Issue metadata
Sign in to add a comment
|
GPU Linux Ozone Builder flaking on bot_update step on GPU.FYI waterfall |
||||||||||||||||||||||
Issue descriptionPast flaky failures: https://ci.chromium.org/buildbot/chromium.gpu.fyi/GPU%20Linux%20Ozone%20Builder/15308 https://ci.chromium.org/buildbot/chromium.gpu.fyi/GPU%20Linux%20Ozone%20Builder/15287 https://ci.chromium.org/buildbot/chromium.gpu.fyi/GPU%20Linux%20Ozone%20Builder/15180 Failures seem to have the form: ===Running git checkout --force 60854de15bfeef95cb803aaab521f89352c07c3d -- === In directory: /b/c/b/GPU_Linux_Ozone_Builder/src fatal: reference is not a tree: 60854de15bfeef95cb803aaab521f89352c07c3d ===Failed in 0.0 mins of git checkout --force 60854de15bfeef95cb803aaab521f89352c07c3d -- === Something failed: git checkout --force 60854de15bfeef95cb803aaab521f89352c07c3d -- failed with code 128 in /b/c/b/GPU_Linux_Ozone_Builder/src.. Ran 35.355741024 seconds past deadline. Aborting. Traceback (most recent call last): File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 1252, in <module> sys.exit(main()) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 1236, in main checkout(options, git_slns, specs, revisions, step_text, shallow) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 1142, in checkout gclient_output = ensure_checkout(**checkout_parameters) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 841, in ensure_checkout cleanup_dir) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 586, in git_checkouts cleanup_dir) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 647, in _git_checkout force_solution_revision(name, url, revisions, sln_dir) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 543, in force_solution_revision git('checkout', '--force', treeish, '--', cwd=cwd) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 209, in git return call(*cmd, **kwargs) File "/b/rr/tmpWt9QJM/rw/checkout/scripts/slave/.recipe_deps/depot_tools/recipes/recipe_modules/bot_update/resources/bot_update.py", line 190, in call code, outval) __main__.SubprocessFailed: git checkout --force 60854de15bfeef95cb803aaab521f89352c07c3d -- failed with code 128 in /b/c/b/GPU_Linux_Ozone_Builder/src. step returned non-zero exit code: 1 Ken/Sunny/Frank, I'm not sure what's going on here. Can one of you help cc the right people?
,
Jan 3 2018
#1: Correct, it isn't the same issue. As mentioned in 798765, that (less than helpful) error message is just an indication that bot_update failed. This looks like it may be either failing to update the bot's cache of chromium/src or it's affected by gerrit replication latency (b/67501786 -> b/35998605).
,
Jan 3 2018
+cc hinoka: any thoughts here?
,
Jan 3 2018
,
Jan 4 2018
#4: thanks, I'd forgotten about that. Curious, though -- why blocking this on that rather than duping it?
,
Jan 4 2018
Only in case we were going to continue to investigate this on the Chromium side, since the other bug's closed. I see that some progress has been made on b/35998605 but these problems are clearly still happening. Do we need to comment on that bug that issues are still being seen? Would you or another Infra team member be willing to do that and then dupe this into Issue 772453 ?
,
Jan 4 2018
agable mentioned on the other bug (https://bugs.chromium.org/p/chromium/issues/detail?id=772453#c9) that there's nothing for us to continue to do. With progress being made but clearly not yet finished on the bug, I'm not sure I see much benefit to prodding that bug.
,
Jan 4 2018
If everyone agrees b/35998605 is responsible, I'm fine with let things sit. There's a lot of people watching that bug because of general Gerrit laggyness, and the infra people are already pretty aware of the need. I'll assign this to myself to watch for the flakes to clear up when the issue is closed, so setting a next action for a month from now.
,
Feb 1 2018
The NextAction date has arrived: 2018-02-01
,
Feb 1 2018
Don't see any flakes in the last 200 builds. Please re-open if flakiness re-occurs. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by kbr@chromium.org
, Jan 3 2018Components: Infra>Client>Chrome