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

Issue 667151 link

Starred by 3 users

Issue metadata

Status: WontFix
Owner: ----
Closed: Feb 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 1
Type: Bug



Sign in to add a comment

compile failure on chromium/Mac ("a no-op build wasn't a no-op")

Project Member Reported by mgiuca@chromium.org, Nov 21 2016

Issue description

compile failure on chromium/Mac

Type: build-failure

Builders failed on: 
- Mac: 
  https://build.chromium.org/p/chromium/builders/Mac

https://build.chromium.org/p/chromium/builders/Mac/builds/21269

Relevant part of the log:
/b/rr/tmpNJ4fpA/rw/checkout/scripts/slave/.recipe_deps/depot_tools/ninja -w dupbuild=err -C /b/c/b/mac_ng/src/out/Release all chrome -j80 -d explain -n
ninja: Entering directory `/b/c/b/mac_ng/src/out/Release'
ninja explain: output test_data/chrome/safe_browsing_dmg/mach_o_in_dmg.dmg doesn't exist
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_GPTSPUD.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_NONE.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_SPUD.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_GPTSPUD.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_NONE.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_SPUD.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/hfs_plus.img is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/hfsx_case_sensitive.img is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/mach_o_in_dmg.dmg is dirty
ninja explain: obj/chrome/test/mac_safe_browsing_test_data.stamp is dirty
ninja explain: obj/chrome/test/mac_safe_browsing_test_data.stamp is dirty
[1/3] ACTION //chrome/test:mac_safe_browsing_test_data(//build/toolchain/mac:clang_x64)
[2/3] STAMP obj/chrome/test/mac_safe_browsing_test_data.stamp
[3/3] STAMP obj/chrome/test/unit_tests.inputdeps.stamp
<Thread(Thread-1, started 4438282240)> ProcessRead: proc.stdout finished.
<Thread(Thread-1, started 4438282240)> ProcessRead: cleaning up.
<Thread(Thread-2, started daemon 4442488832)> TimedFlush: Finished
<Thread(Thread-1, started 4438282240)> ProcessRead: finished.
Failing build because ninja reported work to do.
This means that after completing a compile, another was run and
it resulted in still having work to do (that is, a no-op build
wasn't a no-op). Consult the first "ninja explain:" line for a
likely culprit.

It looks like mach_o_in_dmg.dmg went missing in between the first and second builds, resulting in it having to be rebuilt a second time? I'm not sure what's going on but this closed the tree. I'm going to re-open and see if a new build fixes it (because the committed CL doesn't seem to have anything to do with this failure).
 

Comment 1 by mgiuca@chromium.org, Nov 21 2016

Cc: jparent@chromium.org tansell@chromium.org
Owner: estaab@chromium.org
+estaab suggested by tansell.

Tim mentioned this has failed on a bunch of bots flakily (don't know which though...).
It happened again: https://build.chromium.org/p/chromium/builders/Mac/builds/21274

The blame range is a single patch that adds a comment.

Comment 3 by mgiuca@chromium.org, Nov 21 2016

Ack.

I don't know how to fix it but tansell@ mentioned there's a flag to make this not an error. Maybe we can do that in the short term.

Comment 4 by mgiuca@chromium.org, Nov 21 2016

Labels: -Pri-2 Pri-1
Bumping up to pri 1 since we've seen two tree closers in one day.
Cc: dpranke@chromium.org
Components: Infra>Client>Chrome
Owner: andyb...@chromium.org
Sorry, I missed this. This falls in a grey area between infra and dev tools so perhaps Andy or Dirk can help figure out next steps.
Sheriff here: I could not interpret this error and it took me a while to even find that section in the logs. Is this an infra issue? Any suggestions on how to better direct it?
Components: -Infra>Client>Chrome -Infra>Platform>Buildbot Build
Owner: ----
I'd consider this to be a dev-side problem. Most likely when this happens there are missing build dependencies. We have this set to be checked and be a build error on many but not all builders. 

I believe there may be a known case at the moment where this is still broken on Android; if you're seeing this elsewhere at the moment (e.g., on Mac like the subject says), that'd be a new bug.
This is definitely happening on mac every now and then. It doesn't seem to be reproducible (IE rerunning the same change doesn't cause the same failure).
Components: Services>Safebrowsing
Labels: Hotlist-Sheriff-Chromium
This is still happening recently and frequently.

https://build.chromium.org/p/chromium/builders/Mac?numbuilds=200

The last one is https://build.chromium.org/p/chromium/builders/Mac/builds/23536

Adding Safebrowsing component to have this investigated by SafeBrowsing folks as the log is listing files with "safe_browsing" in their file names.
Cc: rhalavati@chromium.org
+cc rhalavati@ could this be related to  issue 690296  ?
Don't think so. There is no trace of cloud_policy.proto there and ninja explain says this file doesn't exist which doesn't seem related to that proto: test_data/chrome/safe_browsing_dmg/hfsx_case_sensitive.img 

Comment 12 by perkj@chromium.org, Feb 15 2017

Labels: -Hotlist-Sheriff-Chromium

Comment 13 by perkj@chromium.org, Feb 16 2017

This does happen from time to time on Mac builder:
https://build.chromium.org/p/chromium/builders/Mac


ninja explain: output test_data/chrome/safe_browsing_dmg/hfsx_case_sensitive.img doesn't exist
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDBZ_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDCO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDRW_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_GPTSPUD.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_NONE.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDSP_SPUD.sparseimage is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_GPTSPUD.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_NONE.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDTO_SPUD.cdr is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UDZO_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_GPTSPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_NONE.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/dmg_UFBI_SPUD.dmg is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/hfs_plus.img is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/hfsx_case_sensitive.img is dirty
ninja explain: test_data/chrome/safe_browsing_dmg/mach_o_in_dmg.dmg is dirty
ninja explain: obj/chrome/test/mac_safe_browsing_test_data.stamp is dirty
ninja explain: obj/chrome/test/mac_safe_browsing_test_data.stamp is dirty
@@@STEP_TEXT@This should have been a no-op, but it wasn't.@@@
@@@STEP_FAILURE@@@
I don't think this is necessarily mac specific? I just had a similar random failure on a windows build - tryserver.chromium.win/win_chromium_rel_ng/382022

https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Ftryserver.chromium.win%2Fwin_chromium_rel_ng%2F382022%2F%2B%2Frecipes%2Fsteps%2Fcompile__with_patch__confirm_no-op%2F0%2Fstdout

---------------------

ninja explain: output obj/chrome/browser/test_support_ui/password_manager_test_base.obj older than most recent input gen/third_party/WebKit/public/web/window_features.mojom.h (508747538 vs 508747567)
ninja explain: obj/chrome/browser/test_support_ui/password_manager_test_base.obj is dirty
ninja explain: output obj/chrome/browser/test_support_ui/web_ui_test_handler.obj older than most recent input gen/third_party/WebKit/public/web/window_features.mojom.h (508747538 vs 508747567)
ninja explain: obj/chrome/browser/test_support_ui/web_ui_test_handler.obj is dirty
ninja explain: obj/chrome/browser/test_support_ui.lib is dirty
ninja explain: browser_tests.exe is dirty
ninja explain: browser_tests.exe is dirty
ninja explain: obj/chrome/test/browser_tests_run.stamp is dirty
ninja explain: obj/chrome/browser/test_support_ui.lib is dirty
ninja explain: chrome_app_unittests.exe is dirty
ninja explain: chrome_app_unittests.exe is dirty
ninja explain: obj/chrome/test/chrome_app_unittests_run.stamp is dirty
ninja explain: obj/chrome/browser/test_support_ui.lib is dirty
ninja explain: interactive_ui_tests.exe is dirty
ninja explain: interactive_ui_tests.exe is dirty
ninja explain: obj/chrome/test/interactive_ui_tests_run.stamp is dirty
ninja explain: obj/chrome/browser/test_support_ui.lib is dirty
ninja explain: sync_integration_tests.exe is dirty
ninja explain: sync_integration_tests.exe is dirty
ninja explain: obj/chrome/test/sync_integration_tests_run.stamp is dirty
ninja explain: browser_tests.exe is dirty
ninja explain: obj/chrome/test/tab_capture_end2end_tests_run.stamp is dirty
ninja explain: obj/chrome/browser/test_support_ui.lib is dirty
ninja explain: unit_tests.exe is dirty
ninja explain: unit_tests.exe is dirty
ninja explain: obj/chrome/test/unit_tests_run.stamp is dirty
---------------------------

Comment 15 by perkj@chromium.org, Feb 16 2017

Cc: rsesek@chromium.org
Here is a similar no-op failure on a Linux builder...

https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_rel_ng/builds/391551/steps/compile%20%28with%20patch%29%20confirm%20no-op/logs/stdio
------------------------
ninja explain: output obj/pdf/pdf_unittests_run.stamp older than most recent input pdf_unittests (1487230486 vs 1487230487)
ninja explain: obj/pdf/pdf_unittests_run.stamp is dirty
ninja explain: obj/pdf/pdf_unittests_run.stamp is dirty
@@@STEP_TEXT@This should have been a no-op, but it wasn't.@@@
@@@STEP_FAILURE@@@
------------------------
The initial report is a dupe of  issue 602169  and just means macOS is in a bad state and the bot should be rebooted.

But there are a bunch of other unrelated issues reported here, so I'm not sure how useful this bug is.

Comment 18 by vakh@chromium.org, Feb 24 2017

Status: WontFix (was: Available)
It seems like this problem got fixed somehow and is not happening anymore since everything on https://build.chromium.org/p/chromium/builders/Mac?numbuilds=200 is green.

Sign in to add a comment