Hi,

registry runs as a separate container and its data can be persistently
stored
in a volume, mounted to the container. As I outlined in my previous
response.

I think that `/registry` or `/var/lib/registry` is the mount point for the
registry.

Registry is a service that allows people to share their images. But you
don't
need it if you work alone on your laptop - docker daemon keeps images in
`/var/lib/docker`.

Here is a comprehensive guide on registry in openshift, but it was written
for a beta release[0].

[0]
https://github.com/openshift/training/blob/master/07-Registry-and-Storage.md

Josef Karasek, xPaaS

On Tue, Jan 10, 2017 at 2:47 PM, Pri <[email protected]> wrote:

> Hi Thankyou,I have one more query. This document here
> https://access.redhat.com/sites/default/files/attachments/ocp-on-aws-7.pdf,
> uses  S3 for the persistent storage of the OpenShift registry. Is it
> different from docker storage. What it stores, when we say images does it
> pushes images to S3?? or metadata such as binaries, temp files, logs etc
> is stored in S3?
>
> Would be really helpful if you can help understand this.
>
> Thanks,
> Priya
>
> On Tue, Jan 10, 2017 at 3:44 PM, Josef Karasek <[email protected]>
> wrote:
>
>> Hi Pri,
>>
>> docker storage backend you use has huge impact on performance and
>> container density per machine[0].
>> For example docker in rhel distribution uses LVM loopback device. This is
>> very easy to configure
>> and therefore nice default. But also fairly slow. By that I mean slow to
>> start a new container or build a new image.
>> Red Hat's docker distribution comes with a tool that makes setting up
>> docker storage backend much easier.
>> Check it out[1]. Here is a slide deck that briefly summarizes steps
>> needed in order to setup the backend[2]
>>
>> Here is a little old, but still very good performance analysis done by
>> Jeremy[3].
>>
>> Kubernetes persistent volumes are quite different concept. They're not
>> used by docker daemon
>> but by containers in a pod. These volumes are mounted to a container at
>> it's creation time and their purpose
>> is to persistently store data that is created over the container's
>> lifetime.
>> Such example can be NFS storage for a PostgreSQL that runs in openshift.
>>
>> [0] https://developers.redhat.com/blog/2016/10/25/docker-project
>> -can-you-have-overlay2-speed-and-density-with-devicemapper-yep/
>> [1] http://www.projectatomic.io/docs/docker-storage-recommendation/
>> [2] https://josefkarasek.github.io/docker101/#/165
>> [3] https://developers.redhat.com/blog/2014/09/30/overview-stora
>> ge-scalability-docker/
>>
>> Here are some more links that I find highly useful when working with
>> docker storage:
>> http://www.projectatomic.io/blog/2016/05/docker-lvm-plugin/
>> http://www.projectatomic.io/blog/2015/06/notes-on-fedora-cen
>> tos-and-docker-storage-drivers/
>> http://www.projectatomic.io/blog/2015/06/using-volumes-with-
>> docker-can-cause-problems-with-selinux/
>>
>> Please note that docker and openshift are living projects and things
>> change from time to time, so some
>> articles can include information that is no longer accurate.
>>
>> Josef Karasek, xPaaS
>>
>>
>> On Tue, Jan 10, 2017 at 4:47 AM, Pri <[email protected]>
>> wrote:
>>
>>> Hi Akram & Andy
>>>
>>> Thanks for the response, really helpful. Just to be sure, do we need
>>> both for production setup? Will it effect my running app if I don't setup
>>> docker storage on each node?
>>>
>>> Thanks,
>>> Priya
>>>
>>> On Mon, Jan 9, 2017 at 8:25 PM, Andy Goldstein <[email protected]>
>>> wrote:
>>>
>>>>
>>>>
>>>> On Mon, Jan 9, 2017 at 9:45 AM, Akram Ben Aissi <
>>>> [email protected]> wrote:
>>>>
>>>>> Hi Pri,
>>>>>
>>>>> docker storage is locally used by docker on each node to store
>>>>> container runtime data (binaries, temp files, logs).
>>>>>
>>>>
>>>> More specifically, it stores docker images and any modifications made
>>>> to a container's file system after you've started the container.
>>>>
>>>>
>>>>>
>>>>> persistent volumes are attachable volumes, automatically mounted on
>>>>> container landing that is used to store data intended to be permanent
>>>>> accross containers failures, restarts or re-scheduling;
>>>>>
>>>>>
>>>>> On 9 January 2017 at 14:31, Pri <[email protected]> wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I would like to understand the difference between docker
>>>>>> storage-setup and persistent volumes for pods. Aren't both same?
>>>>>>
>>>>>> Do we need to configure both for production purpose?
>>>>>>
>>>>>> Would be very helpful if someone can explain this.
>>>>>>
>>>>>> Thanks a lot
>>>>>> Priya
>>>>>>
>>>>>> _______________________________________________
>>>>>> dev mailing list
>>>>>> [email protected]
>>>>>> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> dev mailing list
>>>>> [email protected]
>>>>> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
>>>>>
>>>>>
>>>>
>>>
>>> _______________________________________________
>>> dev mailing list
>>> [email protected]
>>> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
>>>
>>>
>>
>
_______________________________________________
dev mailing list
[email protected]
http://lists.openshift.redhat.com/openshiftmm/listinfo/dev

Reply via email to