Bug #3236
logrotate.d script from debian repo does not work
Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Development
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
it uses initctl, which does only work with recentish versions of upstart, which are not in wheezy or sysvinit.
History
#1 Updated by Sage Weil almost 11 years ago
- Status changed from New to Need More Info
the logrotate script is an 'all of the above':
invoke-rc.d ceph reload >/dev/null || service ceph reload >/dev/null
# upstart reload isn't very helpful here:
# https://bugs.launchpad.net/upstart/+bug/1012938
for type in mon osd mds; do
initctl list \
| perl -ne 'print "$+{service} cluster=$+{cluster} id=$+{id}\n" if m{^(?<service>ceph-(mon|osd|mds)+)\s+\((?<cluster>[^/)]+)/(?<id>[^)]+)\) start/}' \
| while read l; do
initctl reload -- $l 2>/dev/null || :
done
done
it reloads via sysvinit and upstart. can you be more specific about what didn't work?
#2 Updated by Tobias Florek almost 11 years ago
- File logrotate.d.ceph.diff View added
i did not read closely. you are, of course, right. the problem is, that on all debian systems, initctl won't exist, so you will get a message from cron about it.
#3 Updated by Sage Weil almost 11 years ago
- Status changed from Need More Info to Fix Under Review
Tobias Florek wrote:
i did not read closely. you are, of course, right. the problem is, that on all debian systems, initctl won't exist, so you will get a message from cron about it.
Looks good. I can add your Signed-off-by I assume?
Thanks!
#4 Updated by Tobias Florek almost 11 years ago
yes.
#5 Updated by Sage Weil almost 11 years ago
- Status changed from Fix Under Review to Resolved