Re: Deleting a pod sometimes deletes the docker container and sometimes it doesn't

2016-09-20 Thread v 

All right, in that case aggregated logging it is.

Thank you very much for your support. :)


Am 2016-09-20 um 15:46 schrieb Andy Goldstein:

If the pod crashes immediately, they should be able to use 'oc debug' to try to 
determine what is happening. Otherwise, I would recommend using aggregated 
logging.

Andy

On Tue, Sep 20, 2016 at 3:07 AM, v mailto:vekt...@gmx.net>> 
wrote:

Hello,

our use case is that the developers sometimes want to access the logs of 
crashed pods in order to see why their app crashed. That's not possible via the 
OpenShift Console, therefore we backup the logs of exited containers and make 
them accessible to the devs.

I think that this could be accomplished via aggregated logging too (could 
it?) but last time we tried that it turned out to be a very heavyweight 
solution that constantly required our attention and oversight.

Can you give us any recommendation?

Regards
v




Am 2016-09-19 um 15:57 schrieb Andy Goldstein:

When you delete a pod, its containers should **always** be deleted. If they 
are not, this is a bug.

Could you please elaborate what use case(s) you have for keeping the 
containers around?

Thanks,
Andy

On Mon, Sep 19, 2016 at 5:09 AM, v mailto:vekt...@gmx.net>> wrote:

Hello,

we have an issue with docker/openshfit.
With Openshift 1.1.4 and Docker 1.8.2 we can delete a pod via "oc delete po" and the docker 
container is only stopped but not deleted. That means the container is still visible via "docker ps 
-a", /var/lib/docker/containers/[hash] still persits, "docker logs" still works etc.

With Openshift 1.1.6 and Docker 1.9.1 the docker container is sometimes DELETED 
when we delete a pod via "oc delete po" and sometimes it is just stopped like 
with Openshift 1.1.4/Docker 1.8.2.

Is there any way we can influence this? Ideally we would like "oc delete 
po" to just stop the pod but not delete it.

Regards
v


___
users mailing list
users@lists.openshift.redhat.com 

http://lists.openshift.redhat.com/openshiftmm/listinfo/users 








___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users


___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: Deleting a pod sometimes deletes the docker container and sometimes it doesn't

2016-09-20 Thread Andy Goldstein 
If the pod crashes immediately, they should be able to use 'oc debug' to
try to determine what is happening. Otherwise, I would recommend using
aggregated logging.

Andy

On Tue, Sep 20, 2016 at 3:07 AM, v  wrote:

> Hello,
>
> our use case is that the developers sometimes want to access the logs of
> crashed pods in order to see why their app crashed. That's not possible via
> the OpenShift Console, therefore we backup the logs of exited containers
> and make them accessible to the devs.
>
> I think that this could be accomplished via aggregated logging too (could
> it?) but last time we tried that it turned out to be a very heavyweight
> solution that constantly required our attention and oversight.
>
> Can you give us any recommendation?
>
> Regards
> v
>
>
>
>
> Am 2016-09-19 um 15:57 schrieb Andy Goldstein:
>
> When you delete a pod, its containers should **always** be deleted. If
> they are not, this is a bug.
>
> Could you please elaborate what use case(s) you have for keeping the
> containers around?
>
> Thanks,
> Andy
>
> On Mon, Sep 19, 2016 at 5:09 AM, v  wrote:
>
>> Hello,
>>
>> we have an issue with docker/openshfit.
>> With Openshift 1.1.4 and Docker 1.8.2 we can delete a pod via "oc delete
>> po" and the docker container is only stopped but not deleted. That means
>> the container is still visible via "docker ps -a",
>> /var/lib/docker/containers/[hash] still persits, "docker logs" still
>> works etc.
>>
>> With Openshift 1.1.6 and Docker 1.9.1 the docker container is sometimes
>> DELETED when we delete a pod via "oc delete po" and sometimes it is just
>> stopped like with Openshift 1.1.4/Docker 1.8.2.
>>
>> Is there any way we can influence this? Ideally we would like "oc delete
>> po" to just stop the pod but not delete it.
>>
>> Regards
>> v
>>
>>
>> ___
>> users mailing list
>> users@lists.openshift.redhat.com
>> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>>
>
>
>
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: Deleting a pod sometimes deletes the docker container and sometimes it doesn't

2016-09-20 Thread v 

Hello,

our use case is that the developers sometimes want to access the logs of 
crashed pods in order to see why their app crashed. That's not possible via the 
OpenShift Console, therefore we backup the logs of exited containers and make 
them accessible to the devs.

I think that this could be accomplished via aggregated logging too (could it?) 
but last time we tried that it turned out to be a very heavyweight solution 
that constantly required our attention and oversight.

Can you give us any recommendation?

Regards
v



Am 2016-09-19 um 15:57 schrieb Andy Goldstein:

When you delete a pod, its containers should **always** be deleted. If they are 
not, this is a bug.

Could you please elaborate what use case(s) you have for keeping the containers 
around?

Thanks,
Andy

On Mon, Sep 19, 2016 at 5:09 AM, v mailto:vekt...@gmx.net>> 
wrote:

Hello,

we have an issue with docker/openshfit.
With Openshift 1.1.4 and Docker 1.8.2 we can delete a pod via "oc delete po" and the docker 
container is only stopped but not deleted. That means the container is still visible via "docker ps 
-a", /var/lib/docker/containers/[hash] still persits, "docker logs" still works etc.

With Openshift 1.1.6 and Docker 1.9.1 the docker container is sometimes DELETED when 
we delete a pod via "oc delete po" and sometimes it is just stopped like with 
Openshift 1.1.4/Docker 1.8.2.

Is there any way we can influence this? Ideally we would like "oc delete 
po" to just stop the pod but not delete it.

Regards
v


___
users mailing list
users@lists.openshift.redhat.com 
http://lists.openshift.redhat.com/openshiftmm/listinfo/users 





___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: Deleting a pod sometimes deletes the docker container and sometimes it doesn't

2016-09-19 Thread Andy Goldstein 
When you delete a pod, its containers should **always** be deleted. If they
are not, this is a bug.

Could you please elaborate what use case(s) you have for keeping the
containers around?

Thanks,
Andy

On Mon, Sep 19, 2016 at 5:09 AM, v  wrote:

> Hello,
>
> we have an issue with docker/openshfit.
> With Openshift 1.1.4 and Docker 1.8.2 we can delete a pod via "oc delete
> po" and the docker container is only stopped but not deleted. That means
> the container is still visible via "docker ps -a",
> /var/lib/docker/containers/[hash] still persits, "docker logs" still
> works etc.
>
> With Openshift 1.1.6 and Docker 1.9.1 the docker container is sometimes
> DELETED when we delete a pod via "oc delete po" and sometimes it is just
> stopped like with Openshift 1.1.4/Docker 1.8.2.
>
> Is there any way we can influence this? Ideally we would like "oc delete
> po" to just stop the pod but not delete it.
>
> Regards
> v
>
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Deleting a pod sometimes deletes the docker container and sometimes it doesn't

2016-09-19 Thread v 

Hello,

we have an issue with docker/openshfit.
With Openshift 1.1.4 and Docker 1.8.2 we can delete a pod via "oc delete po" and the docker 
container is only stopped but not deleted. That means the container is still visible via "docker ps 
-a", /var/lib/docker/containers/[hash] still persits, "docker logs" still works etc.

With Openshift 1.1.6 and Docker 1.9.1 the docker container is sometimes DELETED when we 
delete a pod via "oc delete po" and sometimes it is just stopped like with 
Openshift 1.1.4/Docker 1.8.2.

Is there any way we can influence this? Ideally we would like "oc delete po" to 
just stop the pod but not delete it.

Regards
v


___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users