[Doc] correct ceph migration steps

update/correct few ceph rbd migration steps
openstack-k8s-operators · Dec 12, 2024 · 43e9c56 · 43e9c56
1 parent f3818fe
commit 43e9c56
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 14 deletions.
diff --git a/docs_user/modules/proc_migrating-mgr-from-controller-nodes.adoc b/docs_user/modules/proc_migrating-mgr-from-controller-nodes.adoc
@@ -53,10 +53,10 @@ $ sudo nft list ruleset | grep ceph_mgr
 label to the target node:
 +
 ----
-$ ceph orch host label add <target_node> mgr; done
+$ sudo cephadm shell -- ceph orch host label add <target_node> mgr
 ----
 
-. Repeat steps 1-3 for each target node that hosts a Ceph Manager daemon.
+. Repeat steps 1-7 for each target node that hosts a Ceph Manager daemon.
 
 . Get the Ceph Manager spec:
 +

diff --git a/docs_user/modules/proc_migrating-mon-from-controller-nodes-drain-host.adoc b/docs_user/modules/proc_migrating-mon-from-controller-nodes-drain-host.adoc
@@ -2,7 +2,7 @@
 
 = Draining the source node
 
-Drain the existing Controller nodes and remove the source node host from the {CephCluster} cluster.
+Drain the source node and remove the source node host from the {CephCluster} cluster.
 
 .Procedure
 
@@ -19,7 +19,7 @@ $ sudo cp -R /etc/ceph $HOME/ceph_client_backup
 $ sudo cephadm shell -- ceph mgr stat
 ----
 
-. Fail the `ceph-mgr` if it is active on the source node or target node:
+. Fail the `ceph-mgr` if it is active on the source node:
 +
 ----
 $ sudo cephadm shell -- ceph mgr fail <mgr_instance>
@@ -31,28 +31,28 @@ $ sudo cephadm shell -- ceph mgr fail <mgr_instance>
 +
 ----
 $ for label in mon mgr _admin; do
-    sudo cephadm shell -- ceph orch host rm label <source_node> $label;
+    sudo cephadm shell -- ceph orch host label rm <source_node> $label;
 done
 ----
 +
 * Replace `<source_node>` with the hostname of the source node.
 
-. Remove the running Ceph Monitor daemon from the source node:
+. (Optional) Ensure that you remove the Ceph Monitor daemon from the source node if it is still running:
 +
 ----
-$ sudo cephadm shell -- ceph orch daemon rm mon.<source_node> --force"
+$ sudo cephadm shell -- ceph orch daemon rm mon.<source_node> --force
 ----
 
-. Drain the source node:
+. Drain the source node to remove any leftover daemons:
 +
 ----
-$ sudo cephadm shell -- ceph drain <source_node>
+$ sudo cephadm shell -- ceph orch host drain <source_node>
 ----
 
 . Remove the source node host from the {CephCluster} cluster:
 +
 ----
-$ sudo cephadm shell -- ceph orch host rm <source_node> --force"
+$ sudo cephadm shell -- ceph orch host rm <source_node> --force
 ----
 +
 [NOTE]

diff --git a/docs_user/modules/proc_migrating-mon-from-controller-nodes-network.adoc b/docs_user/modules/proc_migrating-mon-from-controller-nodes-network.adoc
@@ -11,12 +11,19 @@ IP address migration assumes that the target nodes are originally deployed by
 // w/ an EDPM node that has already been adopted.
 .Procedure
 
-. Get the original Ceph Monitor IP address from the existing `/etc/ceph/ceph.conf` file on the `mon_host` line, for example:
+. Get the original Ceph Monitor IP addresses from `$HOME/ceph_client_backup/ceph.conf` file on the `mon_host` line, for example:
 +
 ----
 mon_host = [v2:172.17.3.60:3300/0,v1:172.17.3.60:6789/0] [v2:172.17.3.29:3300/0,v1:172.17.3.29:6789/0] [v2:172.17.3.53:3300/0,v1:172.17.3.53:6789/0]
 ----
 
+. Match the IP address retrieved in the previous step with the storage network IP addresses on the source node, and find the Ceph Monitor IP address:
+----
+[tripleo-admin@controller-0 ~]$ ip -o -4 a | grep 172.17.3
+9: vlan30    inet 172.17.3.60/24 brd 172.17.3.255 scope global vlan30\       valid_lft forever preferred_lft forever
+9: vlan30    inet 172.17.3.13/32 brd 172.17.3.255 scope global vlan30\       valid_lft forever preferred_lft forever
+----
+
 . Confirm that the Ceph Monitor IP address is present in the `os-net-config` configuration that is located in the `/etc/os-net-config` directory on the source node:
 +
 ----

diff --git a/docs_user/modules/proc_migrating-mon-from-controller-nodes-redeploy-mon.adoc b/docs_user/modules/proc_migrating-mon-from-controller-nodes-redeploy-mon.adoc
@@ -39,7 +39,7 @@ The Ceph Monitor daemons are marked as `unmanaged`, and you can now redeploy the
 . Delete the existing Ceph Monitor on the target node:
 +
 ----
-$ sudo cephadm shell -- ceph orch daemon add rm mon.<target_node> --force
+$ sudo cephadm shell -- ceph orch daemon rm mon.<target_node> --force
 ----
 +
 * Replace `<target_node>` with the hostname of the target node that is included in the {Ceph} cluster.
@@ -84,7 +84,7 @@ The new Ceph Monitor runs on the target node with the original IP address.
 . Identify the running `mgr`:
 +
 ----
-$ sudo cephadm shell --  mgr stat
+$ sudo cephadm shell -- ceph mgr stat
 ----
 +
 . Refresh the Ceph Manager information by force-failing it:
@@ -101,5 +101,5 @@ $ sudo cephadm shell -- ceph orch reconfig osd.default_drive_group
 
 .Next steps
 
-Repeat the procedure for each node that you want to decommission.
+Repeat the procedure starting from step xref:draining-the-source-node_{context}[Draining the source node] for each node that you want to decommission.
 Proceed to the next step xref:verifying-the-cluster-after-ceph-mon-migration_{context}[Verifying the {CephCluster} cluster after Ceph Monitor migration].