Comments in-line, thanks for the feedback Paul ! ----- Original Message ----- > From: "Paul Belanger" <[email protected]> > To: "David Moreau Simard" <[email protected]> > Cc: "Matthieu Huin" <[email protected]>, [email protected], > [email protected], "Alan Pevec" > <[email protected]>, "Haikel Guemar" <[email protected]>, > [email protected] > Sent: Thursday, September 15, 2016 6:40:10 PM > Subject: Re: jobs and nodepool API wishlist > > On Thu, Sep 15, 2016 at 11:51:06AM -0400, David Moreau Simard wrote: > > + Paul Belanger, I think his input here would be welcome > > > > David Moreau Simard > > Senior Software Engineer | Openstack RDO > > > > dmsimard = [irc, github, twitter] > > > > > > On Thu, Sep 15, 2016 at 11:30 AM, Matthieu Huin <[email protected]> wrote: > > > o/ > > > > > > Next wednesday we intend to groom user stories related to improvements in > > > the manageSF API regarding jobs (jenkins) and nodepool (slave image > > > rebuild etc). > > > > > > There's an etherpad with a tentative list of calls we could implement: > > > http://softwarefactory-project.io/etherpad/p/jobs_nodepool_API_wishlist > > > > Specific to nodepool, there is an open spec upstream for a REST API for > nodepool[1], I would encourage people to comment and contribute there. > Additionally, with the current work that is going on for zuulv3, nodepool > itself > is getting some good upgrades. I believe the REST API will be one of the > next > things we'll be working on.
It's starting to be a big need for SF users, so unless this can land quickly upstream I think we'll go for a temporary solution until the official REST API is out. > I cannot comment on the jenkins part, because we do not use jenkins upstream > any > more. However, some thought on how to extract the information would be via > openstack-health[2] and firehose.openstack.org[3]. Again, a good topic to > raise > upstream. I saw openstack-health in Austin and thought it was very cool, but IIRC it requires jobs to export data in the subunit format. This is not something we do now, nor something we should expect to enforce. I haven't heard of firehose so I'll have a look. > [1] https://review.openstack.org/#/c/141016/ > [2] http://status.openstack.org/openstack-health > [3] > https://specs.openstack.org/openstack-infra/infra-specs/specs/firehose.html > > > > Also, a pre-requisite might be to migrate jenkins to a mysql backend: > > > http://softwarefactory-project.io/redmine/issues/1544 which needs to be > > > groomed as well. > > > > The only downside I see with doing this, is you might lose the ability to run > JJB and mange your job configurations with it. Pretty big blocker :/ Thanks for mentioning that as I had no idea. We might have to compromise there. > > > As advanced users and consumers of SF do not hesitate to have a look at > > > the etherpad beforehand to suggest anything we might have overlooked > > > regarding jobs and slave mgmt. > > > > > > > > > Thanks for your feedback, > > > > > > mhu > _______________________________________________ Softwarefactory-dev mailing list [email protected] https://www.redhat.com/mailman/listinfo/softwarefactory-dev
