Today we had a quiet, civilized, meeting where the following points were raised:
- For a week, I've missed a report by Itzik about vdsm starting to send <feature><acpi/> tag to libvirt when asked to. https://lists.fedorahosted.org/pipermail/vdsm-devel/2012-August/001335.html Could this be related to http://gerrit.ovirt.org/6846 or http://gerrit.ovirt.org/5369 ? I suppose not, but maybe the authors can help me understand what's going on here.
- Robert: about to release a Jenkins job, that compiles rpm per patch, to help people test patches.
- Federico: agrees to look into why we get fishy
Exception RuntimeError: RuntimeError('cannot notify on un-acquired lock',) in <module 'threading' from '/usr/lib64/python2.7/threading.pyc'> ignored
messages from the unit tests, even though it is not due to his recent RLock work.
- Adam: libvdsm still has few shortcomings; plans to enumerate known issue by tomorrow, and push to the world. Asks that Federico reviews his autobuild changes.
- Saggi: when that's out, will work on Java binding for the benefit of Engine. Dan: we must get Engine's consent for libvdsm.
- We need to set the priorities for ovirt-3.2 release, planned for November, to fit into Fedora 19. Adam and Dan thinks that libvsdm is high on that list.
- Deepack: requests review for his gluster work http://gerrit.ovirt.org/#/c/6856/
- Adam: mom integration: VMs are to have policy tags, such as Gold/ Silver/ Bronze to be enforced by MoM.
- Robert: should we auto-test mom? Adam to add `make check` with basic stuff such as pyflakes.
- Dan: should we abandon old patches? Federico: no, people may want they Work-in-Progress patches to stay around and inspire others. A nag mail for a patch that is not abandoned and not advancing anywhere is more reasonable.
If someone has some time to put into gerrit scripting, I won't mind to see such a nag bot. It would help we separate important patches that were somehow forgotten by their author/reviewers and stuff that should probably be recycled.
- I'm pretty sure we had more participants, and more issues raised. Please send a patch to this email with corrections/ additions.
Regards,
Dan.
On 08/27/2012 06:01 PM, Dan Kenigsberg wrote: ...
Dan: should we abandon old patches? Federico: no, people may want they Work-in-Progress patches to stay around and inspire others. A nag mail for a patch that is not abandoned and not advancing anywhere is more reasonable.
If someone has some time to put into gerrit scripting, I won't mind to see such a nag bot. It would help we separate important patches that were somehow forgotten by their author/reviewers and stuff that should probably be recycled.
I was thinking of a bot to send emails to patch owners if patch is untouched for more than 60 days, remind on 75 days, and abandon on 90 days.
any modification of the patch (review comment, updated patch, etc every 2 months) would keep it alive.
On Mon, Aug 27, 2012 at 08:17:36PM +0300, Itamar Heim wrote:
On 08/27/2012 06:01 PM, Dan Kenigsberg wrote: ...
Dan: should we abandon old patches? Federico: no, people may want they Work-in-Progress patches to stay around and inspire others. A nag mail for a patch that is not abandoned and not advancing anywhere is more reasonable.
If someone has some time to put into gerrit scripting, I won't mind to see such a nag bot. It would help we separate important patches that were somehow forgotten by their author/reviewers and stuff that should probably be recycled.
I was thinking of a bot to send emails to patch owners if patch is untouched for more than 60 days, remind on 75 days, and abandon on 90 days.
any modification of the patch (review comment, updated patch, etc every 2 months) would keep it alive.
Consensus was that dead but un-abandoned patches are not annoying enough to auto-abandon them. Let us try to avoid the abandon whip for now, and start by polite reminders.
Dan.
vdsm-devel@lists.fedorahosted.org