> At the moment, the debian-i386 and freebsd82-amd64 builders are > running "make check" with warnings instead of failing.
What good is it going to do if no one knows to look? Or if they always fail and therefore no one pays attention. Until there is a stable set of tests that are known to consistently succeed and which will not deadlock the build slaves if they fail in just the right way? > We already have multiple pools of builders, which aren't mutually > exclusive. We have two main pools: the gerrit-triggered pool > (debian-i386, freebsd82-amd64,opensuse-x86_64, win2008 builders), and > the scheduled builders (the rest). Some branches only build on certain > builders. The only builders that get regular attention are the ones that break the verification of patchsets in gerrit. All of the others are ignored. To fix that part, failing builders that are not gerrit triggered should send an e-mail. Perhaps to openafs-devel. > We can split the builders and slaves however we like. The main issue > is that it would be nice to run "make check" on as many platforms as > possible, but we need more slaves, possibly double, if we want the > same platform coverage and a completely isolated testing pool. You do not need more slaves. You just need to configure the slaves differently. All of the Windows builders are the same machine. The builders just need to be configured to offer additional build configurations. Jeffrey Altman
signature.asc
Description: OpenPGP digital signature
