Ceph

03/13/2020

12:57 PM RADOS Bug #44184: Slow / Hanging Ops after pool creation

Jan Fajerski wrote:
> Can confirm seeing issues with osd map pruning ("oldest_map": 41985 vs "newest_map": 83376) an...

03/05/2020

08:19 PM RADOS Bug #44453: mon: fix/improve mon sync over small keys

Nautilus backport: https://github.com/ceph/ceph/pull/33765

07:32 PM RADOS Bug #42830: problem returning mon to cluster

Workaround in our case is: `ceph config set mon mon_sync_max_payload_size 4096`
We have 5 mons again!

02/27/2020

03:26 PM RADOS Bug #39525: lz4 compressor corrupts data when buffers are unaligned

Erik Lindahl wrote:
> Hi,
>
> Oops; Sorry Dan, but I just realised I misled you. While we do have aggressive LZ4 ...

10:34 AM RADOS Bug #39525: lz4 compressor corrupts data when buffers are unaligned

I got confirmation from Troy and Erik -- both are using lz4 compression like us.
I'm trying to reproduce using uni...

02/25/2020

01:11 AM RADOS Bug #39525: lz4 compressor corrupts data when buffers are unaligned

Might be a stretch, but I just noticed that our bits are flipped nearby the 128k boundary, which is ?coincidentally? ...

02/24/2020

03:50 PM RADOS Bug #42830: problem returning mon to cluster

I noticed there is very little osdmap caching in the leader mon -- here we see only 1 single osdmap in the mempool.
...

05:24 AM RADOS Bug #39525: lz4 compressor corrupts data when buffers are unaligned

More ftr: the corruption occurs in the crush part of the osdmap:...

02/23/2020

12:13 PM RADOS Bug #39525: lz4 compressor corrupts data when buffers are unaligned

A bit more about our incident ftr.
The cluster has 1301 osds in total: 752 filestore and 549 bluestore. The filest...

02/22/2020

01:25 PM RADOS Backport #44206: nautilus: osd segv in ceph::buffer::v14_2_0::ptr::release (PGTempMap::decode)

Started a backport here https://github.com/ceph/ceph/pull/33483