proxmox-mirrors/mirror_frr
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_frr synced 2025-07-24 13:21:51 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
25,370 Commits 4 Branches 191 Tags 125 MiB
16f3db2d8c
Commit Graph

4529 Commits

Author SHA1 Message Date
Philippe Guibert
7a52f27e75 zebra: RTM_GETNEIGH messages may be used by nhrp 
When NHRP registers to zebra to receive link layer events related to
gre interfaces, then it is interested in receiving also RTM_GETNEIGH
messages.

Fixes ("b3b751046495") nhrpd: link layer registration to notifications

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
 2021-08-17 09:07:31 +02:00
Renato Westphal
1dfa8b8991
Merge pull request #9380 from mjstapp/fix_static_lsp_cli 
zebra: mpls validation and static lsp fixes
 2021-08-16 12:06:01 -03:00
Igor Ryzhov
f0010840e8
Merge pull request #9389 from mjstapp/fix_netlink_if_name_sa 
zebra: interface name must be a valid string
 2021-08-14 02:14:44 +03:00
Mark Stapp
e9f79fff57 zebra: interface name must be a valid string 
Validate incoming netlink interface name strings.

Signed-off-by: Mark Stapp <mjs.ietf@gmail.com>
 2021-08-13 16:06:07 -04:00
Igor Ryzhov
1523c0f9ee
Merge pull request #9371 from donaldsharp/zebra_evpn_getl 
zebra: Ensure stream is long enough
 2021-08-13 14:06:37 +03:00
Donald Sharp
a876da9b08
Merge pull request #9374 from mjstapp/fix_nhg_add_leak 
zebra: clean up nhg allocations in error path
 2021-08-12 15:34:07 -04:00
Donald Sharp
86d87c5352 zebra: Ensure stream is long enough 
In zebra_evpn_proc_remote_nh if we do not pass in a long
enough stream, the stream reads will fail.  Ensure that
we have enough data.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-12 15:29:47 -04:00
Mark Stapp
1722cef455
Merge pull request #9304 from donaldsharp/zebra_random_stuff 
Zebra random stuff
 2021-08-12 10:16:46 -04:00
Mark Stapp
a44e310631 zebra: mpls validation and static lsp fixes 
Handle TYPE_IFINDEX nexthops more consistently in a few places;
be more specific about a few integer return values that were
being treated as booleans.

Signed-off-by: Mark Stapp <mjs.ietf@gmail.com>
 2021-08-12 08:53:53 -04:00
Mark Stapp
fd99142ab7 zebra: clean up nhg allocations in error path 
Clean up allocated nhgs in error path in zread_nhg_add().

Signed-off-by: Mark Stapp <mjs.ietf@gmail.com>
 2021-08-11 10:41:53 -04:00
Donald Sharp
c472a97080
Merge pull request #9367 from mjstapp/fix_rt_netlink_af 
zebra: ignore unknown address-family in netlink route msg
 2021-08-11 08:11:39 -04:00
Mark Stapp
deb28338de zebra: ignore unknown address-family in netlink route msg 
Ignore AFs we don't handle in incoming netlink route
updates.

Signed-off-by: Mark Stapp <mjs.ietf@gmail.com>
 2021-08-10 11:44:08 -04:00
Sri Mohana Singamsetty
dd4c59d79a
Merge pull request #9236 from AnuradhaKaruppiah/v6-nh-rmac 
zebra: use a separate dummy prefix for referencing v6 nexthops
 2021-08-10 08:20:55 -07:00
Igor Ryzhov
bdb7b7c5d9
Merge pull request #9321 from donaldsharp/no_leak_re 
zebra: Prevent memory leak if route is rejected early
 2021-08-10 11:39:30 +03:00
Donald Sharp
38c764dde4 zebra: Properly note add/update for rib_add_multipath_nhe 
When calling rib_add_multipath_nhe ensure that we have
well aligned return codes that mean something so that
interersted parties can properly handle the situation.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-09 08:06:33 -04:00
Donald Sharp
f94a7703c0 zebra: Prevent memory leak if route is rejected early 
When receiving a route via zapi, if the route is rejected
there exists a code path where we would not free the corresponding
re created.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-09 07:55:07 -04:00
Donald Sharp
572bc3167f zebra: Delete rib_lookup_and_dump since it is not used 
The rib_lookup_and_dump function is never used, remove

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-06 10:04:40 -04:00
Donald Sharp
42aa465ed1 zebra: Remove rib_lookup_and_pushup function 
The rib_lookup_and_pushup function, from what I can tell, was
used more when static route processing and connected routes were
more closely integrated in zebra.  The goal was when we were adding
a new address to remove the connected route and then allow processing
of the new address.  With the re-org a few years ago to seperate
out connected routes as well as static routes, I believe this is
no longer needed.

on BSD, without this code change we have this log:
2021/08/05 14:33:38 ZEBRA: [QEVVE-G3FQQ] rib_meta_queue_add: (0:0):10.40.30.0/24: queued rn 0x802022bb0 into sub-queue 4
2021/08/05 14:33:38 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 200 Type: RTM_DELETE
2021/08/05 14:33:38 ZEBRA: [V3NSB-BPKBD] Kernel: GATEWAY DONE PROTO1
2021/08/05 14:33:38 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 15
2021/08/05 14:33:38 ZEBRA: [MJD4M-0AAAR] Kernel: pid 53305, rtm_addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:33:38 ZEBRA: [Y9Y5K-JJ7NT] rtm_read: got rtm of type 2 (RTM_DELETE) addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:33:38 ZEBRA: [V17DT-1FJEN] kernel_rtm: 10.40.30.0/24: successfully did NH 9.8.6.7
2021/08/05 14:33:38 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 164 Type: RTM_NEWADDR
2021/08/05 14:33:38 ZEBRA: [V3NSB-BPKBD] Kernel:
2021/08/05 14:33:38 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 4664
2021/08/05 14:33:38 ZEBRA: [MJD4M-0AAAR] Kernel: pid 0, rtm_addrs {DST}
2021/08/05 14:33:38 ZEBRA: [M09CX-TKB4N] ifam_read_mesg: ifindex 1, ifname vtnet0, ifam_addrs {NETMASK,IFP,IFA,BRD}, ifam_flags 0x0, addr 10.40.30.1/24 broad 10.40.30.255 dst (unspec) gateway (unspec)
2021/08/05 14:33:38 ZEBRA: [MFYWV-KH3MC] rib_add_multipath_nhe: (0:0):10.40.30.0/24: Inserting route rn 0x802022bb0, re 0x8032973a0 (connected) existing 0x0, same_count 0
2021/08/05 14:33:38 ZEBRA: [Q4T2G-E2SQF] rib_add_multipath_nhe: dumping RE entry 0x8032973a0 for 10.40.30.0/24 vrf default(0)
2021/08/05 14:33:38 ZEBRA: [M5M58-9PD2R] 10.40.30.0/24: uptime == 1379355, type == 2, instance == 0, table == 0
2021/08/05 14:33:38 ZEBRA: [RVZMM-N7DME] 10.40.30.0/24: metric == 1, mtu == 0, distance == 0, flags == None status == None
2021/08/05 14:33:38 ZEBRA: [Q1NW5-NWY7P] 10.40.30.0/24: nexthop_num == 1, nexthop_active_num == 0
2021/08/05 14:33:38 ZEBRA: [TFHQ8-TC30H] 10.40.30.0/24: NH vtnet0[1] vrf default(0)  with flags
2021/08/05 14:33:38 ZEBRA: [SCETK-GQ9E4] 10.40.30.0/24: dump complete
2021/08/05 14:33:38 ZEBRA: [QEVVE-G3FQQ] rib_meta_queue_add: (0:0):10.40.30.0/24: queued rn 0x802022bb0 into sub-queue 2
2021/08/05 14:33:38 ZEBRA: [MFYWV-KH3MC] rib_add_multipath_nhe: (0:?):10.40.30.0/24 (MRIB): Inserting route rn 0x802022f30, re 0x803297340 (connected) existing 0x0, same_count 0
2021/08/05 14:33:38 ZEBRA: [Q4T2G-E2SQF] rib_add_multipath_nhe: dumping RE entry 0x803297340 for 10.40.30.0/24 vrf default(0)
2021/08/05 14:33:38 ZEBRA: [M5M58-9PD2R] 10.40.30.0/24: uptime == 1379355, type == 2, instance == 0, table == 0
2021/08/05 14:33:38 ZEBRA: [RVZMM-N7DME] 10.40.30.0/24: metric == 1, mtu == 0, distance == 0, flags == None status == None
2021/08/05 14:33:38 ZEBRA: [Q1NW5-NWY7P] 10.40.30.0/24: nexthop_num == 1, nexthop_active_num == 0
2021/08/05 14:33:38 ZEBRA: [TFHQ8-TC30H] 10.40.30.0/24: NH vtnet0[1] vrf default(0)  with flags
2021/08/05 14:33:38 ZEBRA: [SCETK-GQ9E4] 10.40.30.0/24: dump complete
2021/08/05 14:33:38 ZEBRA: [GCGMT-SQR82] rib_link: (0:?):10.40.30.0/24 (MRIB): rn 0x802022f30 adding dest
2021/08/05 14:33:38 ZEBRA: [QEVVE-G3FQQ] rib_meta_queue_add: (0:0):10.40.30.0/24 (MRIB): queued rn 0x802022f30 into sub-queue 2
2021/08/05 14:33:38 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 240 Type: RTM_ADD
2021/08/05 14:33:38 ZEBRA: [V3NSB-BPKBD] Kernel: UP PINNED
2021/08/05 14:33:38 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 0
2021/08/05 14:33:38 ZEBRA: [MJD4M-0AAAR] Kernel: pid 0, rtm_addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:33:38 ZEBRA: [K0KVE-2GJA1] default(0:0):10.40.30.0/24: Processing rn 0x802022bb0
2021/08/05 14:33:38 ZEBRA: [RWCK7-TX4HT] default(0:0):10.40.30.0/24: Examine re 0x8032973a0 (connected) status: Changed flags: None dist 0 metric 1
2021/08/05 14:33:38 ZEBRA: [RWCK7-TX4HT] default(0:0):10.40.30.0/24: Examine re 0x8032970a0 (static) status: None flags: Recursion RR Distance dist 1 metric 0
2021/08/05 14:33:38 ZEBRA: [NYYJJ-0Q8QG] default(0:0):10.40.30.0/24: After processing: old_selected 0x0 new_selected 0x8032973a0 old_fib 0x0 new_fib 0x8032973a0
2021/08/05 14:33:38 ZEBRA: [RT9DY-ZS2KN] default(0:0):10.40.30.0/24: Adding route rn 0x802022bb0, re 0x8032973a0 (connected)
2021/08/05 14:33:38 ZEBRA: [PP3BZ-RABJN] default(0:0):10.40.30.0/24: rn 0x802022bb0 dequeued from sub-queue 2
2021/08/05 14:33:38 ZEBRA: [K0KVE-2GJA1] default(0:0):10.40.30.0/24: Processing rn 0x802022f30
2021/08/05 14:33:38 ZEBRA: [RWCK7-TX4HT] default(0:0):10.40.30.0/24: Examine re 0x803297340 (connected) status: Changed flags: None dist 0 metric 1
2021/08/05 14:33:38 ZEBRA: [NYYJJ-0Q8QG] default(0:0):10.40.30.0/24: After processing: old_selected 0x0 new_selected 0x803297340 old_fib 0x0 new_fib 0x803297340
2021/08/05 14:33:38 ZEBRA: [RT9DY-ZS2KN] default(0:0):10.40.30.0/24: Adding route rn 0x802022f30, re 0x803297340 (connected)
2021/08/05 14:33:38 ZEBRA: [PP3BZ-RABJN] default(0:0):10.40.30.0/24: rn 0x802022f30 dequeued from sub-queue 2
2021/08/05 14:33:38 ZEBRA: [K0KVE-2GJA1] default(0:0):10.40.30.0/24: Processing rn 0x802022bb0
2021/08/05 14:33:38 ZEBRA: [RWCK7-TX4HT] default(0:0):10.40.30.0/24: Examine re 0x8032973a0 (connected) status: Queued flags: Selected dist 0 metric 1
2021/08/05 14:33:38 ZEBRA: [RWCK7-TX4HT] default(0:0):10.40.30.0/24: Examine re 0x8032970a0 (static) status: None flags: Recursion RR Distance dist 1 metric 0
2021/08/05 14:33:38 ZEBRA: [NYYJJ-0Q8QG] default(0:0):10.40.30.0/24: After processing: old_selected 0x8032973a0 new_selected 0x8032973a0 old_fib 0x8032973a0 new_fib 0x8032973a0
2021/08/05 14:33:38 ZEBRA: [PP3BZ-RABJN] default(0:0):10.40.30.0/24: rn 0x802022bb0 dequeued from sub-queue 4
2021/08/05 14:33:38 ZEBRA: [GHWHS-ZKQM5] update_from_ctx: default(0:0):10.40.30.0/24: SELECTED, re 0x8032973a0
2021/08/05 14:33:38 ZEBRA: [TS3SH-1276M] default(0:0):10.40.30.0/24 update_from_ctx(): no fib nhg
2021/08/05 14:33:38 ZEBRA: [HKQXC-4STSK] default(0:0):10.40.30.0/24 update_from_ctx(): rib nhg matched, changed 'false'
2021/08/05 14:33:38 ZEBRA: [HBZNK-5H1X0] (0:0):10.40.30.0/24: Redist update re 0x8032973a0 (connected), old 0x0 (None)
2021/08/05 14:33:38 ZEBRA: [GHWHS-ZKQM5] update_from_ctx: default(0:0):10.40.30.0/24: SELECTED, re 0x8032973a0
2021/08/05 14:33:38 ZEBRA: [TS3SH-1276M] default(0:0):10.40.30.0/24 update_from_ctx(): no fib nhg
2021/08/05 14:33:38 ZEBRA: [HKQXC-4STSK] default(0:0):10.40.30.0/24 update_from_ctx(): rib nhg matched, changed 'false'
2021/08/05 14:33:38 ZEBRA: [HBZNK-5H1X0] (0:0):10.40.30.0/24: Redist update re 0x8032973a0 (connected), old 0x0 (None)

With this code change:

2021/08/05 14:41:24 ZEBRA: [MFYWV-KH3MC] rib_add_multipath_nhe: (0:?):10.10.40.0/24: Inserting route rn 0x802022f30, re 0x8021cbe60 (static) existing 0x0, same_count 0
2021/08/05 14:41:24 ZEBRA: [RT9DY-ZS2KN] default(0:0):10.10.40.0/24: Adding route rn 0x802022f30, re 0x8021cbe60 (static)
2021/08/05 14:41:24 ZEBRA: [V17DT-1FJEN] kernel_rtm: 10.10.40.0/24: successfully did NH 9.8.6.7
2021/08/05 14:41:24 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 200 Type: RTM_ADD
2021/08/05 14:41:24 ZEBRA: [V3NSB-BPKBD] Kernel: UP GATEWAY DONE PROTO1
2021/08/05 14:41:24 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 0
2021/08/05 14:41:24 ZEBRA: [MJD4M-0AAAR] Kernel: pid 60818, rtm_addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:41:24 ZEBRA: [Y9Y5K-JJ7NT] rtm_read: got rtm of type 1 (RTM_ADD) addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:41:24 ZEBRA: [TS3SH-1276M] default(0:0):10.10.40.0/24 update_from_ctx(): no fib nhg
2021/08/05 14:41:24 ZEBRA: [HKQXC-4STSK] default(0:0):10.10.40.0/24 update_from_ctx(): rib nhg matched, changed 'true'
2021/08/05 14:41:24 ZEBRA: [HBZNK-5H1X0] (0:0):10.10.40.0/24: Redist update re 0x8021cbe60 (static), old 0x0 (None)
2021/08/05 14:42:06 ZEBRA: [ZJ4AV-JEMJ3] dplane_intf_addr_set
2021/08/05 14:42:06 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 164 Type: RTM_NEWADDR
2021/08/05 14:42:06 ZEBRA: [V3NSB-BPKBD] Kernel:
2021/08/05 14:42:06 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 4664
2021/08/05 14:42:06 ZEBRA: [MJD4M-0AAAR] Kernel: pid 0, rtm_addrs {DST}
2021/08/05 14:42:06 ZEBRA: [M09CX-TKB4N] ifam_read_mesg: ifindex 1, ifname vtnet0, ifam_addrs {NETMASK,IFP,IFA,BRD}, ifam_flags 0x0, addr 10.10.40.3/24 broad 10.10.40.255 dst (unspec) gateway (unspec)
2021/08/05 14:42:06 ZEBRA: [MFYWV-KH3MC] rib_add_multipath_nhe: (0:0):10.10.40.0/24: Inserting route rn 0x802022f30, re 0x80308c4c0 (connected) existing 0x0, same_count 0
2021/08/05 14:42:06 ZEBRA: [MFYWV-KH3MC] rib_add_multipath_nhe: (0:?):10.10.40.0/24 (MRIB): Inserting route rn 0x802023160, re 0x80308c460 (connected) existing 0x0, same_count 0
2021/08/05 14:42:06 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 240 Type: RTM_ADD
2021/08/05 14:42:06 ZEBRA: [V3NSB-BPKBD] Kernel: UP PINNED
2021/08/05 14:42:06 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 0
2021/08/05 14:42:06 ZEBRA: [MJD4M-0AAAR] Kernel: pid 0, rtm_addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:42:06 ZEBRA: [RG9Y6-E93A0] default(0:0):10.10.40.0/24: Updating route rn 0x802022f30, re 0x80308c4c0 (connected) old 0x8021cbe60 (static)
2021/08/05 14:42:06 ZEBRA: [RT9DY-ZS2KN] default(0:0):10.10.40.0/24: Adding route rn 0x802023160, re 0x80308c460 (connected)
2021/08/05 14:42:06 ZEBRA: [THSYN-E2XFY][EC 100663299] rtm_write: write : Address already in use (48)
2021/08/05 14:42:06 ZEBRA: [RV5F2-MQGZG][EC 100663303] kernel_rtm: 10.10.40.0/24: rtm_write() unexpectedly returned -5 for command RTM_DELETE
2021/08/05 14:42:06 ZEBRA: [ZPR30-5G1FB] Kernel: Len: 200 Type: RTM_DELETE
2021/08/05 14:42:06 ZEBRA: [V3NSB-BPKBD] Kernel: UP PROTO1
2021/08/05 14:42:06 ZEBRA: [HDTM1-ENZNM] Kernel: message seq 1
2021/08/05 14:42:06 ZEBRA: [MJD4M-0AAAR] Kernel: pid 60818, rtm_addrs {DST,GATEWAY,NETMASK}
2021/08/05 14:42:06 ZEBRA: [XASXT-GF69Y] kernel_rtm: No useful nexthops were found in RIB prefix 10.10.40.0/24
2021/08/05 14:42:06 ZEBRA: [TS3SH-1276M] default(0:0):10.10.40.0/24 update_from_ctx(): no fib nhg
2021/08/05 14:42:06 ZEBRA: [HKQXC-4STSK] default(0:0):10.10.40.0/24 update_from_ctx(): rib nhg matched, changed 'false'
2021/08/05 14:42:06 ZEBRA: [HBZNK-5H1X0] (0:0):10.10.40.0/24: Redist update re 0x80308c4c0 (connected), old 0x8021cbe60 (static)

netstat -rn:

10.10.40.0/24      link#1             U        vtnet0
10.10.40.3         link#1             UHS         lo0

show ip route:
C>* 10.10.40.0/24 [0/1] is directly connected, vtnet0, 00:18:48
S   10.10.40.0/24 [1/0] via 9.8.6.7, vtnet0, weight 1, 00:19:30

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-06 10:04:40 -04:00
Donald Sharp
38ef05ea33 zebra: debug zebra kernel msgdump is linux specific 
The command `debug zebra kernel msgdump is netlink specific.
There is no point at all to allow this to be configed on non
netlink platforms.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-06 10:04:40 -04:00
Donald Sharp
e658173ae6 zebra: Convert srcdest_rnode2str to %pRN in zebra_rib.c 
There were a bunch of places where we converted the
route node to a prefix string via srcdest_rnode2str when
we should have been using %pRN in zebra_rib.c.  Just
convert over the ones we should to use it.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-06 10:04:40 -04:00
Donald Sharp
f0afc61d58 zebra: short-circuit rib_process when nothing to do 
When we are calling rib_process and the route_node
in question has no dest, there is no work to do here
at all.  As such we should just return before
attempting to do any other work.  This is just a tiny bit
of simplification being done.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-06 10:02:53 -04:00
Donald Sharp
6140b3b41b zebra: prevent crash when nhlfe is NULL 
There exists a call path where the nhlfe_alloc can return NULL
for blackhole nexthops.  In this case we were still trying
to save the nhlfe pointer causing a crash when we attempted
to add it to a self-contained list.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-04 13:38:25 -04:00
Donald Sharp
10cc80cafd zebra: don't use default case when switching over enum nexthop 
Do not use the `default` case when switching over an enumerated
type.  This allows the code to fail to compile when we add a
new enumeration.  Thus allowing us developers to know all
the places in the code we'll need to touch.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-08-04 13:34:03 -04:00
Russ White
11093fc905
Merge pull request #9231 from idryzhov/zebra-rmap-set-src 
zebra: remove checks for src address existence when using "set src"
 2021-08-03 09:22:18 -04:00
Russ White
1358f9d10a
Merge pull request #9259 from opensourcerouting/moar-json 
*: can't get enough JSON
 2021-08-03 09:13:12 -04:00
Donatas Abraitis
71c06f610f
Merge pull request #9258 from mjstapp/fix_rule_strlcpy 
zebra: use strlcpy in dplane_rule_init
 2021-08-03 09:12:38 +03:00
Renato Westphal
488599bfa2
Merge pull request #9232 from idryzhov/interface-node-cleanup 
*: cleanup interface node installation
 2021-08-02 21:13:29 -03:00
Renato Westphal
c15dc24f79 zebra: add "json" option to "show interface" 
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
 2021-08-02 17:19:45 -03:00
Mark Stapp
bc86b347db zebra: use strlcpy in dplane_rule_init 
Use strlcpy for safety in dplane rule init api.

Signed-off-by: Mark Stapp <mjs.ietf@gmail.com>
 2021-08-02 12:35:50 -04:00
Igor Ryzhov
1f74d96c41 zebra: remove checks for src address existence when using "set src" 
1. This check is absolutely useless. Nothing keeps user from deleting
   the address right after this check.
2. This check prevents zebra from correctly reading the user config with
   "set src" because of a race with interface startup (see #4249).
3. NO OPERATIONAL DATA USAGE ON VALIDATION STAGE.

Fixes #7319.

Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
 2021-08-02 18:35:30 +03:00
Igor Ryzhov
72928fa1aa
Merge pull request #9238 from leonshaw/fix/netns-delete 
lib, zebra: Preserve user-configured VRF on netns deletion
 2021-08-02 18:12:19 +03:00
Xiao Liang
6910315f6f lib, zebra: Preserve user-configured VRF on netns deletion 
Don't clear VRF's user-configured flag when netns is deleted.

Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
 2021-07-30 14:53:45 +08:00
Anuradha Karuppiah
82732723da zebra: use a separate dummy prefix for referencing v6 nexthops 
v4 and v6 host/refernce prefixes need to be setup separately for
[RMAC, VTEP] entries as the VTEP is always normalized to a v4 addr.

Signed-off-by: Anuradha Karuppiah <anuradhak@nvidia.com>
 2021-07-29 17:25:11 -07:00
Igor Ryzhov
9da01b0b7b *: cleanup interface node installation 
The only difference in daemons' interface node definition is the config
write function. No need to define the node in every daemon, just pass
the callback as an argument to a library function and define the node
there.

Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
 2021-07-29 21:35:25 +03:00
batmancn
5306e6cf00 zebra: bugfix of error quit of zebra, due to no nexthop ACTIVE 
There exists some rare situations where fpm will attempt
to send a route update with no valid nexthops.  In that
case an assert would be hit.  This is not good for
trying to keep your routing daemons up and running
when we can safely just recover the situation.

Fixes #7588
Signed-off-by: batmancn <batmanustc@gmail.com>
<fixed commit message, and used zlog_err>
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-07-28 16:13:59 -04:00
Jafar Al-Gharaibeh
213d980ff9
Merge pull request #9007 from donaldsharp/pbr_stuff 
add ability to match on proto to pbr
 2021-07-27 15:09:29 -05:00
David Lamparter
631fce38ff
Merge pull request #9107 from donaldsharp/label_destruction 
zebra: On client shutdown cleanup any vrf labels associated with it
 2021-07-27 14:28:13 +02:00
David Lamparter
9c9d8a6129
Merge pull request #9088 from donaldsharp/zebra_redistribute_wrong_tables 
zebra: Do not allow redistribution for non-vrf tables
 2021-07-27 14:14:23 +02:00
Trey Aspelund
fb0b54b361 zebra: Remove MM seq from evpn rmac json output 
Currently 'show evpn rmac vni .. mac .. json' includes fields for
localSequence and remoteSequence, which are misleading since they
aren't applicable to a macs in the IP-VRF mac table (RMAC).
This removes the localSequence + remoteSequence fields from the output.

Signed-off-by: Trey Aspelund <taspelund@nvidia.com>
 2021-07-22 20:23:56 +00:00
Donald Sharp
9fbbcbeb1f
Merge pull request #9091 from gord1306/remove_lst_vlan 
zebra: trigger remove all access vlans info for access port
 2021-07-22 07:04:20 -04:00
Donald Sharp
06302ecb88 zebra: On client shutdown cleanup any vrf labels associated with it 
When a vrf label is created by a client and the client disconnects
we should clean up any vrf labels associated with that client.

eva# show mpls table
 Inbound Label  Type   Nexthop  Outbound Label
 -----------------------------------------------
 1000           SHARP  RED      -

eva# exit
sharpd@eva ~/f/zebra (label_destruction)> ps -ef | grep frr
root     4017793       1  0 13:57 ?        00:00:00 /usr/lib/frr/watchfrr -d -F datacenter --log file:/var/log/frr/watchfrr.log --log-level debug zebra bgpd ospfd isisd pimd eigrpd sharpd staticd
frr      4017824       1  0 13:57 ?        00:00:00 /usr/lib/frr/zebra -d -F datacenter --log file:/tmp/zebra.log -r --graceful_restart 60 -A 127.0.0.1 -s 90000000
frr      4017829       1  0 13:57 ?        00:00:00 /usr/lib/frr/bgpd -d -F datacenter -M rpki -A 127.0.0.1
frr      4017836       1  0 13:57 ?        00:00:00 /usr/lib/frr/ospfd -d -F datacenter -A 127.0.0.1
frr      4017839       1  0 13:57 ?        00:00:00 /usr/lib/frr/isisd -d -F datacenter -A 127.0.0.1
frr      4017842       1  0 13:57 ?        00:00:00 /usr/lib/frr/pimd -d -F datacenter -A 127.0.0.1
frr      4017865       1  0 13:57 ?        00:00:00 /usr/lib/frr/eigrpd -d -F datacenter -A 127.0.0.1
frr      4017869       1  0 13:57 ?        00:00:00 /usr/lib/frr/sharpd -d -F datacenter -A 127.0.0.1
frr      4017888       1  0 13:57 ?        00:00:00 /usr/lib/frr/staticd -d -F datacenter -A 127.0.0.1
sharpd   4018624 3938423  0 14:02 pts/10   00:00:00 grep --color=auto frr
sharpd@eva ~/f/zebra (label_destruction)> sudo kill -9 4017869

sharpd@eva ~/f/zebra (label_destruction)> sudo vtysh -c "show mpls table"
sharpd@eva ~/f/zebra (label_destruction)>

Fixes: #1787
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-07-21 14:04:36 -04:00
David Lamparter
63116a7008 build: fix AM_LDFLAGS usage (and gcov) 
like the other automake variables, setting `xyz_LDFLAGS` causes
`AM_LDFLAGS` to be ignored for `xyz`.  For some reason I had in my mind
that automake doesn't do this for LDFLAGS, but... it does.  (Which is
consistent with `_CFLAGS` and co.)

So, all the libraries and modules have been ignoring `AM_LDFLAGS` (which
includes `SAN_FLAGS` too).  Set up new `LIB_LDFLAGS` and
`MODULE_LDFLAGS` to handle all of this correctly (and move these bits to
a central location.)

Fixes: #9034
Fixes: 0c4285d77e ("build: properly split CFLAGS from AC_CFLAGS")
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
 2021-07-21 17:10:08 +02:00
Donald Sharp
ecff5258a0 zebra: Mark some bsd interface prefixes as SECONDARY 
Notice when a ip address on a bsd interface is considered
an alias, let's mark the connected prefix we generate as
a SECONDARY.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-07-20 10:12:04 -04:00
gord_chen
ec8977510e zebra: trigger remove all access vlans for access port 
When port was removed from last access vlan, the linux kernel
won't send any vlan info in the netlink message, it might affact
the evpn mh not withdraw EAD-EVI routes.

Signed-off-by: Gord Chen <gord_chen@edge-core.com>
 2021-07-20 09:39:45 +00:00
Donald Sharp
79a9ad1450 zebra: Do not allow redistribution for non-vrf tables 
Current code was allowing redistribution of kernel routes from
the non-default non vrf tables once FRR was already up and running.

In the case where we add `redistribute kernel` in an upper level
protocol we never consider the non-default vrf or non-vrf tables
so it is never accepted.

In the case where a kernel route is added after `redistribute kernel`
is already in place we were never looking at the fact that the
route was in a non-default non-vrf table.  This code fixes
that issue.

Fixes: #9073
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
 2021-07-19 20:04:03 -04:00
Mark Stapp
80ff3f05ea zebra: replace ipaddr2str in dplane module 
Replace a couple of ipaddr2str calls with pIA in the dplane
module.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
 2021-07-19 10:36:12 -04:00
Mark Stapp
7e5b0b2b36 zebra: process EVPN remote VTEP updates from the workqueue 
Move remote VTEP updates from immediate, inline processing
in their ZAPI message handlers to the main workqueue.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
 2021-07-19 10:36:12 -04:00
Mark Stapp
7f7e49d11a zebra: use workqueue for vxlan remote macip updates 
Enqueue incoming vxlan remote macip updates on the main
workqueue, instead of performing the updates immediately,
in-line.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
 2021-07-19 10:36:12 -04:00
Mark Stapp
1a3bd37f7c zebra: use more const 
Use const in many more evpn apis, especially for macaddr,
ipaddr arguments.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
 2021-07-19 10:36:12 -04:00
Mark Stapp
32367e7a3b zebra: add workqueue support for EVPN updates 
Add workqueue subqueue for EVPN/VxLAN updates; migrate the
evpn route and remote ES processing from their ZAPI handlers
to the workqueue.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
 2021-07-19 10:36:12 -04:00