1166964 - Disable Breakpad INFO logging

Status: VERIFIED FIXED

(Toolkit :: Crash Reporting, defect)

Description

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Breakpad's INFO logging level prints lots of:

No module at 0x0000 

messages when the stack walker gets invoked, possibly via the hang monitor.

These messages don't seem offer any value, so it would be good to hide them so more important console logs are easy to find.

Comment 1

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Ted, is there an approach that you'd recommend here?  I could add a BPLOG_INFO_DISABLED define or something.

Comment 2

[(not currently active) Ted Mielczarek]

What I *really* want is to rip the Breakpad stack walking code out of the client (bug 1069558), but you should be able to do something like what Breakpad's Mac unit tests do:
https://code.google.com/p/google-breakpad/source/browse/trunk/src/client/mac/Breakpad.xcodeproj/project.pbxproj#2249
https://code.google.com/p/google-breakpad/source/browse/trunk/src/client/mac/tests/testlogging.h
https://code.google.com/p/google-breakpad/source/browse/trunk/src/client/mac/tests/exception_handler_test.cc#43

You probably want to use something that just dumps the logs on the floor though instead of a stringstream like that.

Comment 3

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Attachment: MozReview Request: bz://1166964/jryans

/r/9161 - Bug 1166964 - Disable Breakpad INFO logging. r=ted

Pull down this commit:

hg pull -r be0c80e8f884fd5bcd5ba021bdb518df2de6eb1b https://reviewboard-hg.mozilla.org/gecko/

Comment 4

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Try: https://treeherder.mozilla.org/#/jobs?repo=try&revision=cd01334d9e44

Comment 5

[(not currently active) Ted Mielczarek]

https://reviewboard.mozilla.org/r/9161/#review9101

Ship It!

Comment 6

[(not currently active) Ted Mielczarek]

Thanks for the patch!

Comment 7

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Ryan, when do you think you can get your patch landed?

Comment 8

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to Henrik Skupin (:whimboo) from comment #7)
> Ryan, when do you think you can get your patch landed?

Hopefully soon!  I had to fix a Mac build issue, so I'll request another review for the updated version now.

Comment 9

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Fixed build issue with Mac breakpad client:

https://treeherder.mozilla.org/#/jobs?repo=try&revision=23142792a8fa

Comment 10

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Attachment: MozReview Request: Bug 1166964 - Disable Breakpad INFO logging. r=ted

Bug 1166964 - Disable Breakpad INFO logging. r=ted

Comment 11

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Ted, the only additional change here is adding USE_LIBS: "breakpad_logging" to toolkit/crashreporter/client/moz.build for the client to compile (on Try it only mattered for Mac, so I only added it there).

I wasn't sure if it was required to be a static lib like the others, so another review seemed best.

Comment 12

[(not currently active) Ted Mielczarek]

Comment on attachment 8616756 [details]
MozReview Request: Bug 1166964 - Disable Breakpad INFO logging. r=ted

https://reviewboard.mozilla.org/r/9161/#review9225

Ship It!

Comment 13

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/51e393cd5b5a

Comment 14

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/51e393cd5b5a

Comment 15

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Thanks for getting it fixed. Logs are much cleaner now! Verified with latest Nightly builds on all 3 major platforms.

Comment 16

[jgotts]

This debug logging was not present in 38.0.1esr but it showed up in 38.1.0esr.

Can you guys generate a 38.1.1esr with it turned off?

Thanks!

John

Comment 17

[(not currently active) Ted Mielczarek]

This is not the sort of thing we'd spin a new build for. I'm not sure why it would show up in 38.1 when it wasn't present in 38.0, I suspect you just hit a condition in 38.1 that manifested this bug. This is fixed for Firefox 41, it's unlikely that we'd backport this patch to any earlier release.

Comment 18

[jgotts]

You are right. The bug for our situation is triggered when moving from some version of Firefox to 38.1.0 ESR.

Now that the profile that I'm using has been moved to 38.1.0ESR, there is no longer any of this debugging. If it only happens when changing I change versions then the issue is minor.

I'm looking forward to the release of 45 ESR. Have a nice day!

John