[Bug 1102751] New: Got "Error running removeDevice" after kill -9 docker

[Red Hat Bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1102751

            Bug ID: 1102751
           Summary: Got "Error running removeDevice" after kill -9 docker
           Product: Fedora
           Version: 20
         Component: docker-io
          Assignee: lsm5(a)redhat.com
          Reporter: mfojtik(a)redhat.com
        QA Contact: extras-qa(a)fedoraproject.org
                CC: admiller(a)redhat.com, golang(a)lists.fedoraproject.org,
                    lsm5(a)redhat.com, mattdm(a)redhat.com,
                    mgoldman(a)redhat.com, s(a)shk.io, vbatts(a)redhat.com



Description of problem:

While I was playing with Docker today, I found interesting behavior:

When you 'kill -9' the 'docker -d' process, then this happens:

1. The 'docker -d' process gets killed ;-)
2. All containers gets killed
3. The 'docker -d' restarts, because of Restart=on-failure

Now, while all this is expected and OK, I get this when I tried to 'restart'
the container that was killed in 2):


May 29 13:24:56 ip-10-146-193-102 systemd[1]: Starting Container origin-db-1...
May 29 13:24:56 ip-10-146-193-102 sh[19731]: Reusing
b80104263de02f39bbc6d742c977ddadecb0660f5c50386eaf1cf645f3515b9c
May 29 13:25:07 ip-10-146-193-102 docker[19739]: Error: Cannot destroy
container origin-db-1: Driver devicemapper failed to remove root filesystem
e536cf328e0083e2414c750434deb1127517d00399fddac84903825d6f003787: Error running
removeDevice
May 29 13:25:07 ip-10-146-193-102 docker[19739]: 2014/05/29 13:25:07 Error:
failed to remove one or more containers
May 29 13:25:07 ip-10-146-193-102 gear[21193]: user: unknown user
ctr-origin-db-1
May 29 13:25:09 ip-10-146-193-102 docker[21192]: 2014/05/29 13:25:09 Error:
Cannot start container
bd1a88a35e84ef6d71fa3e1df0b4e2998318dfed6d60fd3ae381567fff2ac2a3: Cannot find
child for /origin-db-1
May 29 13:25:09 ip-10-146-193-102 systemd[1]: ctr-origin-db-1.service: main
process exited, code=exited, status=1/FAILURE


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:

1. Start up some containers
2. kill -9 (the 'docker -d' process pid)
3. Try to start the dead container again
4. Get the error above.

Actual results:


Expected results:

Containers should be successfully started back.


Additional info:

-- 
You are receiving this mail because:
You are on the CC list for the bug.