Hi!
I have a Ceph cluster version 16.2.7 built with cephadm. Ceph logs in
/var/log/ceph/ all rotate nicely, but container logs in
/var/lib/docker/containers/ keep growing. For example, a log of a monitor
container has reached almost 3 GB since December 2021:
09:18 [root@ceph02
/var/lib/docker/containers/04807066d68dab96c506e03c2a1e2d3d42bde6677353db5728c5a2ed637579fe]#
ls -al *log
-rw-r----- 1 root root 3011531294 Jan 27 09:18
04807066d68dab96c506e03c2a1e2d3d42bde6677353db5728c5a2ed637579fe-json.log
The log is full of debug info, such as
{"log":"cluster 2022-01-27T09:13:47.026820+0000 mgr.ceph01.vankui
(mgr.55010800) 2575830 : cluster [DBG] pgmap v2156823: 1824 pgs: 1824
active+clean; 4.4 TiB data, 28 TiB used, 549 TiB / 577 TiB avail; 2.6 MiB/s
rd, 7.0 MiB/s wr, 915
op/s\n","stream":"stderr","time":"2022-01-27T09:13:48.274941081Z"}
{"log":"audit 2022-01-27T09:13:58.659145+0000 mon.ceph04 (mon.3) 427114 :
audit [DBG] from='client.? a.b.c.d:0/1284725261' entity='client.cinder'
cmd=[{\"prefix\":\"osd pool get-quota\", \"pool\": \"volumes\",
\"format\":\"json\"}]:
dispatch\n","stream":"stderr","time":"2022-01-27T09:13:59.26827113Z"}
Other containers, including managers and OSDs, also appear to log excessive
debug information. What is the best way to reduce logging of all containers
to reasonable levels and trim the existing container logs?
I would very much appreciate any advice.
Best regards,
Zakhar
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
