proxmox-mirrors/pve-docs
1
0
Fork 0
mirror of https://git.proxmox.com/git/pve-docs synced 2025-04-28 04:35:36 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

pveceph: destroy osd: rework to make it easier to read

Browse Source 
* rephrased some parts
* split large paragraphs
* put CLI commands into own code blocks instead of being inline
* added step to wait until rebalance after OUT is finished

Signed-off-by: Aaron Lauterer <a.lauterer@proxmox.com>
This commit is contained in:
Aaron Lauterer 2025-03-24 17:34:58 +01:00
parent c49d3f328c
commit 0c2dcfce35
1 changed files with 49 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

71
pveceph.adoc
View File

@ -506,36 +506,63 @@ If you experience problems with an OSD or its disk, try to
xref:pve_ceph_mon_and_ts[troubleshoot] them first to decide if a
xref:pve_ceph_osd_replace[replacement] is needed.
To destroy an OSD:
To destroy an OSD, navigate to the __<Node> -> Ceph -> OSD__ panel or use the
mentioned CLI commands on the node where the OSD is located.
. Either open the web interface and select any {pve} node in the tree
view, or open a shell on the node where the OSD to be deleted is
located.
. Make sure the cluster has enough space to handle the removal of the OSD.
In the __Ceph -> OSD__ panel,if the to-be destroyed OSD is still `up` and `in`
(non-zero value at `AVAIL`), make sure that all OSDs have their `Used (%)`
value well below the `nearfull_ratio` of default `85%`.
+
This way you can reduce the risk from the upcoming rebalancing, which may cause
OSDs to run full and thereby blocking I/O on Ceph pools.
+
Use the following command to get the same information on the CLI:
+
[source,bash]
----
ceph osd df tree
----
. Go to the __Ceph -> OSD__ panel (`ceph osd df tree`). If the OSD to
be deleted is still `up` and `in` (non-zero value at `AVAIL`), make
sure that all OSDs have their `Used (%)` value well below the
`nearfull_ratio` of default `85%`. In this way you can reduce the risk
from the upcoming rebalancing, which may cause OSDs to run full and
thereby blocking I/O on Ceph pools.
. If the to-be destroyed OSD is not `out` yet, select the OSD and click on
**Out**. This will exclude it from data distribution and start a rebalance.
+
The following command does the same:
+
[source,bash]
----
ceph osd out <id>
----
. If the deletable OSD is not `out` yet, select the OSD and click on
**Out** (`ceph osd out <id>`). This will exclude it from data
distribution and starts a rebalance.
. If you can, wait until Ceph has finished the rebalance to always have enough
replicas. The OSD will be empty; once it is, it will show __0 PGs__.
. Click on **Stop**, if stopping is not safe yet, a warning will
appear and you should click on **Cancel**, please try again shortly
afterwards. When using the shell, check if it is safe to stop by
reading the output from `ceph osd ok-to-stop <id>`, once true, run
`pveceph stop --service osd.<id>` .
. Click on **Stop**. If stopping is not safe yet, a warning will
appear, and you should click on **Cancel**. Try it again in a few moments.
+
The following commands can be used to check if it is safe to stop and stop
the OSD:
+
[source,bash]
----
ceph osd ok-to-stop <id>
pveceph stop --service osd.<id>
----
. Finally:
+
[WARNING]
To remove the OSD from Ceph and delete all disk data, first click on
**More -> Destroy**. Use the cleanup option to clean up the partition
table and similar, enabling an immediate reuse of the disk in {pve}.
Finally, click on **Remove** (`pveceph osd destroy <id> [--cleanup]`).
**More -> Destroy**. Enable the cleanup option to clean up the partition table
and other structures. This makes it possible to immediately reuse the disk in
{pve}.
Then, click on **Remove**.
+
The CLI command to destroy the OSD is:
+
[source,bash]
----
pveceph osd destroy <id> [--cleanup]
----
[[pve_ceph_pools]]