New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 608446 link

Starred by 2 users

Issue metadata

Status: WontFix
Owner: ----
Closed: Jun 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug

Blocked on:
issue 602778



Sign in to add a comment

"update_scripts" is flaky

Project Member Reported by chromium...@appspot.gserviceaccount.com, May 2 2016

Issue description

"update_scripts" is flaky.

This issue was created automatically by the chromium-try-flakes app. Please find the right owner to fix the respective test/step and assign this issue to them. If the step/test is not infrastructure-related (e.g. flaky test), please add Sheriff-Chromium label and change issue status to Untriaged. When done, please remove the issue from Trooper Bug Queue by removing the Infra-Troopers label.

We have detected 3 recent flakes. List of all flakes can be found at https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww.

Flaky tests should be disabled within 30 minutes unless culprit CL is found and reverted. Please see more details here: https://sites.google.com/a/chromium.org/dev/developers/tree-sheriffs/sheriffing-bug-queues#triaging-auto-filed-flakiness-bugs

This flaky test/step was previously tracked in  issue 606486 .
 

Comment 1 by smut@chromium.org, May 2 2016

Cc: hinoka@chromium.org
Components: -Tests>Flaky Infra>Labs
Labels: OS-Windows
IOError: [Errno 28] No space left on device

Please clean up disk space on the vms backing this builder:
https://build.chromium.org/p/tryserver.chromium.win/builders/win_optional_gpu_tests_rel
Owner: pschmidt@chromium.org
Status: Assigned (was: Untriaged)
Cc: pgervais@chromium.org
Components: -Infra>Labs Infra
Owner: ----
Status: Untriaged (was: Assigned)
Seems to be that these small pool of vm's have become a builder dumping ground.  There's too many assigned.  I've deleted the win & win.chrome build directories on vm{181,185}-m4 which are the current full slaves so as to provide temporary relief.

Marking this back to untriaged.

win_chrome_official
win_chromium_compile_rel_ng
win_chromium_dbg_ng
win_chromium_syzyasan_rel
win_chromium_variable
win_chromium_variable_archive
win_chromium_variable_chrome
win_chromium_variable_gn
win_chromium_variable_webkit_builder
win_chromium_variable_webkit_layout
win_chromium_variable_webkit_win7_builder
win_chromium_variable_webkit_win7_builder_x64
win_nacl_sdk
win_nacl_sdk_build
win_optional_gpu_tests_rel
vm187-m4 and vm192-m4 will be in the same situation soon if nothing is done.

There are (at least) two pools in this situation:

master.tryserver.chromium.mac:ios_cq_dbg
master.tryserver.chromium.win:win_optional_chromium

An idea to prevent this would be to add a presubmit check that would reject pools with more than, say, five builders.

Status: Available (was: Untriaged)
Related: http://crbug.com/608554
Issue 608554 has been merged into this issue.
You didn't hear this from me but I can increase the size of the disk on these to provide some breathing room until a permanent fix to this builder pool problem is fixed.
The problem will come back in no time, and on other pools as well. We have to fix the issue for real. 

Filed http://crbug.com/608565 to track that.
pschmidt: if you have time to upgrade disks, feel free to chip away at http://crbug.com/605604
Upgraded disks on vm{181,185..187,192}-m4.  This should provide some head room until 608565 is addressed.
Issue 608855 has been merged into this issue.
I'm seeing several bots running out of space for similar reasons; vm673-m4, vm900-m4, vm898-m4, vm1026-m4, vm676-m4, vm675-m4, build86-a4, build87-a4, vm899-m4
Cc: smut@chromium.org
smut, can you fix the pool for tryserver.chromium.mac? Are you the best person to do that?

Comment 14 by smut@chromium.org, May 3 2016

Blockedon: 602778
Once the blocking bug is resolved, anyone will be able to rm -rf /b/build/slave/ios_* to free up disk space on the iOS try bot VMs.
I take it we can silence the alerts on vm1026-m4 in the meantime?

Comment 16 by dnj@google.com, May 9 2016

Labels: -Pri-1 -Infra-Troopers Pri-2
Owner: pgervais@chromium.org
Assigning to pgervais@ since he seems to have the most insight here. Please re-add troopers if there is anything trooper can do to help.
Project Member

Comment 17 by chromium...@appspot.gserviceaccount.com, May 25 2016

Detected 3 new flakes for test/step "update_scripts". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Trooper Bug Queue (unless already there).
Project Member

Comment 18 by chromium...@appspot.gserviceaccount.com, Jun 2 2016

Detected 3 new flakes for test/step "update_scripts". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww. This message was posted automatically by the chromium-try-flakes app.
Owner: ----
Project Member

Comment 20 by chromium...@appspot.gserviceaccount.com, Jun 9 2016

Labels: Infra-Troopers
Detected 3 new flakes for test/step "update_scripts". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Trooper Bug Queue (unless already there).
Project Member

Comment 21 by chromium...@appspot.gserviceaccount.com, Jun 9 2016

Detected 3 new flakes for test/step "update_scripts". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww. This message was posted automatically by the chromium-try-flakes app.
Project Member

Comment 22 by chromium...@appspot.gserviceaccount.com, Jun 9 2016

Detected 3 new flakes for test/step "update_scripts". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyGQsSBUZsYWtlIg51cGRhdGVfc2NyaXB0cww. This message was posted automatically by the chromium-try-flakes app.
Status: WontFix (was: Available)
All recent flakes don't seem to be disk space issues. It doesn't seem to prefer one OS over another and they all fail due to no-output timeouts. Not sure what would cause update_scripts to hang (some git lock file hanging around?) but it happens so rarely that I really don't think it's worth spending time investigating.

Sign in to add a comment