1366112 - 6,600 instances of "unable to create file '/home/worker/.cache/dconf/user': Permission denied. dconf will not work properly.: 'glib warning'" emitted from toolkit/xre/nsSigHandlers.cpp during linux64 debug testing

Status: RESOLVED WORKSFORME

(Toolkit :: Startup and Profile System, defect, P3)

Description

[Eric Rahm [:erahm]]

> 6562 WARNING: unable to create file '/home/worker/.cache/dconf/user': Permission denied.  dconf will not work properly.: 'glib warning', file toolkit/xre/nsSigHandlers.cpp, line 142

This warning [1] shows up in the following test suites:

>   1920 - test-linux64/debug-mochitest-browser-chrome-e10s-15 bc15
>    702 - test-linux64/debug-reftest-e10s-6 R6
>    702 - test-linux64/debug-reftest-no-accel-e10s-6 Ru6
>    438 - test-linux64/debug-reftest-no-accel-e10s-1 Ru1
>    438 - test-linux64/debug-reftest-e10s-1 R1
>    385 - test-linux64/debug-crashtest-e10s C
>    354 - test-linux64/debug-reftest-e10s-4 R4
>    354 - test-linux64/debug-reftest-no-accel-e10s-4 Ru4
>    312 - test-linux64/debug-mochitest-browser-chrome-e10s-1 bc1
>    228 - test-linux64/debug-reftest-e10s-3 R3
>    228 - test-linux64/debug-reftest-no-accel-e10s-3 Ru3
>    114 - test-linux64/debug-reftest-e10s-5 R5
>    114 - test-linux64/debug-reftest-no-accel-e10s-5 Ru5
>     48 - test-linux64/debug-mochitest-browser-chrome-e10s-10 bc10
>     42 - test-linux64/debug-reftest-no-accel-e10s-8 Ru8
>     42 - test-linux64/debug-reftest-e10s-8 R8
>     42 - test-linux64/debug-mochitest-clipboard-e10s cl
>     30 - test-linux64/debug-mochitest-browser-chrome-e10s-6 bc6
>     30 - test-linux64/debug-reftest-e10s-2 R2
>     30 - test-linux64/debug-reftest-no-accel-e10s-2 Ru2
>      6 - test-linux64/debug-mochitest-browser-chrome-e10s-3 bc3
>      3 - test-linux64/debug-mochitest-browser-chrome-e10s-4 bc4

It shows up in 280 tests. A few of the most prevalent:

>   1896 - [e10s] browser/base/content/test/general/browser_bug590206.js
>    312 - [e10s] dom/xhr/tests/browser_blobFromFile.js
>     42 - [e10s] toolkit/components/passwordmgr/test/browser/browser_context_menu.js
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/printing/960822.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/printing/960822-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/bugs/409659-1a.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/bugs/409659-1-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/bugs/243519-6.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/bugs/243519-6-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/w3c-css/submitted/css21/pagination/moz-css21-rowgroup-page-break-inside-avoid-5.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/w3c-css/submitted/css21/pagination/moz-css21-rowgroup-page-break-inside-avoid-5-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/css-animations/print-no-animations.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/css-animations/print-no-animations-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/printing/318022-1.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/printing/318022-1-ref.html
>     24 - [e10s] file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/w3c-css/submitted/css21/pagination/moz-css21-float-page-break-inside-avoid-9.html == file:///home/worker/workspace/build/tests/reftest/tests/layout/reftests/w3c-css/submitted/css21/pagination/moz-css21-float-page-break-inside-avoid-9-ref.html

[1] https://hg.mozilla.org/mozilla-central/annotate/cc65f9233e5b/toolkit/xre/nsSigHandlers.cpp#l142

Comment 1

[Eric Rahm [:erahm]]

Bisection points to bug 1355984: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=a91f922e14ad404d73b5a9b4a4349c3b97a7d04e&tochange=cf7f6cdfaaf2a59f932ea104847eae915b29666b

I'm somewhat skeptical of that, but Ryan can you take a quick look?

Comment 2

[Eric Rahm [:erahm]]

Hmmm seems to be a bug in mozregression, that build is definitely good.

Comment 3

[Eric Rahm [:erahm]]

The range on central seems more legit:

https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=3243c8fc3ce7831dda843b60d6bb2d7e4acf1fd4&tochange=198effec6e11460e977f13e9cb83050a21a64d27

It looks like there are a couple of merges in there, that probably tripped things up.

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/fef30348e68c ?

Comment 5

[Andrew McCreight [:mccr8]]

I don't see any similar warnings in the top 100.