openstack team mailing list archive

[Mark McLoughlin <markmc@xxxxxxxxxx>]

On Fri, 2011-11-11 at 08:57 -0800, James E. Blair wrote:
> Mark McLoughlin <markmc@xxxxxxxxxx> writes:
> 
> > On Fri, 2011-11-11 at 12:11 +0400, Yuriy Taraday wrote:
> >> I wonder if we should keep Change ID consistent in stable and master
> >> branches so that if one merged something into master, reviewers
> >> and archaeologists can easily find both related changes in master and all
> >> backports of specific change.
> >> 
> >> The simple scenario is: push change into master, then cherry-pick it on top
> >> of stable branch(es). Change-Id will be the same, Gerrit will allow one to
> >> find all such backports by clicking on Change-Id.
> >
> > If gerrit can handle it, that would be great. But I'm not sure it does
> 
> It does work as Yuriy described, and seems to be in keeping with gerrit
> philosophy.  Maybe we should update the wiki to incorporate that.
> 
> Here's an example: 
> 
> https://review-dev.openstack.org/#q,I1729eb6fb7027808650bae9a87b2d95cc5c5a0f7,n,z

Cool, I'll update the wiki.

> > In the mean time, we make sure that all commits to the stable branch
> > include "cherry picked from XXXXX" in the commit message to help
> > tracking.
> >
> > Also, I'm experimenting with using git-notes to keep track of e.g. why
> > patches on master weren't cherry-picked into stable:
> >
> >   http://wiki.openstack.org/StableBranch#Keeping_Notes
> 
> Why not (also) leave review comments to that effect in gerrit?  If you
> started them out with something like "Reviewed for stable inclusion",
> they'd be easy to spot when scanning the collapsed comments.

Well, what I want to do is occasionally go over all the commits that
have been made on master since the last time I reviewed them. And I
don't really want to have to go to gerrit for every single commit on
master to add comments.

Cheers,
Mark.