Hi Joffrey, That's good to know. Please note that you can switch back to "high_client_ops" profile once all the recoveries are completed. This is to ensure clients operations get higher priority when there aren't many recoveries or no recoveries going on. -Sridhar On Tue, Mar 7, 2023 at 2:51 PM Joffrey <joff.au@gmail.com> wrote:
Hi, I changed de mclock priority last Friday with "ceph tell 'osd.*' injectargs '--osd_mclock_profile=high_recovery_ops'" and now, the Health is OK. So, you(re right, I need to change the mclock scheduler to changes recovery priority.
Thanks you
Le sam. 4 mars 2023 à 08:12, Sridhar Seshasayee <sseshasa@redhat.com> a écrit :
Hello Joffrey, You could be hitting the slow backfill/recovery issue with mclock_scheduler. To confirm the above could you please provide the output of the following commands?
1. ceph versions 2. ceph config show osd.<id> | grep osd_max_backfills 3. ceph config show osd.<id> | grep osd_recovery_max_active 4. ceph config show osd.<id> | grep osd_mclock
where 'id' can be any valid osd id.
With the mclock_scheduler enabled and with 17.2.5, it is not possible to override recovery settings like 'osd_max_backfills' and other recovery related config options.
-Sridhar _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
-- Sridhar Seshasayee Partner Engineer Red Hat <https://www.redhat.com> <https://www.redhat.com>