Hi all. About 2 days my Ceph cluster goes 1 millions IO/s on reads default.rgw.buckets.index. When this happens my PUT requests goes up to 200 req/s but I had 500 req/s before and there were no high IO/s on that pool. When this happens my rgw nodes and OSDs goes up to 100% cpu usage.
Do you have any indea what’s going on here that this pool gets 1 millions IO/s?
Also I have upgraded to 14.2.8 but problem still persists.
Thanks for you help :)
_______________________________________________ Dev mailing list -- dev@xxxxxxx To unsubscribe send an email to dev-leave@xxxxxxx
