Compile confirm no-op step is failing on Windows try bots |
|||||
Issue descriptionIt appears to be flaky, in that it failed, then passed twice, then failed continuously on one of the bots. https://build.chromium.org/p/chromium.win/builders/Win%20Builder https://build.chromium.org/p/chromium.win/builders/Win%20x64%20Builder https://build.chromium.org/p/chromium.win/builders/Win%20x64%20Builder%20%28dbg%29
,
Jun 29 2017
Looping in troopers; these failures do not appear to be flaky anymore. :(
,
Jun 29 2017
This is failing on Windows try bots and preventing all patches from landing. Example build: http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_ng/builds/479013
,
Jun 29 2017
It looks goma cache contamination. We'll rollback our client soon :/
,
Jun 29 2017
For example: https://build.chromium.org/p/chromium/builders/Win/builds/56585 compile confirm no-op log is here: https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium%2FWin%2F56585%2F%2B%2Frecipes%2Fsteps%2Fcompile_confirm_no-op%2F0%2Fstdout We can see a string `crwingoma`, which is goma canary builder name. This should not been shown here. We're changed cache key generation algorithm in goma release 132, and after goma roll, Win builder started failing. Maybe other builders' result are wrongly used here.
,
Jun 29 2017
Issue 737917 has been merged into this issue.
,
Jun 29 2017
We've rolled back to version 131. Some of goma cache items (DepsCache) form was changed in version 132. However version 131 does not accept that well, so we have needed to purge the cache. Workaround was in Issue 737896. I believe everything is ok now. If not please let me know...
,
Jun 29 2017
Changing to fixed. Please change back if there is still an issue.
,
Jun 30 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/goma/client/+/51f2af255eccc557e117137b19809224529810a6 commit 51f2af255eccc557e117137b19809224529810a6 Author: Takuto Ikuta <tikuta@google.com> Date: Fri Jun 30 06:13:04 2017 |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by benwells@chromium.org
, Jun 29 2017