On 10/6/21 00:06, Brian Kim wrote:
Dear ceph-users,
We have a ceph cluster with 3 MDS's and recently had to replay our cache
which is taking an extremely long time to complete. Is there some way to
speed up this process as well as apply some checkpoint so it doesn't have
to start all the way from the beginning?
What do you consider extremely long?
Short of "put all metadata on the fastests disk you have / can find",
not that I know of. And even then it can take a considerable amount of
time. During cache replays (after crashes / failover) I have seen ~ 150K
ceph metadata read IOPS for ~ 5 minutes. Most of them came from OSD
cache. If those IOPS have to come from disk, especially spinners, I can
imagine it would take a huge amount of time.
Gr. Stefan
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
