[Container-tools] using openshift origin v3 to back centos build pipeline

[Václav Pavlín]

On 07/20/2015 03:42 PM, Aaron Weitekamp wrote:
> On Mon, Jul 20, 2015 at 7:18 AM, Karanbir Singh <kbsingh at redhat.com 
> <mailto:kbsingh at redhat.com>> wrote:
>
>     On 20/07/15 10:53, Tomas Tomecek wrote:
>
>     > Quoting Aaron Weitekamp (2015-07-17 21:23:53)
>
>     >> Looking deeper into the capabilities of openshift origin[1] have we
>
>     >> considered this as the core tech for pushing image content
>     through the
>
>     >> pipeline? I don't think I'm on the right mailing lists for this
>     discussion.
>
>     >> I'm just curious on the direction here.
>
>     >>
>
>     >> [1] https://docs.openshift.com/enterprise/3.0/dev_guide/builds.html
>
>     >>
>
>     > We have discussed this (using fullblown osbs [1][2] for pipeline).
>
>     >
>
>     > Not sure what's the current architecture for pipeline atm.
>
>     >
>
>     > [1] https://github.com/projectatomic/osbs-client
>
>     > [2]
>     https://github.com/projectatomic/osbs-client/blob/master/docs/osbs_instance_setup.md
>
>     >
>
>
>
>     does'nt OSOv3 also run OSBS under the hood ?
>
>
>
>     we looked at OSBS and the layers that get added on dont really scale
>
>     very well to multiple, on demand sort of installations. For that,
>     we're
>
>     just using atomicapp-builder and dock ( now atomic-reactor ).
>
>
> ​Thanks for the info. I'm trying to sync what we're building vs. what 
> tools ​end-users might use to build out a CI/CD workflow. OpenShift 
> makes a lot of sense and I would assume scaling issues will be 
> resolved in upcoming releases.
It's not about OSv3 scalability (i.e. about building thousands of 
images, that's something OpenShift should be able to handle). It's about 
being able to quickly reproduce the build environment as CentOS uses 
baremetal machines and needs to be able to quickly provision them with 
parts of pipeline that need to run atm. From our discussions with KB, I 
understood that we want to start with fresh empty environment, deploy 
builder, pull existing deps, build what is missing, run some sanity 
tests, inspect the result, push it and tear down the deployment so that 
we can start fresh again.

Thus the scalability here means - how much time does it take to deploy 
the builder? How hard it is to maintain the deployment configuration?

Vašek
>
>
>
>
>     regards,
>
>
>
>     --
>
>     Karanbir Singh, The CentOS Project, London, UK
>
>     RH Ext. 8274455 | DID: 0044 207 009 4455
>
>
>
>     _______________________________________________
>
>     Container-tools mailing list
>
>     Container-tools at redhat.com <mailto:Container-tools at redhat.com>
>
>     https://www.redhat.com/mailman/listinfo/container-tools
>
>
>
>
>
>
> _______________________________________________
> Container-tools mailing list
> Container-tools at redhat.com
> https://www.redhat.com/mailman/listinfo/container-tools
>
>


-- 

Architect - Senior Software Engineer
Developer Experience
Brno, Czech Republic
Phone: +420 739 666 824


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/container-tools/attachments/20150721/eae93e15/attachment.htm>