Bug #65130
opencrimson: crimson-rados did not detect reintroduction of https://tracker.ceph.com/issues/61875
0%
Description
https://github.com/ceph/ceph/pull/56376 would have reintroduced https://tracker.ceph.com/issues/61875 as it puts the snap mapper keys back into the pg meta object. Oddly, a teuthology run on that branch which seems to have included tests with both snapshots and osd restarts did not show crashes associated with this regression and at least one case that seems like it should have exercised the relevant code passed. A quick glance over PGLog.cc::FuturizedShardStoreReader doesn't show any changes, so it should have crashed in the final else branch of FuturizedShardStoreLogReader::process_entry at e.decode_with_checksum.
Tasks:
- Confirm that the crimson-rados suite actually combines snapshots with OSD restarts
- Work out why the existing suite didn't fail the above PR
- Amend the tests to cover the gap
Updated by Samuel Just about 1 month ago
Note that the defect addressed by the above PR is going to be handled another way -- this bug is about why the test suite didn't catch this specific regression.
Updated by Matan Breizman about 1 month ago
- Status changed from New to Fix Under Review
- Assignee set to Matan Breizman
- Pull request ID set to 56511
Updated by Matan Breizman about 1 month ago
- Related to Bug #65247: ObjectContext::drop_recovery_read(): Assertion `recovery_read_marker' failed. added
Updated by Matan Breizman 18 days ago
Added label: crimson-replicated-recovery to track all the required fixes
https://github.com/ceph/ceph/pulls?q=+is%3Apr+label%3Acrimson-replicated-recovery