Actions
Bug #57917
opencephadm: duplicate log entry for /var/log/ceph/cephadm.log
% Done:
0%
Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
Configuration of logrotate for /var/lib/ceph/cephadm.log was added (and backported to Pacific) via this PR:
https://github.com/ceph/ceph/pull/41283
However, a global config of for /var/lib/ceph/*.log is already present in the ceph-base RPM.
Using cephadm while also having the ceph-base RPM installed causes logrotate to fail with a duplicate log entry message
node1:~ # logrotate --force /etc/logrotate.d/*
error: /etc/logrotate.d/cephadm:2 duplicate log entry for /var/log/ceph/cephadm.log
node1:~ # cat /etc/logrotate.d/ceph | head -1
/var/log/ceph/*.log {
node1:~ # rpm -qf /etc/logrotate.d/ceph
ceph-base-16.2.9.536+g41a9f9a5573-150300.3.3.1.x86_64
node1:~ # cat /etc/logrotate.d/cephadm | head -2
# created by cephadm
/var/log/ceph/cephadm.log {
Updated by Michael Fritch over 1 year ago
this is also problematic for users upgrading from Octopus based clusters to Pacific.
Updated by Michael Fritch over 1 year ago
As a workaround, I tried a `prerotate` script:
node3:~ # cat /etc/logrotate.d/ceph
/var/log/ceph/*.log {
rotate 7
daily
compress
nosharedscripts
prerotate
bash -c "[[ ! $1 =~ cephadm ]]"
endscript
postrotate
killall -q -1 ceph-mon ceph-mgr ceph-mds ceph-osd ceph-fuse radosgw rbd-mirror cephfs-mirror || pkill -1 -x "ceph-mon|ceph-mgr|ceph-mds|ceph-osd|ceph-fuse|radosgw|rbd-mirror|cephfs-mirror" || true
endscript
missingok
notifempty
su root ceph
}
But, logrotate will fail with an additional error:
node3:~ # logrotate --force /etc/logrotate.d/ error: cephadm:2 duplicate log entry for /var/log/ceph/cephadm.log error: found error in file cephadm, skipping error: error running non-shared prerotate script for /var/log/ceph/cephadm.log of '/var/log/ceph/*.log '
Actions