[
https://issues.apache.org/jira/browse/MESOS-5632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15335650#comment-15335650
]
haosdent commented on MESOS-5632:
---------------------------------
If restart Mesos Agent work for you?
> Orphaned docker container not killed if executor has exited
> -----------------------------------------------------------
>
> Key: MESOS-5632
> URL: https://issues.apache.org/jira/browse/MESOS-5632
> Project: Mesos
> Issue Type: Bug
> Components: docker, slave
> Reporter: Mansheng Yang
>
> [This ticket|https://issues.apache.org/jira/browse/MESOS-3573] is marked as
> resolved but it was only partially fixed.
> As mentioned in that ticket, if you start a docker container, kill the
> docker-executor process, then a new container will be started but the old one
> will still be there.
> Some logs:
> {noformat}
> I0617 15:01:22.851604 7285 docker.cpp:877] Recovering container
> '71695f70-afad-421d-8636-deb6724ecaca' for executor
> 'kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d' of framework
> '317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000'
> I0617 15:01:22.853303 7285 docker.cpp:2107] Executor for container
> '71695f70-afad-421d-8636-deb6724ecaca' has exited
> I0617 15:01:22.853327 7285 docker.cpp:1826] Destroying container
> '71695f70-afad-421d-8636-deb6724ecaca'
> I0617 15:01:22.853575 7285 docker.cpp:1954] Running docker stop on container
> '71695f70-afad-421d-8636-deb6724ecaca'
> I0617 15:01:22.853607 7285 docker.cpp:1956] Running docker stop on container
> 'mesos-cbb3d52c-b6dd-4b7e-864d-705fc2fab983-S4.71695f70-afad-421d-8636-deb6724ecaca'0
> I0617 15:01:22.854801 7283 slave.cpp:4767] Sending reconnect request to
> executor 'kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d' of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000 at executor(1)@127.0.1.1:56304
> E0617 15:01:22.855870 7283 process.cpp:2040] Failed to shutdown socket with
> fd 10: Transport endpoint is not connected
> E0617 15:01:22.855974 7283 slave.cpp:4118] Termination of executor
> 'kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d' of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000 failed: Unknown container:
> 71695f70-afad-421d-8636-deb6724ecaca
> I0617 15:01:22.857015 7283 slave.cpp:3257] Handling status update
> TASK_FAILED (UUID: b5dfa1dc-62db-4fb5-93c8-958d22f930df) for task
> kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000 from @0.0.0.0:0
> W0617 15:01:22.858330 7288 docker.cpp:1403] Ignoring updating unknown
> container: 71695f70-afad-421d-8636-deb6724ecaca
> I0617 15:01:22.858819 7288 status_update_manager.cpp:320] Received status
> update TASK_FAILED (UUID: b5dfa1dc-62db-4fb5-93c8-958d22f930df) for task
> kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000
> I0617 15:01:22.858986 7288 status_update_manager.cpp:824] Checkpointing
> UPDATE for status update TASK_FAILED (UUID:
> b5dfa1dc-62db-4fb5-93c8-958d22f930df) for task
> kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000
> W0617 15:01:22.920336 7289 slave.cpp:3601] Dropping status update
> TASK_FAILED (UUID: b5dfa1dc-62db-4fb5-93c8-958d22f930df) for task
> kafka2.3802f3c9-3459-11e6-bf06-6e0c5199624d of framework
> 317ab6ce-d599-4ad4-bae2-eb74a6c42d87-0000 sent by status update manager
> because the agent is in RECOVERING state
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
