[Gluster-devel] [Gluster-Maintainers] Why a -2 need to be carried over to next patch set?

[Niels de Vos]

On Wed, Oct 12, 2016 at 11:09:37AM +0530, Atin Mukherjee wrote:
> Although a -2 on a patch indicates that reviewer has strongly disagreed on
> the changes done on the patch, but is it right to carry forward the same
> vote on the subsequent patch set(s)? What if the changes on the following
> patch sets are in line with the comments on the patch set where a -2 was
> mentioned? As it stands until the same reviewer revokes the -2, the patch
> can't be merged. Is this what was intended for?

Yes, otherwise the reviewer should have used -1.

> My primary concern here is if the concerned person is unavailable (for
> various reason) the acceptance of the patch gets delayed even if we have co
> maintainers for the same module acking the patch?

If people give -2 for components they are not responsoble for, I think
we should educate them. Maintainers of components or stable versions
should be allowed to give -2 (or +2), others can give that as well, but
it should be seen as a -1 (or +1) in that case.

For patches that are really urgent, and the -2 reviewer does not respond
in a few days/week, the Gerrit admins can remove the -2. For these cases
I would expect an email to the reviewer, with gluster-devel on CC.

> What do others think here? Should we continue to carry over a -2 on the
> subsequent patch sets?

Yes.

Niels
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <http://www.gluster.org/pipermail/gluster-devel/attachments/20161012/bc3a9765/attachment-0001.sig>