[Bug 2127242] [NEW] Existing mds restarted unexpectedly when adding new ceph-fs unit

[dongdong tao]

Public bug reported:

Steps to reproduce:
1. juju deploy a fresh squid ceph cluster

2. juju add-unit ceph-fs

3. Observe the existing mds got restarted unexpectedly.


The time required for mds failover process to complete can take several minutes, especially for large scaled cephfs cluster. During the failover, clients operation will be completely stalled, which may lead to notable business impact.

We should only avoid this unnecessary mds restart.

** Affects: ceph (Ubuntu)
     Importance: Undecided
         Status: New

-- 
You received this bug notification because you are a member of Ubuntu
OpenStack, which is subscribed to ceph in Ubuntu.
https://bugs.launchpad.net/bugs/2127242

Title:
  Existing mds restarted unexpectedly when adding new ceph-fs unit

Status in ceph package in Ubuntu:
  New

Bug description:
  Steps to reproduce:
  1. juju deploy a fresh squid ceph cluster

  2. juju add-unit ceph-fs

  3. Observe the existing mds got restarted unexpectedly.

  
  The time required for mds failover process to complete can take several minutes, especially for large scaled cephfs cluster. During the failover, clients operation will be completely stalled, which may lead to notable business impact.

  We should only avoid this unnecessary mds restart.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/2127242/+subscriptions