[Gluster-devel] Reviews older than 90 days

Shyam srangana at redhat.com
Mon Mar 27 15:04:07 UTC 2017


On 03/17/2017 03:22 AM, Nigel Babu wrote:
> Hello folks,
>
> I put this up for discussion at the last community meeting, but I was unable to
> attend the meeting. We have a lot of reviews that are quite old[1]. They have,
> at some point, become unmergable or the person who proposed them lost interest.
> This is on us. We've let reviews slip through the cracks. There are also the
> odd cases when a second patch has been merged which fixes the actual issue, but
> the old one isn't abandoned.
>
> There's no way we can work our way through the entire list. I suggest that we
> abandon all patches with no updates in the last 90 days. If it hits on a patch
> that someone is particularly passionate to get merged in, they're welcome to
> re-open it.

I agree with this, also it makes the dashboard more manageable.

>
> A. Current master dashboard: https://review.gluster.org/#/projects/glusterfs,dashboards/dashboard:master-dashboard
> B. Master dashboard (older than 90 days): http://bit.ly/2m8FX6m
> C. Master dashboard (only patches newer than 90 days): http://bit.ly/2mB0xsl
>
> If we abandon older patches, our master dashboard will look like (C) above.
> This means that there's already patches that are quite a few reviews that need
> some hand holding. Is it possible for us to catch up to the ones that are
> currently pending?

We should focus on getting things out of the way at least in 90 days, 
one way or the other. So, I would assume as we start using the 
dashboard, we will have better focus on the "Awaiting Reviews Or 
Regression" section as we go bottom up on that.

So I would say we can catch up here (as in C), but not with B, so the 
overall idea seems to be useful to reduce noise at first, and possibly 
never get into this problem in the future as we catch up.

>
> [1]: http://bit.ly/2nfBq1N
>
> --
> nigelb
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at gluster.org
> http://lists.gluster.org/mailman/listinfo/gluster-devel
>


More information about the Gluster-devel mailing list