qa: not starting ceph-mds daemons
These runs both "Timed out waiting for MDS daemons to become healthy". Grepping through the teuthology logs, it looks like they never invoked ceph-mds. How does that happen?!?!?
#1 Updated by Zheng Yan almost 4 years ago
- Status changed from New to Verified
the bug is introduced by commit 685d76a77ca16ca601a99148ef507cfde1fb3593 "ceph: wait for CephFS to be healthy before proceeding. that commit adds a wait_for_daemons() to the funtion that setups MDS (the function does not start MDS)
#2 Updated by Greg Farnum almost 4 years ago
- Status changed from Verified to Need More Info
That commit is now about 16 hours old, and these jobs were started about 24 hours ago — although they didn't get machines locked until after that commit...is that how the qa-suite works now? I thought it took what existed when it was scheduled.