These are the processes in the iotop in 1 node. I think it's compacting but it is always like this, never finish. 59936 be/4 ceph 0.00 B/s 10.08 M/s 0.00 % 53.07 % ceph-osd -f --cluster ceph --id 46 --setuser ceph --setgroup ceph [bstore_kv_sync] 66097 be/4 ceph 0.00 B/s 6.96 M/s 0.00 % 43.11 % ceph-osd -f --cluster ceph --id 48 --setuser ceph --setgroup ceph [bstore_kv_sync] 63145 be/4 ceph 0.00 B/s 5.82 M/s 0.00 % 40.49 % ceph-osd -f --cluster ceph --id 47 --setuser ceph --setgroup ceph [bstore_kv_sync] 51150 be/4 ceph 0.00 B/s 3.21 M/s 0.00 % 10.50 % ceph-osd -f --cluster ceph --id 43 --setuser ceph --setgroup ceph [bstore_kv_sync] 53909 be/4 ceph 0.00 B/s 2.91 M/s 0.00 % 9.98 % ceph-osd -f --cluster ceph --id 44 --setuser ceph --setgroup ceph [bstore_kv_sync] 57066 be/4 ceph 0.00 B/s 2.18 M/s 0.00 % 8.66 % ceph-osd -f --cluster ceph --id 45 --setuser ceph --setgroup ceph [bstore_kv_sync] 36672 be/4 ceph 0.00 B/s 2.68 M/s 0.00 % 7.82 % ceph-osd -f --cluster ceph --id 42 --setuser ceph --setgroup ceph [bstore_kv_sync] Istvan Szabo Senior Infrastructure Engineer --------------------------------------------------- Agoda Services Co., Ltd. e: istvan.szabo@xxxxxxxxx --------------------------------------------------- -----Original Message----- From: Stefan Kooman <stefan@xxxxxx> Sent: Monday, September 20, 2021 2:13 PM To: Szabo, Istvan (Agoda) <Istvan.Szabo@xxxxxxxxx>; ceph-users <ceph-users@xxxxxxx> Subject: Re: Adding cache tier to an existing objectstore cluster possible? Email received from the internet. If in doubt, don't click any link nor open any attachment ! ________________________________ On 9/20/21 06:15, Szabo, Istvan (Agoda) wrote: > Hi, > > I'm running out of idea why my wal+db nvmes are maxed out always so thinking of I might missed the cache tiering in front of my 4:2 ec-pool. IS it possible to add it later? Maybe I missed a post where you talked about WAL+DB being maxed out. What Ceph version do you use? Maybe you suffer from issue #52244 which is ifxed in Pacific 16.2.6 with PR [1]. Gr. Stefan [1]: https://github.com/ceph/ceph/pull/42773 _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |