Merge lp:~ballot/charms/xenial/daisy-retracer/lp1845257 into lp:~daisy-pluckers/charms/xenial/daisy-retracer/trunk

Proposed by Benjamin Allot
Status: Rejected
Rejected by: Brian Murray
Proposed branch: lp:~ballot/charms/xenial/daisy-retracer/lp1845257
Merge into: lp:~daisy-pluckers/charms/xenial/daisy-retracer/trunk
Diff against target: 18 lines (+2/-1)
1 file modified
hooks/common (+2/-1)
To merge this branch: bzr merge lp:~ballot/charms/xenial/daisy-retracer/lp1845257
Reviewer Review Type Date Requested Status
Brian Murray Disapprove
Junien F Approve
Review via email: mp+382547@code.launchpad.net
To post a comment you must log in.
206. By Benjamin Allot

Add RestartSec directive and increase StartLimitBurst

The rationale behind this change is that the default RestartSec is 100ms.
Upon error of the service (because of a crash or whatever), it attempts to restart
very quickly 5 times, then stay in a failed state until manually restarted.

As seen in systemd.unit manual for the StartLimitBurst setting:

  Note that units which are configured for Restart= and which reach the start limit
  are not attempted to be restarted anymore;

The new limit try one restart every 10 seconds and the StartLimitInterval is set to
StartLimitBurst * RestartSec to avoid reaching this state.

Revision history for this message
Junien F (axino) wrote :

+1

review: Approve
Revision history for this message
Brian Murray (brian-murray) wrote :

In bug 1845257 Dimitri has asked for the unit log files so we can try to figure out why exactly the units are failing to restart. Could those please be provided to the bug report? Thanks in advance.

review: Needs Information
Revision history for this message
Brian Murray (brian-murray) wrote :

I'm rejecting this in favor of another change I made to the retracing service.

review: Disapprove

Unmerged revisions

206. By Benjamin Allot

Add RestartSec directive and increase StartLimitBurst

The rationale behind this change is that the default RestartSec is 100ms.
Upon error of the service (because of a crash or whatever), it attempts to restart
very quickly 5 times, then stay in a failed state until manually restarted.

As seen in systemd.unit manual for the StartLimitBurst setting:

  Note that units which are configured for Restart= and which reach the start limit
  are not attempted to be restarted anymore;

The new limit try one restart every 10 seconds and the StartLimitInterval is set to
StartLimitBurst * RestartSec to avoid reaching this state.

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
1=== modified file 'hooks/common'
2--- hooks/common 2019-06-12 16:13:09 +0000
3+++ hooks/common 2020-04-19 15:31:17 +0000
4@@ -189,12 +189,13 @@
5 [Unit]
6 Description="Error Tracker retracer for $arch"
7 StartLimitInterval=60
8-StartLimitBurst=5
9+StartLimitBurst=6
10
11 [Service]
12 Environment="PYTHONPATH=$LOCAL_CONFIG_LOCATION:$CODE_LOCATION:$APPORT_LOCATION:$DEPLOY_ROOT" "PATH=$APPORT_LOCATION/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
13 ExecStart=$CODE_LOCATION/daisy/retracer.py --config-dir $CODE_LOCATION/retracer/config $SANDBOX --output $LOG_DIR/retracer-$arch.log --architecture $arch --core-storage $VAR_DIR $CACHE_DEBS $CLEANUP_SANDBOX $CLEANUP_DEBS $FAILED_QUEUE $STACKTRACE_SOURCE $RETRACER_VERBOSE
14 Restart=always
15+RestartSec=10
16 TimeoutStopSec=180
17
18 [Install]

Subscribers

People subscribed via source and target branches

to all changes: