Ceph

03/09/2023

08:14 AM RADOS Bug #6297: ceph osd tell * will break when FD limit reached, messenger should close pipes as nece...

Hi Brad, yes I can. I tried with 1300 and it works fine. I added "ulimit -n 2048" to the script as a work-around.
...

03/07/2023

12:25 PM RADOS Bug #6297: ceph osd tell * will break when FD limit reached, messenger should close pipes as nece...

Just run into this problem as well. I'm scraping OSD perf dumps to a file in a script and I get...

12/16/2022

09:44 AM bluestore Bug #44010: changing osd_memory_target currently requires restart, should update at runtime

Just to add to this from my side, after upgrading to octopus 15.2.17 the memory target can be adjusted at run time.

11/10/2022

01:14 PM Ceph Bug #58002 (New): mon_max_pg_per_osd is not checked per OSD

The warning for exceeding mon_max_pg_per_osd seems to be triggered only when the average PG count over all OSDs excee...

09/20/2022

07:07 AM CephFS Bug #56529: ceph-fs crashes on getfattr

Hi, just a confirmation. The problem is solved in ML 5.19.10-1.el7 and probably all other stable kernel lines, includ...

09/07/2022

12:47 PM RADOS Bug #46847: Loss of placement information on OSD reboot

The PR https://github.com/ceph/ceph/pull/40849 for adding the test was marked stale. I left a comment and it would be...

08:57 AM Ceph Bug #56995: PGs go inactive after failed OSD comes up and is marked as in

The problem is still present in octopus 15.2.17. Almost certainly all newer versions are affected.

09/01/2022

10:36 AM RADOS Bug #49231: MONs unresponsive over extended periods of time

OK, I did some more work and it looks like I can trigger the issue with some certainty by failing an MDS that was up ...

09:35 AM Ceph Bug #56995: PGs go inactive after failed OSD comes up and is marked as in

I cannot reproduce this with mimic-13.2.10.

08/31/2022

12:34 PM Ceph Bug #57348 (New): crush map fails: (1) chose and choseleaf for type OSD not identical and (2) ret...

We observe two issues with crush. Related ceph-user thread (split up into two for some reason): https://lists.ceph.io...