Bug #65770
openqa: failed to be set on mds daemons: {'mds.imported', 'mds.exported'}
0%
Description
This issue has been seen in QA runs for a couple of months but it incorrectly got marked as known issue. https://tracker.ceph.com/issues/54108 was used to mark it known and it can't be that issue since a patch fixing it was merged 2 years ago.
Following are are some instances where this issue occurred -
https://pulpito.ceph.com/pdonnell-2024-04-30_05:04:19-fs-wip-pdonnell-testing-20240429.210911-debug-distro-default-smithi/7680522
https://pulpito.ceph.com/pdonnell-2024-04-30_05:04:19-fs-wip-pdonnell-testing-20240429.210911-debug-distro-default-smithi/7680641
https://pulpito.ceph.com/pdonnell-2024-04-20_23:33:17-fs-wip-pdonnell-testing-20240420.180737-debug-distro-default-smithi/7666020
https://pulpito.ceph.com/pdonnell-2024-04-02_11:52:43-fs-wip-batrick-testing-20240402.004512-distro-default-smithi/7635633
Updated by Venky Shankar 12 days ago
Jos, start by checking if the workload isn't heavy enough to trigger subtree export/import (which would then update the respective perf counters). If that's the case check-counters would trip since it expects there counters to be present in `perf dump`. Also, do the same for other counters that show up in failed runs.
Updated by Venky Shankar 7 days ago
Updated by Jos Collin 5 days ago
Venky Shankar wrote in #note-2:
Jos, start by checking if the workload isn't heavy enough to trigger subtree export/import (which would then update the respective perf counters). If that's the case check-counters would trip since it expects there counters to be present in `perf dump`. Also, do the same for other counters that show up in failed runs.
I'll check that. check_counter.py and l_mds_imported/l_mds_exported are there since long time. So wondering what made the workload suddenly isn't heavy enough?