/14501>`_, Marcus Watts)
* rgw,tests: qa/rgw: add cluster name to path when s3tests scans rgw log (`pr#14845 <https://github.com/ceph/ceph/pull/14845>`_, Casey Bodley)
* rgw,tests: qa/rgw: don't scan radosgw logs for encryption keys on jewel upgrade test (`pr#14697 <https://github.com/ceph/ceph/pull/14697>`_, Casey Bodley)
* rgw,tests: qa/rgw: fix assertions in radosgw_admin task (`pr#14842 <https://github.com/ceph/ceph/pull/14842>`_, Casey Bodley)
* rgw,tests: test/rgw: fixes for test_multi_period_incremental_sync() (`pr#13067 <https://github.com/ceph/ceph/pull/13067>`_, Casey Bodley)
* rgw,tests: test/rgw: fix for empty lists as default arguments (`pr#14816 <https://github.com/ceph/ceph/pull/14816>`_, Casey Bodley)
* rgw: update Beast for streaming reads in asio frontend (`pr#14273 <https://github.com/ceph/ceph/pull/14273>`_, Casey Bodley)
* rgw: using the same bucket num_shards as master zg when create bucket in secondary zg (`issue#19745 <http://tracker.ceph.com/issues/19745>`_, `pr#14388 <https://github.com/ceph/ceph/pull/14388>`_, Shasha Lu)
* rgw: when create_bucket use the same num_shards with info.num_shards (`issue#19745 <http://tracker.ceph.com/issues/19745>`_, `pr#15010 <https://github.com/ceph/ceph/pull/15010>`_, Shasha Lu)
* tests: ceph_test_rados_api_tier: tolerate ENOENT from 'pg scrub' (`pr#14807 <https://github.com/ceph/ceph/pull/14807>`_, Sage Weil)
* tests: cephtool/test.sh error on full tests (`issue#19698 <http://tracker.ceph.com/issues/19698>`_, `pr#14647 <https://github.com/ceph/ceph/pull/14647>`_, Willem Jan Withagen, David Zafman)
* tests: Don't dump core when using EXPECT_DEATH (`pr#14821 <https://github.com/ceph/ceph/pull/14821>`_, Kefu Chai, Brad Hubbard)
* tests: fio_ceph_objectstore: fixes improper write request data lifetime (`pr#14338 <https://github.com/ceph/ceph/pull/14338>`_, Adam Kupczyk)
* tests: fix broken links in upgrade/hammer-jewel-x/stress-split (`issue#19793 <http://tracker.ceph.com/issues/19793>`_, `pr#14831 <https://github.com/ceph/ceph/pull/14831>`_, Nathan Cutler)
* tests: include/denc: support ENCODE_DUMP (`pr#14962 <https://github.com/ceph/ceph/pull/14962>`_, Sage Weil)
* tests: libradosstriper: do not assign garbage to returned value (`pr#15009 <https://github.com/ceph/ceph/pull/15009>`_, Kefu Chai)
* tests: qa/erasure-code: override min_size to 2 (`issue#19770 <http://tracker.ceph.com/issues/19770>`_, `pr#14872 <https://github.com/ceph/ceph/pull/14872>`_, Kefu Chai)
* tests: qa/suites/jewel-x/point-to-point: don't scane for keys on second s3tests either (`pr#14788 <https://github.com/ceph/ceph/pull/14788>`_, Sage Weil)
* tests: qa/suites: Reduce fs combination tests for smoke, use bluestore (`pr#14854 <https://github.com/ceph/ceph/pull/14854>`_, Vasu Kulkarni)
* tests: qa/suites: Revert "qa/suites: add mon-reweight-min-pgs-per-osd = 4" (`pr#14584 <https://github.com/ceph/ceph/pull/14584>`_, Kefu Chai)
* tests: qa/suites/upgrade/jewel-x: add mgr.x role (`pr#14689 <https://github.com/ceph/ceph/pull/14689>`_, Sage Weil)
* tests: qa/suites/upgrade/kraken-x: misc fixes (`pr#14887 <https://github.com/ceph/ceph/pull/14887>`_, Sage Weil)
* tests: qa/tasks/ceph_manager: always fix pgp_num when done with thrashosd task (`issue#19771 <http://tracker.ceph.com/issues/19771>`_, `pr#14931 <https://github.com/ceph/ceph/pull/14931>`_, Kefu Chai)
* tests: qa/tasks: few fixes to get ceph-deploy 1node to working state (`pr#14400 <https://github.com/ceph/ceph/pull/14400>`_, Vasu Kulkarni)
* tests: rados: move cephtool.yaml to new singleton/bluestore subsuite (`issue#19797 <http://tracker.ceph.com/issues/19797>`_, `pr#14847 <https://github.com/ceph/ceph/pull/14847>`_, Nathan Cutler)
* tests: set -x in suites/iozone.sh workunit (`issue#19740 <http://tracker.ceph.com/issues/19740>`_, `pr#14713 <https://github.com/ceph/ceph/pull/14713>`_, Nathan Cutler)
* tests: test/compressor: disable isal tests if not available (`pr#14929 <https://github.com/ceph/ceph/pull/14929>`_, Kefu Chai)
* tests: test: c_read_operations.cc: silence warning from -Wsign-compare (`pr#14888 <https://github.com/ceph/ceph/pull/14888>`_, Jos Collin)
* tests: test/fio_ceph_objectstore: fix fio plugin build failure by engine_data (`pr#15044 <https://github.com/ceph/ceph/pull/15044>`_, lisali)
* tests: test: fixing assert that creates warning: comparison between signed and unsigned integer expressions (`pr#14794 <https://github.com/ceph/ceph/pull/14794>`_, Jos Collin)
* tests: test/fsx: Remove the dead code associated with aio backend (`pr#14905 <https://github.com/ceph/ceph/pull/14905>`_, Zhou Zhengping)
* tests: test/objectstore/: Check put_ref return value (`pr#15007 <https://github.com/ceph/ceph/pull/15007>`_, zhanglei)
* tests: test/osd/osd-dup.sh: lower wb fd throttle limits (`pr#14984 <https://github.com/ceph/ceph/pull/14984>`_, Dan Mick)
* tests: test: use 7130 for crush-classes.sh (`pr#14783 <https://github.com/ceph/ceph/pull/14783>`_, Loic Dachary)
* tests: test: warning: comparison between signed and unsigned integer expressions (`pr#14705 <https://github.com/ceph/ceph/pull/14705>`_, Jos Collin)
* tests,tools: test: kill warnings (`pr#14892 <https://github.com/ceph/ceph/pull/14892>`_, Kefu Chai)
* tools: change compare_exchange_weak to compare_exchange_strong (`pr#15030 <https://github.com/ceph/ceph/pull/15030>`_, Jesse Williamson)
* tools: rados: check for negative return value of rados_create_with_context() as its comment put (`pr#10893 <https://github.com/ceph/ceph/pull/10893>`_, zhang.zezhu)
* tools: rados: out json 'df' values as numbers, not strings (`issue#15546 <http://tracker.ceph.com/issues/15546>`_, `pr#14644 <https://github.com/ceph/ceph/pull/14644>`_, Sage Weil)
v12.0.2 Luminous (dev)
======================
This is the third development checkpoint release of Luminous, the next long term
@ -10,40 +827,6 @@ stable release.
Major changes from v12.0.1
--------------------------
* The original librados rados_objects_list_open (C) and objects_begin
(C++) object listing API, deprecated in Hammer, has finally been
removed. Users of this interface must update their software to use
either the rados_nobjects_list_open (C) and nobjects_begin (C++) API or
the new rados_object_list_begin (C) and object_list_begin (C++) API
before updating the client-side librados library to Luminous.
Object enumeration (via any API) with the latest librados version
and pre-Hammer OSDs is no longer supported. Note that no in-tree
Ceph services rely on object enumeration via the deprecated APIs, so
only external librados users might be affected.
The newest (and recommended) rados_object_list_begin (C) and
object_list_begin (C++) API is only usable on clusters with the
SORTBITWISE flag enabled (Jewel and later). (Note that this flag is
required to be set before upgrading beyond Jewel.)
* CephFS clients without the 'p' flag in their authentication capability
string will no longer be able to set quotas or any layout fields. This
flag previously only restricted modification of the pool and namespace
fields in layouts.
* CephFS directory fragmentation (large directory support) is enabled
by default on new filesystems. To enable it on existing filesystems
use "ceph fs set <fs_name> allow_dirfrags".
* CephFS will generate a health warning if you have fewer standby daemons
than it thinks you wanted. By default this will be 1 if you ever had
a standby, and 0 if you did not. You can customize this using
``ceph fs set <fs> standby_count_wanted <number>``. Setting it
to zero will effectively disable the health check.
* The "ceph mds tell ..." command has been removed. It is superceded
by "ceph tell mds.<id> ..."
* RGW introduces server side encryption of uploaded objects with 3 options for
the management of encryption keys, automatic encryption (only recommended for
test setups), customer provided keys similar to Amazon SSE KMS specification &
using a key management service (openstack barbician)
Notable Changes
---------------
@ -924,51 +1707,6 @@ This is the first development checkpoint release of Luminous series, the next
long term release. We're off to a good start to release Luminous in the spring
of '17.
Major changes from Kraken
-------------------------
* When assigning a network to the public network and not to
the cluster network the network specification of the public
network will be used for the cluster network as well.
In older versions this would lead to cluster services
being bound to 0.0.0.0:<port>, thus making the
cluster service even more publicly available than the
public services. When only specifying a cluster network it
will still result in the public services binding to 0.0.0.0.
* Some variants of the omap_get_keys and omap_get_vals librados
functions have been deprecated in favor of omap_get_vals2 and
omap_get_keys2. The new methods include an output argument
indicating whether there are additional keys left to fetch.
Previously this had to be inferred from the requested key count vs
the number of keys returned, but this breaks with new OSD-side
limits on the number of keys or bytes that can be returned by a
single omap request. These limits were introduced by kraken but
are effectively disabled by default (by setting a very large limit
of 1 GB) because users of the newly deprecated interface cannot
tell whether they should fetch more keys or not. In the case of
the standalone calls in the C++ interface
(IoCtx::get_omap_{keys,vals}), librados has been updated to loop on
the client side to provide a correct result via multiple calls to
the OSD. In the case of the methods used for building
multi-operation transactions, however, client-side looping is not
practical, and the methods have been deprecated. Note that use of
either the IoCtx methods on older librados versions or the
deprecated methods on any version of librados will lead to
incomplete results if/when the new OSD limits are enabled.
* In previous versions, if a client sent an op to the wrong OSD, the OSD
would reply with ENXIO. The rationale here is that the client or OSD is
clearly buggy and we want to surface the error as clearly as possible.
We now only send the ENXIO reply if the osd_enxio_on_misdirected_op option
is enabled (it's off by default). This means that a VM using librbd that
previously would have gotten an EIO and gone read-only will now see a
blocked/hung IO instead.
* When configuring ceph-fuse mounts in /etc/fstab, a new syntax is
available that uses "ceph.<arg>=<val>" in the options column, instead
of putting configuration in the device column. The old style syntax
still works. See the documentation page "Mount CephFS in your
file systems table" for details.
Notable Changes
@ -1475,7 +2213,7 @@ Upgrading from Jewel
to Kraken 11.2.z (or, eventually, Luminous 12.2.z).
* The ``sortbitwise`` flag must be set on the Jewel cluster before upgrading
to Kraken. The latest Jewel (10.2.4+) releases issue a health warning if
to Kraken. The latest Jewel (10.2.8+) releases issue a health warning if
the flag is not set, so this is probably already set. If it is not, Kraken
OSDs will refuse to start and will print and error message in their log.

3
ceph/doc/start/documenting-ceph.rst
View File

@ -346,7 +346,8 @@ Then, download them from a mirror and install them. For example::
sudo yum install ditaa-0.9-13.r74.fc21.noarch.rpm
Once you have installed all these packages, build the documentation by following
the steps given in ``Build the Source``.
the steps given in `Build the Source`_.
Commit the Change
-----------------

4
ceph/doc/start/get-involved.rst
View File

@ -12,11 +12,11 @@ These are exciting times in the Ceph community! Get involved!
| **Blog** | Check the Ceph Blog_ periodically to keep track | http://ceph.com/community/blog/ |
| | of Ceph progress and important announcements. | |
+----------------------+-------------------------------------------------+-----------------------------------------------+
| **Planet Ceph** | Check the blog aggregation on Planet Ceph for | http://ceph.com/community/planet-ceph/ |
| **Planet Ceph** | Check the blog aggregation on Planet Ceph for | https://ceph.com/category/planet/ |
| | interesting stories, information and | |
| | experiences from the community. | |
+----------------------+-------------------------------------------------+-----------------------------------------------+
| **Wiki** | Check the Ceph Wiki is a source for more | https://wiki.ceph.com/ |
| **Wiki** | Check the Ceph Wiki is a source for more | http://wiki.ceph.com/ |
| | community and development related topics. You | |
| | can find there information about blueprints, | |
| | meetups, the Ceph Developer Summits and more. | |

9
ceph/doc/start/quick-ceph-deploy.rst
View File

@ -38,14 +38,11 @@ Create a Cluster
================
If at any point you run into trouble and you want to start over, execute
the following to purge the configuration::
ceph-deploy purgedata {ceph-node} [{ceph-node}]
ceph-deploy forgetkeys
To purge the Ceph packages too, you may also execute::
the following to purge the Ceph packages, and erase all its data and configuration::
ceph-deploy purge {ceph-node} [{ceph-node}]
ceph-deploy purgedata {ceph-node} [{ceph-node}]
ceph-deploy forgetkeys
If you execute ``purge``, you must re-install Ceph.

12
ceph/examples/librados/Makefile
View File

@ -1,7 +1,7 @@
CXX?=g++
CXX_FLAGS?=-std=c++11 -Wall -Wextra -Werror -g
CXX_LIBS?=-lboost_system -lrados
CXX_LIBS?=-lboost_system -lrados -lradosstriper
CXX_INC?=$(LOCAL_LIBRADOS_INC)
CXX_CC=$(CXX) $(CXX_FLAGS) $(CXX_INC) $(LOCAL_LIBRADOS) $(CXX_LIBS)
@ -13,11 +13,12 @@ CC_CC=$(CC) $(CC_FLAGS) $(CC_INC) $(LOCAL_LIBRADOS) $(CC_LIBS)
# Relative path to the Ceph source:
CEPH_SRC_HOME?=../../src
CEPH_BLD_HOME?=../../build
LOCAL_LIBRADOS?=-L$(CEPH_SRC_HOME)/.libs/ -Wl,-rpath,$(CEPH_SRC_HOME)/.libs
LOCAL_LIBRADOS?=-L$(CEPH_BLD_HOME)/lib/ -Wl,-rpath,$(CEPH_BLD_HOME)/lib
LOCAL_LIBRADOS_INC?=-I$(CEPH_SRC_HOME)/include
all: hello_world_cpp hello_world_c
all: hello_world_cpp hello_radosstriper_cpp hello_world_c
# Build against the system librados instead of the one in the build tree:
all-system: LOCAL_LIBRADOS=
@ -27,9 +28,12 @@ all-system: all
hello_world_cpp: hello_world.cc
$(CXX_CC) -o hello_world_cpp hello_world.cc
hello_radosstriper_cpp: hello_radosstriper.cc
$(CXX_CC) -o hello_radosstriper_cpp hello_radosstriper.cc
hello_world_c: hello_world_c.c
$(CC_CC) -o hello_world_c hello_world_c.c
clean:
rm -f hello_world_cpp hello_world_c
rm -f hello_world_cpp hello_radosstriper_cpp hello_world_c

102
ceph/examples/librados/hello_radosstriper.cc Normal file
View File

@ -0,0 +1,102 @@
#include "rados/librados.hpp"
#include "radosstriper/libradosstriper.hpp"
#include <iostream>
#include <string>
int main(int argc, char* argv[])
{
if(argc != 6)
{
std::cout <<"Please put in correct params\n"<<
"Stripe Count:\n"<<
"Object Size:\n" <<
"File Name:\n" <<
"Object Name:\n"
"Pool Name:"<< std::endl;
return EXIT_FAILURE;
}
uint32_t strip_count = std::stoi(argv[1]);
uint32_t obj_size = std::stoi(argv[2]);
std::string fname = argv[3];
std::string obj_name = argv[4];
std::string pool_name = argv[5];
int ret = 0;
librados::IoCtx io_ctx;
librados::Rados cluster;
libradosstriper::RadosStriper* rs = new libradosstriper::RadosStriper;
// make sure the keyring file is in /etc/ceph/ and is world readable
ret = cluster.init2("client.admin","ceph",0);
if( ret < 0)
{
std::cerr << "Couldn't init cluster "<< ret << std::endl;
}
// make sure ceph.conf is in /etc/ceph/ and is world readable
ret = cluster.conf_read_file("ceph.conf");
if( ret < 0)
{
std::cerr << "Couldn't read conf file "<< ret << std::endl;
}
ret = cluster.connect();
if(ret < 0)
{
std::cerr << "Couldn't connect to cluster "<< ret << std::endl;
}
else
{
std::cout << "Connected to Cluster"<< std::endl;
}
ret = cluster.ioctx_create(pool_name.c_str(), io_ctx);
if(ret < 0)
{
std::cerr << "Couldn't Create IO_CTX"<< ret << std::endl;
}
ret = libradosstriper::RadosStriper::striper_create(io_ctx,rs);
if(ret < 0)
{
std::cerr << "Couldn't Create RadosStriper"<< ret << std::endl;
delete rs;
}
uint64_t alignment = 0;
ret = io_ctx.pool_required_alignment2(&alignment);
if(ret < 0)
{
std::cerr << "IO_CTX didn't give alignment "<< ret
<< "\n Is this an erasure coded pool? "<< std::endl;
delete rs;
io_ctx.close();
cluster.shutdown();
return EXIT_FAILURE;
}
std::cout << "Pool alignment: "<< alignment << std::endl;
rs->set_object_layout_stripe_unit(alignment);
// how many objects are we striping across?
rs->set_object_layout_stripe_count(strip_count);
// how big should each object be?
rs->set_object_layout_object_size(obj_size);
std::string err = "no_err";
librados::bufferlist bl;
bl.read_file(fname.c_str(),&err);
if(err != "no_err")
{
std::cout << "Error reading file into bufferlist: "<< err << std::endl;
delete rs;
io_ctx.close();
cluster.shutdown();
return EXIT_FAILURE;
}
std::cout << "Writing: " << fname << "\nas: "<< obj_name << std::endl;
rs->write_full(obj_name,bl);
std::cout << "done with: " << fname << std::endl;
delete rs;
io_ctx.close();
cluster.shutdown();
}

9
ceph/install-deps.sh
View File

@ -22,6 +22,7 @@ export LC_ALL=C # the following is vulnerable to i18n
if [ x`uname`x = xFreeBSDx ]; then
$SUDO pkg install -yq \
devel/git \
devel/gperf \
devel/gmake \
devel/cmake \
devel/yasm \
@ -29,7 +30,6 @@ if [ x`uname`x = xFreeBSDx ]; then
devel/boost-python-libs \
devel/valgrind \
devel/pkgconf \
devel/libatomic_ops \
devel/libedit \
devel/libtool \
devel/google-perftools \
@ -52,6 +52,7 @@ if [ x`uname`x = xFreeBSDx ]; then
emulators/fuse \
java/junit \
lang/python27 \
devel/py-pip \
devel/py-argparse \
devel/py-nose \
www/py-flask \
@ -59,6 +60,9 @@ if [ x`uname`x = xFreeBSDx ]; then
sysutils/flock \
sysutils/fusefs-libs \
# Now use pip to install some extra python modules
pip install pecan
exit
else
source /etc/os-release
@ -159,7 +163,8 @@ function populate_wheelhouse() {
# although pip comes with virtualenv, having a recent version
# of pip matters when it comes to using wheel packages
pip --timeout 300 $install 'setuptools >= 0.8' 'pip >= 7.0' 'wheel >= 0.24' || return 1
# workaround of https://github.com/pypa/setuptools/issues/1042
pip --timeout 300 $install 'setuptools >= 0.8,< 36' 'pip >= 7.0' 'wheel >= 0.24' || return 1
if test $# != 0 ; then
pip --timeout 300 $install $@ || return 1
fi

1
ceph/mirroring/MIRRORS
View File

@ -8,3 +8,4 @@ us-east.ceph.com: Tyler Bishop <tyler.bishop@beyondhosting.net>
hk.ceph.com: Mart van Santen <mart@greenhost.nl>
fr.ceph.com: Adrien Gillard <gillard.adrien@gmail.com>
uk.ceph.com: Tim Bishop <T.D.Bishop@kent.ac.uk>
cn.ceph.com: USTC LUG <lug@ustc.edu.cn>

5
ceph/qa/clusters/extra-client.yaml
View File

@ -7,3 +7,8 @@ openstack:
- volumes: # attached to each instance
count: 3
size: 10 # GB
overrides:
ceph:
conf:
osd:
osd shutdown pgref assert: true

4
ceph/qa/clusters/fixed-1.yaml
View File

@ -6,5 +6,9 @@ overrides:
osd crush chooseleaf type: 0
osd pool default pg num: 128
osd pool default pgp num: 128
ceph:
conf:
osd:
osd shutdown pgref assert: true
roles:
- [mon.a, mgr.x, osd.0, osd.1, osd.2, client.0]

5
ceph/qa/clusters/fixed-2.yaml
View File

@ -5,3 +5,8 @@ openstack:
- volumes: # attached to each instance
count: 3
size: 10 # GB
overrides:
ceph:
conf:
osd:
osd shutdown pgref assert: true

5
ceph/qa/clusters/fixed-3-cephfs.yaml
View File

@ -9,3 +9,8 @@ openstack:
log-rotate:
ceph-mds: 10G
ceph-osd: 10G
overrides:
ceph:
conf:
osd:
osd shutdown pgref assert: true

5
ceph/qa/clusters/fixed-3.yaml
View File

@ -6,3 +6,8 @@ openstack:
- volumes: # attached to each instance
count: 3
size: 10 # GB
overrides:
ceph:
conf:
osd:
osd shutdown pgref assert: true

Some files were not shown because too many files have changed in this diff Show More