thanks Boris, there was some previous discussion in related issue https://tracker.ceph.com/issues/74231, and i've proposed a fix for that On Tue, Feb 3, 2026 at 6:07 AM Boris via ceph-users <ceph-users@ceph.io> wrote:
created a tracker with additional information: https://tracker.ceph.com/issues/74729
Am Di., 3. Feb. 2026 um 01:51 Uhr schrieb Boris <bb@kervyn.de>:
Hi,
we just encountered a strange problem:
When we run radosgw-admin bucket stats on the master zonegroup we get some of these errors error getting bucket stats bucket=rregergf ret=-2
This is the debug_rgw=20 output: error getting bucket stats bucket=rregergf ret=2026-02-03T00:50:32.347+0000 722007981a80 20 cls_bucket_head: CLSRGWIssueGetDirHeader() returned -2 -2 failure: (2) No such file or directory:
But the bucketname is listed in a bucket list command.
After some digging we found that the bucket was created after the upgrade to squid in a non master zonegroup.
Doing the radosgw-admin bucket stats command in any zonegroup, except the master, yields results.
We checked, and it is every bucket created after the upgrade, that was not created in the master zonegroup.
We only sync metadata between the zonegrouops, and I just ran a fresh metadata sync on all zonegroups.
Any ideas what this could be?
-- Die Selbsthilfegruppe "UTF-8-Probleme" trifft sich diesmal abweichend im groüen Saal. _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io