Ceph

05/06/2024

08:56 PM RADOS Bug #65749: osd_max_pg_per_osd_hard_ratio 3 is set too low for real life

Radoslaw Zarzynski wrote in #note-4:
> Is there any trace of autoscaler-induced PG splitting visible during the situ... Dan van der Ster

05/03/2024

07:51 PM RADOS Bug #55750: mon: slow request of very long time

Still getting this on reef 18.2.1 Dan van der Ster

05/01/2024

10:12 PM RADOS Bug #65749 (Fix Under Review): osd_max_pg_per_osd_hard_ratio 3 is set too low for real life

Dan van der Ster

10:00 PM RADOS Bug #65749 (In Progress): osd_max_pg_per_osd_hard_ratio 3 is set too low for real life

Dan van der Ster

09:38 PM mgr Bug #65748 (Fix Under Review): Change default upmap_max_deviation to 1

Dan van der Ster

04/30/2024

09:36 PM Ceph Bug #55461 (Fix Under Review): ceph osd crush swap-bucket {old_host} {new_host} where {old_host}={new_host} crashes monitors

Dan van der Ster

03/07/2024

06:40 PM CephFS Support #57952: Pacific: the buffer_anon_bytes of ceph-mds is too large

There was a similar case back in nautilus:
* https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/TNXNORN... Dan van der Ster

02/22/2024

05:21 PM RADOS Feature #56956: osdc: Add objecter fastfail

> There is no point in indefinitely waiting when pg of an object is inactive.
This is not correct for CephFS or RB... Dan van der Ster

02/21/2024

07:32 PM Ceph Bug #64323: PG recovery stuck without making progress

We've seen this in a few clusters and reverted to osd_op_queue = wpq to workaround the issue in mclock scheduler. Dan van der Ster

02/12/2024

09:50 PM RADOS Bug #63389: Failed to encode map X with expected CRC

@Radek I'm confused, are we going with PR 54890 as a fix for this, or your proposed actual fix? Dan van der Ster