Gagan, I am not sure about your question. But if you have (potential) performance issue, have you read https://docs.ceph.com/en/latest/cephfs/multimds/. The Ceph mainstream approach for improving performances, if you have multiple clients/apps, is to run multiple MDS for your file system. Michel Le 30/06/2025 à 17:51, gagan tiwari a écrit :
Thanks Guys for your help. I will go for v19.2.2 Squid.
One thing that I want to know is whether Ceph MDS is still single threaded or multithreaded in the latest v19.2.2 Squid ?
If not, any plan to upgrade it to multithreaded in the new release to take full advantage of multicores server.
Thanks, Gagan
On Mon, Jun 30, 2025 at 12:45 PM Patrick Begou < Patrick.Begou@univ-grenoble-alpes.fr> wrote:
Hi,
just starting with a new (small) cluster running Almalinux9 . I've also deployed 19.2.2 with Cephadm. All went fine. Still not in production, I'm in the benchmarking step.
Patrick
Le 27/06/2025 à 18:15, Michel Jouvin a écrit :
Ho,
If you're starting a new cluster I'd advise to start with 19.2.2 (and deploy the clusster with cephadm). We upgraded our production cluster to this version and had no major issues. Only an issue with deep scrubs scheduling but with an easy workaround, see related thread a month ago...
Best regards,
Michel.
Michel Sent from my mobile Le 27 juin 2025 17:51:20 gagan tiwari <gagan.tiwari@mathisys-india.com> a écrit :
Hi Guys, I am going to set-up a new Ceph Cluster on nodes with OS RHEL8/9. Please let me know which is the most stable version of ceph with all bug fixes that I should go for.
Thanks, Gagan _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io