unit agent fails to start after restart - run.socket: connection refused
Bug #1286213 reported by
Michael Nelson
This bug affects 6 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
juju-core |
Fix Released
|
Critical
|
Dimiter Naydenov |
Bug Description
Steps to repro:
1) juju bootstrap
2) juju deploy --repository=../.. local:whatever
3) Confirm that `juju run --service whatever "echo 'Hello world'"` works
4) sudo lxc-stop --reboot --name=
5) Retry `juju run --service whatever "echo 'Hello world'"`
Result:
error: dial unix /var/lib/
Everything else seems to work fine (debug-hooks, ssh etc.), eg:
$ juju ssh whatever/0 "echo 'Hello world'"
Hello world
Note, you don't have to reboot the machine to trigger this. Just doing "restart jujud-unit-
Related branches
lp:~dimitern/juju-core/320-lp-1286213-juju-run-socket-remove
- Juju Engineering: Pending requested
-
Diff: 124 lines (+23/-19)5 files modifiedcmd/jujud/run.go (+1/-1)
cmd/jujud/run_test.go (+1/-1)
worker/uniter/runlistener.go (+12/-7)
worker/uniter/runlistener_test.go (+6/-6)
worker/uniter/uniter.go (+3/-4)
tags: | added: landscape |
Changed in juju-core: | |
assignee: | nobody → Dimiter Naydenov (dimitern) |
status: | Triaged → In Progress |
Changed in juju-core: | |
status: | In Progress → Fix Committed |
Changed in juju-core: | |
status: | Fix Committed → Fix Released |
To post a comment you must log in.
Is the lxc series different from the host series? This issue may be a dupe of bug 1285901, which has a fix landing at this moment.