Unable to destroy service/machine/unit

[Jeff Pihach]

1.24-beta5-trusty-amd64

I haven't been able to reproduce this since, but somehow my local (lxc) environment ended up in a broken state.

I deployed mysql and before it was done installing I ran
`juju destroy-service mysql`

It put the service in 'dying' state but never came out.
When I ran
`juju destroy-machine --force 4`
nothing changed.

When I tried to destroy the unit I got the following error
`juju destroy-unit mysql/0`
"ERROR no units were destroyed: state changing too quickly; try again soon"

The machine 4's log didn't have anything out of the ordinary except this final line
"FATAL: Could not load /lib/modules/3.13.0-53-generic/modules.dep: No such file or directory"

At this point there was nothing I could do to get the environment back into a functional state and had to destroy it.

Sorry there isn't much here but hopefully these error messages allow someone to work backwards to find the root cause.

[Gabriel Samfira (gabriel-samfira)]

This happened to me as well. Had to go into the mongo database and manually change the workload status from maintenance to error. After which I was able to destroy the unit. I will try to reproduce and post logs.

[Robert Clark (returntoreptar)]

@gabriel-samfira Could you tell me the steps you took to do this? I am having the same error and have been so far unable to get the node out of this state. I am getting the following two errors over and over in juju debug-log

unit-docker-2[1787]: 2015-07-08 16:35:24 ERROR juju.worker.uniter.filter filter.go:137 state changing too quickly; try again soon
unit-docker-2[1787]: 2015-07-08 16:35:24 ERROR juju.worker runner.go:219 exited "uniter": state changing too quickly; try again soon