Low timeout in docker start script
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Snappy |
Invalid
|
Medium
|
Unassigned | ||
snappy-hub |
Fix Committed
|
Undecided
|
Unassigned |
Bug Description
Hi, the docker service is eventually not starting on slow systems (tested on canonistack region 1 and 2, snappy 1504/edge, #133) after a reboot, this is the output of journald:
-- Logs begin at Thu 2015-08-06 11:13:46 UTC, end at Thu 2015-08-06 11:19:48 UTC. --
Aug 06 11:14:13 test systemd[1]: Started The docker app deployment mechanism.
Aug 06 11:14:13 test systemd[1]: Starting The docker app deployment mechanism...
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:14 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
Aug 06 11:14:16 test ubuntu-
When the service starts properly this is the output:
-- Logs begin at Thu 2015-08-06 14:23:45 UTC, end at Thu 2015-08-06 14:24:54 UTC. --
Aug 06 14:24:37 test systemd[1]: Started The docker app deployment mechanism.
Aug 06 14:24:37 test systemd[1]: Starting The docker app deployment mechanism...
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:37 test ubuntu-
Aug 06 14:24:43 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
Aug 06 14:24:44 test ubuntu-
.....
Increasing the timeout value in [1] seems to solve the problem.
Thanks!
Related branches
- Snappy Developers: Pending requested
-
Diff: 12 lines (+1/-1)1 file modifiedpackage-dir/bin/docker.start (+1/-1)
tags: | added: docker |
Changed in snappy: | |
status: | New → Fix Committed |
Changed in snappy-hub: | |
status: | New → Fix Committed |
Changed in snappy: | |
importance: | Undecided → Medium |
Changed in snappy: | |
status: | Fix Committed → Invalid |
The second long in the bug description correspond to a system booted after manually increasing the timeout in /apps/docker/ current/ bin/docker. start to 8, in that case the service started after 6 seconds.
With the default timeout this is the log for a successful service setup (which happens randomly in a slow system as the one depicted in the description):
-- Logs begin at Wed 2015-08-05 07:02:06 UTC, end at Wed 2015-08-05 07:02:48 UTC. -- core-launcher[ 761]: time="2015- 08-05T07: 02:31Z" level=info msg="+job serveapi( unix:// /var/run/ docker. sock)" core-launcher[ 761]: time="2015- 08-05T07: 02:31Z" level=info msg="Listening for HTTP on unix (/var/run/ docker. sock)"
Aug 05 07:02:29 test-16732 systemd[1]: Started The docker app deployment mechanism.
Aug 05 07:02:29 test-16732 systemd[1]: Starting The docker app deployment mechanism...
Aug 05 07:02:31 test-16732 ubuntu-
Aug 05 07:02:31 test-16732 ubuntu-
...
All these logs have been collected with the debug flag activated for both docker and ubuntu- core-launcher.
Thanks!