CQ/commit bot not posting commit hash |
||||||||
Issue descriptionIssue 605429 seems to be happening again. I've landed a few CLs that haven't been tagged with the commit position recently: https://codereview.chromium.org/1929113002 https://codereview.chromium.org/1914233005 Some other recent ones from the waterfall: https://codereview.chromium.org/1925013002 https://codereview.chromium.org/1926813005 https://codereview.chromium.org/1925803003
,
Apr 28 2016
,
Apr 28 2016
Hmm... gnumbd is running so I'm not sure what's up here. Adding some people who might know what is going on.
,
Apr 28 2016
it is not gnumbd that posts those comments, but commit_position_bot last postmortem: https://docs.google.com/document/d/1iVhf_SqRwQlvGiyAqQ1yrcezddBm8kryBwtOaU6hETs/edit
,
Apr 29 2016
I've been tracking my work in https://bugs.chromium.org/p/chromium/issues/detail?id=605429. Short answer: it's still not working, and I'm not sure precisely what to do. Tagging iannucci@ and tandrii@ in case they know something.
,
Apr 29 2016
That system is not part of CQ.
,
Apr 29 2016
Here's the first commit that failed: https://chromium.googlesource.com/chromium/src/+/26c9ab0b74189cc46821a12d210635f5ba1f80af Here's its parent: https://chromium.googlesource.com/chromium/src/+/7819534cfb29bb8d80c61668cde8341ffaca3ab3 Notice anything? The "Review URL" footer changed to "Review-Url" and commit position bot can no longer extract the review URL. This CL changed the footer appended to CQed CLs: https://chromereviews.googleplex.com/412857013 I'm adding Infra>CQ back because a CQ CL was the root case, even if the broken system is not part of CQ.
,
Apr 29 2016
Issue 607953 has been merged into this issue.
,
Apr 30 2016
smut@ you are totally right! Let me quick land a fix to the bot code itself: https://chromereviews.googleplex.com/414277013
,
Apr 30 2016
So, there are two options now: 1. re-wind the state of the bots to re-processing missed commit. 2. or just let them start working from whatever lands next. The #2 has obvious issue that 4 days of commits won't get tagged. #1 downside is that manually landed commits will be tagged twice. IMO, that's lesser evil, so proceeding with #1.
,
Apr 30 2016
The following revision refers to this bug: https://chrome-internal.googlesource.com/infra/infra_internal.git/+/07e30923e075ed96c8f373e2e923c3bcbd7e2370 commit 07e30923e075ed96c8f373e2e923c3bcbd7e2370 Author: tandrii <tandrii@google.com> Date: Sat Apr 30 17:22:17 2016
,
Apr 30 2016
The following revision refers to this bug: https://chrome-internal.googlesource.com/infra/infra_internal.git/+/8df5d176974b3b32a1497bfac712b5ff4a2078a1 commit 8df5d176974b3b32a1497bfac712b5ff4a2078a1 Author: tandrii <tandrii@google.com> Date: Sat Apr 30 17:27:21 2016
,
Apr 30 2016
For chrome - done. Remaining v8, webrtc -> this evening.
,
Apr 30 2016
Also, the very last commit disables double-sending of emails.
,
May 1 2016
|
||||||||
►
Sign in to add a comment |
||||||||
Comment 1 by no...@chromium.org
, Apr 28 2016