1277738 - Mozreview refuses all review request if one reviewer who has r+'d a patch is refusing to review requests

Status: RESOLVED INVALID

(MozReview Graveyard :: General, defect)

Description

[Chris Pearce [:cpearce (Not reading bugmail)]]

If a reviewer has marked in their bugzilla profile that they wish to refuse review requests, and you have one of your commits in a mozreview push marked as r=thatRefusingReviewer, then you can't push to mozreview to get review from *other* reviewers on other changesets.

For example, bug 1271483. :Gijs is refusing review. There are 16 commits in the review push, two of which is r=gijs. We want review on patch 7 from bsmedberg. So we have two commits with "r=gijs" in their commit message, and one with "r?bsmedberg". But mozreview refuses to let us push to mozreview, even though we're not actually seeking review from someone who's refusing review. I just have commits previously r+'d by someone who's now blocking review requests in the same push.

Comment 1

[:glob ✱]

i'm unable to reproduce this using the following steps:
1. commit r?userA
2. commit r?userB
3. push for review
4. userA gives first commit r+
5. userA blocks review requests in bmo
6. update second commit
7. push for review

i've also tried updating the first commit.  in both cases the review was pushed without issue and the r+ from userA was carried forward.


i'll try with a review made before some changes there were pushed out yesterday, with the fixup commit after said changes.


cpearce: can you provide the output that you're seeing please?

Comment 2

[:glob ✱]

(In reply to Byron Jones ‹:glob› from comment #1)
> i'll try with a review made before some changes there were pushed out
> yesterday, with the fixup commit after said changes.

this also worked.

Comment 3

[Chris Pearce [:cpearce (Not reading bugmail)]]

(In reply to Byron Jones ‹:glob› from comment #1)
> cpearce: can you provide the output that you're seeing please?

I made the push from my machine which I don't have with me right now, I can re-try when I get back from mozlondon.

A factor that may also contribute is that I'd been pushing the same changesets to someone else's mozreview. That is, I took over work in bug 1271483 by pulling the changesets out of mozreview, modified one, and pushed back to review. Maybe that's a factor?

Comment 4

[BMO Automation]

MozReview is now obsolete. Please use Phabricator instead. Closing this bug.