[
https://issues.apache.org/jira/browse/MESOS-3699?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
dave choi updated MESOS-3699:
-----------------------------
Description:
Docker container name has been changed at version 0.23.0 from
“DOCKER_NAME_PREFIX + containerId” to "DOCKER_NAME_PREFIX + slaveId +
DOCKER_NAME_SEPERATOR + containerId”.
So docker container of task of version 0.22.X has failed to stop it.
I think that the docker containers should be renamed when a slave recover.
was:
Docker container name has been changed at version 0.23.0 from
“DOCKER_NAME_PREFIX + containerId” to "DOCKER_NAME_PREFIX + slaveId +
DOCKER_NAME_SEPERATOR + containerId”.
So docker container of task of version 0.22.X has failed to stop it.
I think that the docker containers should be renamed when a slave re-register.
> After upgrade version to 0.23.0, then If a task that has been executing from
> version 0.22.X is killed, the docker container has been orphaned.
> ----------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: MESOS-3699
> URL: https://issues.apache.org/jira/browse/MESOS-3699
> Project: Mesos
> Issue Type: Bug
> Components: docker
> Affects Versions: 0.23.0
> Reporter: dave choi
>
> Docker container name has been changed at version 0.23.0 from
> “DOCKER_NAME_PREFIX + containerId” to "DOCKER_NAME_PREFIX + slaveId +
> DOCKER_NAME_SEPERATOR + containerId”.
> So docker container of task of version 0.22.X has failed to stop it.
> I think that the docker containers should be renamed when a slave recover.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
