Take the 2-minute tour ×
Ask Ubuntu is a question and answer site for Ubuntu users and developers. It's 100% free, no registration required.

I installed a lab of openstack through juju, deployed rabbitmq-server on the same node that holds mysql. but I found that rabbitmq-server gives unsolvable hook failed error.

So I tried to destroy the service to deploy it again but it doesn't. I tried destroy service from juju-gui and juju destroy-service rabbitmq-server and it still there. only the unit rabbitmq-server/0 were gone.

Now the service "life" is stuck on "dying"

rabbitmq-server: charm: cs:precise/rabbitmq-server-16 exposed: true life: dying relations: cluster: - rabbitmq-server

I connected to the node and didn't find any logs related to rabbitmq at the logs directory. but I found this logs:

/var/log/juju/unit-rabbitmq-server-0.log
2013-12-12 11:16:33 INFO juju.worker.uniter uniter.go:363 committing "stop" hook
2013-12-12 11:16:33 INFO juju.worker.uniter uniter.go:381 committed "stop" hook
2013-12-12 11:16:33 INFO juju.worker.uniter modes.go:421 ModeContinue starting
2013-12-12 11:16:33 INFO juju.worker.uniter modes.go:84 continuing after "stop" hook
2013-12-12 11:16:33 INFO juju.worker.uniter modes.go:421 ModeTerminating starting
2013-12-12 11:16:33 INFO juju.worker.uniter uniter.go:105 unit "rabbitmq-server/0" shutting down: agent should be terminated
2013-12-12 11:16:33 ERROR juju.worker.uniter.filter filter.go:117 tomb: dying
2013-12-12 11:16:33 ERROR juju runner.go:200 worker: fatal "uniter": agent should be terminated
2013-12-12 11:16:33 ERROR juju runner.go:200 worker: fatal "api": agent should be terminated
2013-12-12 11:16:33 INFO juju supercommand.go:286 command finished

Is there anyway to force it to be destroyed completely?

share|improve this question

2 Answers 2

up vote 3 down vote accepted

You can't remove that service, but you can deploy a new service with the same charm:

juju deploy rabbitmq-server rabbitmq

...and you can even put it on the same machine if you want:

juju deploy rabbitmq-server rabbitmq --to <machine-id>

...but you may wish to update your client, and run:

juju upgrade-juju

...to get your environment running 1.16.5, in which that bug is resolved. It won't fix an environment that's already in that situation, but it will prevent it from happening again with the new service.

share|improve this answer
    
That is a great idea and it took care of everything. Thank you @fwereade –  Harti Dec 15 '13 at 14:40

You need to resolve the error using:

juju resolved rabbitmq-server/0 (assuming that is its unit name)

...to continue past the failing hook. You may need to do this more than once, because the unit may have more hooks to run before it's finally cleaned up, and if it's in a really bad state they might well error too.

share|improve this answer
    
I couldn't resolve it. tried many times but this is the only instatant that refuse resolving. –  Harti Dec 12 '13 at 11:12
    
It's also been in the state "dying" for a day now. This is the juju status output: rabbitmq-server: charm: cs:precise/rabbitmq-server-16 exposed: true life: dying relations: cluster: - rabbitmq-server units: rabbitmq-server/0: agent-state: stopped agent-version: 1.16.3 life: dead machine: "12" public-address: rm-mas-03.RMGS.EG –  Harti Dec 12 '13 at 11:17
    
@user224433 Can you update your question with this information so it's up top instead of buried in a comment? It would help to get that debug information in the question. –  Jorge Castro Dec 13 '13 at 20:36

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.