On Wed, 2007-05-30 at 13:35 -0500, Tom "spot" Callaway wrote: > The very specific logic path is this: > > 1. A build is submitted. > 2. Primary architectures run. If any primary architecture fails, it > stops. > 3. When all primary architectures pass, then the build is sent to all > secondary architectures. > 4. Each secondary architecture runs to completion. Failures by any > secondary architecture do not affect other secondary architectures. > 5. Secondary architectures which failed make a lot of noise. (This is > where automagic bug filing occurs, emailing architecture teams, etc). My concern is that you're tinkering with a balance (between package and arch maintainers) which doesn't currently need to be poked at -- it's working very well as it is. By letting packages go through after a build failure without _any_ intervention from the package maintainer, you're encouraging the less conscientious package maintainers just to let it happen. But _all_ build failures should be investigated, because they might be a generic problem. The procedures should encourage that. Also, I'm not sure how chains of packages should be handled. Would it be _mandatory_ to use 'koji chain-build'? Am I the only person who's used to just submitting the first build then waiting for it to complete before submitting the next? That would break with the above plan, wouldn't it? Spot, I'd like to see the document start with a very clear statement of whatever 'problem' it's trying to solve. -- dwmw2 -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
