-
Notifications
You must be signed in to change notification settings - Fork 51
Docker service is failing to start with latest build of the vagrant box #69
Comments
The devicemapper arguments in
More info: https://github.com/docker/docker/blob/master/man/docker.1.md |
@jasonbrooks Thanks for the pointer. It looks like docker cmd line does not need data and metadata logical volume path and just need path to docker--pool. So do I need to create a lv for docker-pool ? Added later: |
@LalatenduMohanty : As I pointed in #68 (comment) ; there are multiple entries of variable DOCKER_STORAGE_OPTIONS in |
@jasonbrooks : I tried updating the the variable however the service is failing with msg="Error starting daemon: error initializing graphdriver: Non existing device vg001-docker--pool" I think, I am missing some preliminary step(s) before updating the [vagrant@localhost ~]$ rpm -q docker
docker-1.7.1-108.el7.centos.x86_64
[vagrant@localhost ~]$ cat /etc/sysconfig/docker-storage
# DOCKER_STORAGE_OPTIONS = --storage-opt dm.metadatadev=/dev/mylogvol/my-docker-metadata --storage-opt dm.datadev=/dev/mylogvol/my-docker-data
DOCKER_STORAGE_OPTIONS=--storage-driver devicemapper --storage-opt dm.fs=xfs --storage-opt dm.thinpooldev=/dev/mapper/vg001-docker--pool --storage-opt dm.use_deferred_removal=true
[vagrant@localhost ~]$ sudo systemctl restart docker ; sudo systemctl status docker -l
Job for docker.service failed. See 'systemctl status docker.service' and 'journalctl -xn' for details.
docker.service - Docker Application Container Engine
Loaded: loaded (/usr/lib/systemd/system/docker.service; enabled)
Drop-In: /usr/lib/systemd/system/docker.service.d
└─flannel.conf
Active: failed (Result: exit-code) since Thu 2015-08-20 02:30:06 EDT; 12ms ago
Docs: http://docs.docker.com
Process: 19140 ExecStart=/usr/bin/docker -d $OPTIONS $DOCKER_STORAGE_OPTIONS $DOCKER_NETWORK_OPTIONS $ADD_REGISTRY $BLOCK_REGISTRY $INSECURE_REGISTRY (code=exited, status=1/FAILURE)
Main PID: 19140 (code=exited, status=1/FAILURE)
Aug 20 02:30:06 localhost.localdomain systemd[1]: Starting Docker Application Container Engine...
Aug 20 02:30:06 localhost.localdomain docker[19140]: time="2015-08-20T02:30:06.767464326-04:00" level=fatal msg="Error starting daemon: error initializing graphdriver: Non existing device vg001-docker--pool"
Aug 20 02:30:06 localhost.localdomain systemd[1]: docker.service: main process exited, code=exited, status=1/FAILURE
Aug 20 02:30:06 localhost.localdomain systemd[1]: Failed to start Docker Application Container Engine.
Aug 20 02:30:06 localhost.localdomain systemd[1]: Unit docker.service entered failed state.
[vagrant@localhost ~]$ ls /dev/mapper/
control vg001-docker--data vg001-docker--meta vg001-root |
@navidshaikh Yes, I have seen your comment in #68 (comment) . However @jasonbrooks also pointed some more stuff in #69 (comment) . Looks like docker storage options has changed. So trying to understand it. |
@navidshaikh So it seems we need a separate thinpool for the docker pool. |
from The current versions in C7 are more then lvm2-2.02.112
|
|
…cated now docker-storage-setup now smart to use the freespace from the volume group as thinpool i.e. docker pool This also fixes projectatomic#69 Signed-off-by: Lalatendu Mohanty <lmohanty@redhat.com>
As dm.datadev dm.metadatadev is deprecated now. Also now docker-storage-setup is smart to use the freespace from the volume group as thinpool i.e. docker pool which is the recommended way of setting up storage for docker. This also fixes projectatomic#69 Signed-off-by: Lalatendu Mohanty <lmohanty@redhat.com>
Read #68 (comment)
Scratch build: https://cbs.centos.org/koji/taskinfo?taskID=18983
CC @navidshaikh
The text was updated successfully, but these errors were encountered: