Ceph

02/02/2024

09:01 AM CephFS Bug #62501: pacific(?): qa: mgr-osd-full causes OSD aborts due to ENOSPC (fs/full/subvolume_snapshot_rm.sh)

The bug is caused by difference is behaviour of BlueFS.
This PR is the final trigger: https://github.com/ceph/ceph/p... Adam Kupczyk

12/19/2023

01:42 PM bluestore Bug #63769 (Pending Backport): ObjectStore/StoreTestSpecificAUSize.SyntheticMatrixSharding/1 is failing when using bitmap allocator

Adam Kupczyk

12/14/2023

10:13 AM bluestore Bug #63769: ObjectStore/StoreTestSpecificAUSize.SyntheticMatrixSharding/1 is failing when using bitmap allocator

Problem is replicated by:
./bin/ceph_test_objectstore --gtest_filter=\*SyntheticMatrixSharding/1 --bluestore_allocat... Adam Kupczyk

12/13/2023

11:26 AM bluestore Bug #63769: ObjectStore/StoreTestSpecificAUSize.SyntheticMatrixSharding/1 is failing when using bitmap allocator

Replicated, verified. Adam Kupczyk

11/28/2023

04:43 PM bluestore Bug #63618: Allocator configured with 64K alloc unit might get 4K requests

PR https://github.com/ceph/ceph/pull/48854 "os/bluestore: enable 4K allocation unit for BlueFS"
was created with cas... Adam Kupczyk

11/03/2023

07:36 PM bluestore Bug #63436 (Pending Backport): Typo in reshard example

See https://tracker.ceph.com/issues/63353.
I missed the fact that "o"->"O" should be done too. Adam Kupczyk

03:33 PM bluestore Bug #63353: resharding RocksDB after upgrade to Pacific breaks OSDs

It is a bug. Same prefix can be inserted twice and it confuses the tool.
It should be fixed.
Please try to resh... Adam Kupczyk

09/06/2023

07:49 PM bluestore Bug #62730 (New): ceph-bluestore-tool reshard broken

It is possible to specify same prefix twice. Here an example with "p" defined twice.
ceph-bluestore-tool --path /v... Adam Kupczyk

08/31/2023

03:18 PM bluestore Bug #61953 (Won't Fix): bluestore: bluestore slow latency report when rocksdb compaction

Its normal behaviour for BlueStore to be a bit slower when compacting. Adam Kupczyk

02:42 PM bluestore Bug #61949 (Fix Under Review): btree allocator doesn't pass alloctor's UTs

Adam Kupczyk