1401599 - Frequent browser_mixed_content_cert_override.js | application crashed [@ mozilla::net::HttpChannelChild::RecvReportSecurityMessage(nsTString<char16_t> const &,nsTString<char16_t> const &)] (((bool)(!!(!NS_FAILED_impl(rv)))), at HttpChannelChild.cpp:1522)

Status: RESOLVED INCOMPLETE

(Core :: Networking, defect, P5)

Description

[Treeherder Bug Filer]

Filed by: rvandermeulen [at] mozilla.com

https://treeherder.mozilla.org/logviewer.html#?job_id=132235118&repo=mozilla-beta

https://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-beta-win64-debug/mozilla-beta_win10_64-debug_test-mochitest-e10s-browser-chrome-3-bm110-tests1-windows-build16.txt.gz

This is failing frequently on Beta57 on Win10 debug at the moment. These tests were only recently enabled (previously they ran on Win8), so I'm not sure how old the problem really is.

Comment 1

[Ryan VanderMeulen [:RyanVM]]

Christoph, any ideas on what might be happening here?

Comment 2

[Shih-Chiang Chien [:schien] (UTC+8) (use ni? plz)]

This assertion is hit because fail to format localization string at https://searchfox.org/mozilla-central/rev/2ef8bd8a46a02c68ddbb1d5f25fa254dd7be1fbd/netwerk/protocol/http/HttpBaseChannel.cpp#2552.

Comment 3

[Intermittent Failures Robot]

23 failures in 190 pushes (0.121 failures/push) were associated with this bug yesterday.    

Repository breakdown:
* mozilla-beta: 21
* try: 2

Platform breakdown:
* windows10-64: 23

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1401599&startday=2017-09-20&endday=2017-09-20&tree=all

Comment 4

[Christoph Kerschbaumer [:ckerschb]]

(In reply to Ryan VanderMeulen [:RyanVM] from comment #1)
> Christoph, any ideas on what might be happening here?

Tanvi, Kate, can anyone take a look at this one please?

Comment 5

[Kate McKinley [:kmckinley, :Kate]]

I'm not having any luck reproducing this on treeherder[1]. As Shih-Chiang notes above, it looks like the crash is because we use an assertion when we can not get the correct localization. Since this occurs in the content process, it makes me wonder if this could be a sandboxing issue.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=3df4d011c191c7b06481792041e34c261c8d8463

Comment 6

[Intermittent Failures Robot]

25 failures in 943 pushes (0.027 failures/push) were associated with this bug in the last 7 days.    

Repository breakdown:
* mozilla-beta: 23
* try: 2

Platform breakdown:
* windows10-64: 25

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1401599&startday=2017-09-18&endday=2017-09-24&tree=all

Comment 7

[Bob Owen (:bobowen)]

(In reply to Kate McKinley [:kmckinley, :Kate] from comment #5)
> I'm not having any luck reproducing this on treeherder[1]. As Shih-Chiang
> notes above, it looks like the crash is because we use an assertion when we
> can not get the correct localization. Since this occurs in the content
> process, it makes me wonder if this could be a sandboxing issue.
> 
> https://treeherder.mozilla.org/#/
> jobs?repo=try&revision=3df4d011c191c7b06481792041e34c261c8d8463

Sorry about the delay here.

If this were sandboxing I'd guess this would be failing all the time (although in theory it could be a timing thing if it was supposed to happen before we lower the sandbox).
The content process should have read access to the string bundles location anyway.

Gabor - you mentioned something about issues with loading string bundles in our meeting (bug 1385249), could this be a related issue?
The root cause mentioned in comment 2 is because of a failure with nsContentUtils::GetLocalizedString.
Looks like there might be non-fatal localization failures a bit further up the log as well.

Comment 8

[Gabor Krizsanits (INACTIVE)]

(In reply to Bob Owen (:bobowen) from comment #7)
> Gabor - you mentioned something about issues with loading string bundles in
> our meeting (bug 1385249), could this be a related issue?
> The root cause mentioned in comment 2 is because of a failure with
> nsContentUtils::GetLocalizedString.
> Looks like there might be non-fatal localization failures a bit further up
> the log as well.

Yes that I'm pretty sure that it's the same problem.

Comment 9

[Firefox Bug Husbandry Bot]

https://wiki.mozilla.org/Bugmasters#Intermittent_Test_Failure_Cleanup