Back to bug 2180542

Who When What Removed Added
Red Hat One Jira (issues.redhat.com) 2023-03-21 17:42:49 UTC Link ID Red Hat Issue Tracker OSP-23264
Goutham Pacha Ravi 2023-03-21 17:47:05 UTC CC gouthamr
Chuck Copello 2023-03-21 19:16:42 UTC CC ccopello
ndeevy 2023-03-22 09:09:39 UTC Docs Contact ndeevy
Goutham Pacha Ravi 2023-03-23 22:50:13 UTC CC vhariria
QA Contact vhariria lkuchlan
Flags needinfo?(cjanisze)
Goutham Pacha Ravi 2023-03-23 22:59:03 UTC Flags needinfo?(lkuchlan)
CC lkuchlan
Chris Janiszewski 2023-03-24 13:00:34 UTC Flags needinfo?(cjanisze) needinfo?(lkuchlan)
Chris Janiszewski 2023-03-24 13:01:52 UTC Flags needinfo?(lkuchlan)
lkuchlan 2023-03-26 07:23:16 UTC Flags needinfo?(lkuchlan)
Francesco Pantano 2023-03-29 15:43:05 UTC Flags needinfo?(cjanisze)
CC fpantano
Chris Janiszewski 2023-04-12 13:10:37 UTC Flags needinfo?(cjanisze)
Manoj Katari 2023-04-19 11:07:03 UTC CC mkatari
Ashley R 2023-05-09 12:52:19 UTC Severity unspecified medium
Priority unspecified low
CC ashrodri
Keywords Triaged
Francesco Pantano 2023-06-06 08:01:20 UTC Flags needinfo?(gouthamr) needinfo?(alfrgarc)
CC alfrgarc
Goutham Pacha Ravi 2023-06-06 18:43:39 UTC Flags needinfo?(gouthamr)
Alfredo 2023-06-12 15:16:02 UTC Flags needinfo?(alfrgarc)
Goutham Pacha Ravi 2023-06-29 20:24:57 UTC Link ID OpenStack gerrit 887340
Goutham Pacha Ravi 2023-06-29 20:27:18 UTC Assignee openstack-manila-bugs gouthamr
Component openstack-manila tripleo-ansible
Status NEW ON_DEV
QA Contact lkuchlan jhakimra
Ian Frangs 2023-08-03 15:46:23 UTC Flags needinfo?(gouthamr)
Goutham Pacha Ravi 2023-08-04 00:34:11 UTC Target Release --- 17.1
Doc Type If docs needed, set a value Bug Fix
Doc Text Cause: The pacemaker controlled "cephfs-nfs" resource requires a run time directory to store some process data. While this directory was created fine on install or upgrade, a reboot of the controller nodes would remove the directory.

Consequence: The "cephfs-nfs" service would not recover after controller nodes are rebooted. If all controller nodes are rebooted, the service fails permanently

Fix: The directory is now created prior to spawning the "cephfs-nfs" service.

Result: "cephfs-nfs" service does not fail due to a missing runtime directory.
Target Milestone --- z2
Flags needinfo?(gouthamr)
Status ON_DEV POST
Jenny-Anne Lynch 2023-08-04 16:07:50 UTC CC jelynch
Flags needinfo?(gouthamr)
Doc Text Cause: The pacemaker controlled "cephfs-nfs" resource requires a run time directory to store some process data. While this directory was created fine on install or upgrade, a reboot of the controller nodes would remove the directory.

Consequence: The "cephfs-nfs" service would not recover after controller nodes are rebooted. If all controller nodes are rebooted, the service fails permanently

Fix: The directory is now created prior to spawning the "cephfs-nfs" service.

Result: "cephfs-nfs" service does not fail due to a missing runtime directory.
Before this update, the Pacemaker-controlled `ceph-nfs` resource required a runtime directory to store some process data. The directory was created when you installed or upgraded RHOSP. However, a reboot of the Controller nodes removed the directory, and the `ceph-nfs` service did not recover when the Controller nodes were rebooted. If all Controller nodes were rebooted, the `ceph-nfs` service failed permanently. With this update, the directory is created before spawning the `ceph-nfs` service, and the `cephfs-nfs` service does not fail if the runtime directory is missing.
Goutham Pacha Ravi 2023-08-04 20:44:32 UTC Flags needinfo?(gouthamr)
Jenny-Anne Lynch 2023-08-09 10:08:32 UTC Doc Text Before this update, the Pacemaker-controlled `ceph-nfs` resource required a runtime directory to store some process data. The directory was created when you installed or upgraded RHOSP. However, a reboot of the Controller nodes removed the directory, and the `ceph-nfs` service did not recover when the Controller nodes were rebooted. If all Controller nodes were rebooted, the `ceph-nfs` service failed permanently. With this update, the directory is created before spawning the `ceph-nfs` service, and the `cephfs-nfs` service does not fail if the runtime directory is missing. The Pacemaker-controlled `ceph-nfs` resource requires a runtime directory to store some process data. Before this update, the directory was created when you installed or upgraded RHOSP. However, a reboot of the Controller nodes removed the directory, and the `ceph-nfs` service did not recover when the Controller nodes were rebooted. If all Controller nodes were rebooted, the `ceph-nfs` service failed permanently. With this update, the directory is created before spawning the `ceph-nfs` service, and the `cephfs-nfs` service continues through reboots.
James Smith 2023-08-14 14:11:11 UTC CC jamsmith
Flags needinfo?(gouthamr)
Doc Text The Pacemaker-controlled `ceph-nfs` resource requires a runtime directory to store some process data. Before this update, the directory was created when you installed or upgraded RHOSP. However, a reboot of the Controller nodes removed the directory, and the `ceph-nfs` service did not recover when the Controller nodes were rebooted. If all Controller nodes were rebooted, the `ceph-nfs` service failed permanently. With this update, the directory is created before spawning the `ceph-nfs` service, and the `cephfs-nfs` service continues through reboots. The Pacemaker-controlled `ceph-nfs` resource requires a runtime directory to store some process data. The directory is created when you install or upgrade RHOSP. However, a reboot of the Controller nodes removes the directory, and the `ceph-nfs` service does not recover when the Controller nodes are rebooted. If all Controller nodes are rebooted, the `ceph-nfs` service fails permanently.
Doc Type Bug Fix Known Issue
Goutham Pacha Ravi 2023-08-14 22:13:41 UTC Flags needinfo?(gouthamr)
Jenny-Anne Lynch 2023-08-16 14:56:40 UTC CC kgilliga
Flags needinfo?(gouthamr)
Jenny-Anne Lynch 2023-08-16 14:57:05 UTC Docs Contact jelynch
Jenny-Anne Lynch 2023-08-16 15:10:39 UTC Doc Text The Pacemaker-controlled `ceph-nfs` resource requires a runtime directory to store some process data. The directory is created when you install or upgrade RHOSP. However, a reboot of the Controller nodes removes the directory, and the `ceph-nfs` service does not recover when the Controller nodes are rebooted. If all Controller nodes are rebooted, the `ceph-nfs` service fails permanently. The Pacemaker-controlled `ceph-nfs` resource requires a runtime directory to store some process data. The directory is created when you install or upgrade RHOSP. Currently, a reboot of the Controller nodes removes the directory, and the `ceph-nfs` service does not recover when the Controller nodes are rebooted. If all Controller nodes are rebooted, the `ceph-nfs` service fails permanently.
+
Workaround: If you reboot a Controller node, log into the Controller node and create a `/var/run/ceph` directory:
+
`$ mkdir -p /var/run/ceph`
+
Repeat this step on all Controller nodes that have been rebooted. If the `ceph-nfs-pacemaker` service has been
marked as failed, after creating the directory, execute the following command from any of the Controller nodes:
+
`$ pcs resource cleanup`
Goutham Pacha Ravi 2023-08-16 15:52:01 UTC Flags needinfo?(gouthamr)
Jenny-Anne Lynch 2023-08-16 16:20:27 UTC Summary after osp17.0.0 -> osp17.0.1 upgrade manila ceph-nfs failed to start. Error: ceph-nfs start on leaf1-controller-0 returned 'error' because 'failed' After upgrade, manila ceph-nfs fails to start with error: ceph-nfs start on leaf1-controller-0 returned 'error' because 'failed'

Back to bug 2180542