814968 - Something bad happened. It's not you, it's me. https://crash-stats.mozilla.com/report/index/1c26698f-5a42-4eff-911c-748782121124

Status: RESOLVED FIXED

(Socorro :: General, task)

Description

[Fede]

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/17.0 Firefox/17.0
Build ID: 20121119183901

Steps to reproduce:

https://crash-stats.mozilla.com/report/index/1c26698f-5a42-4eff-911c-748782121124


Actual results:

Something bad happened. It's not you, it's me.

Comment 1

[Jesse Ruderman]

I'm hitting what I imagine is the same bug.

https://crash-stats.mozilla.com/report/list?range_value=7&range_unit=days&date=2012-12-03&signature=F2102588022______________________________________________________________________________________________________&version=Firefox%3A17.0.1

Links to

https://crash-stats.mozilla.com/report/index/b74aa9df-9d79-46fe-9826-9fc5c2121201

and other crash reports that fail to load, with the message

"Something bad happened. It's not you, it's me.  Please file an issue in Bugzilla describing what happened, and please include the URL for this page."

Comment 2

[Laura Thomson :laura]

Some crash reports load (e.g. https://crash-stats.mozilla.com/report/index/d880c546-9576-4652-87dc-011892121205), so it is not system wide.

Next step: checking logs.

Comment 3

[Robert Helmer [:rhelmer]]

The frontend is throwing an exception here:

https://github.com/rhelmer/socorro/blob/master/webapp-php/application/views/report/index.php#L360

Comment 4

[Robert Helmer [:rhelmer]]

(In reply to Robert Helmer [:rhelmer] from comment #3)
> The frontend is throwing an exception here:
> 
> https://github.com/rhelmer/socorro/blob/master/webapp-php/application/views/
> report/index.php#L360

2012-12-06 04:06:14 +00:00 --- error: [5xx Error] File: application/views/report/index.php; Line: 360; Message: Undefined offset: 0

Comment 5

[Robert Helmer [:rhelmer]]

The processed JSON for this crash is:
https://crash-stats.mozilla.com/dumps/1c26698f-5a42-4eff-911c-748782121124.jsonz

Comment 6

[Robert Helmer [:rhelmer]]

From irc:

9:35 < ted> there was a breakpad change that let us get info out for some 
             threads even if we couldn't read some of the other threads
09:36 < ted> this was 
             http://code.google.com/p/google-breakpad/source/detail?r=1077

Comment 7

[Robert Helmer [:rhelmer]]

(In reply to Robert Helmer [:rhelmer] from comment #6)
> From irc:
> 
> 9:35 < ted> there was a breakpad change that let us get info out for some 
>              threads even if we couldn't read some of the other threads
> 09:36 < ted> this was 
>              http://code.google.com/p/google-breakpad/source/detail?r=1077

This is not a regression, since before we would have shown a completely blank report.. we are working on a fix for it regardless, should be simple. The view code assumes that it can there is a thread from 0 to the maximum thread number and tries to iterate over each, we can just guard against that case now.

Comment 8

[Robert Helmer [:rhelmer]]

Confirmed that this already works as expected in the django version of crash-stats (VPN only):
http://crash-stats-new-dev.allizom.org/report/index/8aa61ff4-05c7-460f-a7c3-5aaf32121206

Comment 9

[Robert Helmer [:rhelmer]]

schalk r? https://github.com/mozilla/socorro/pull/988

Comment 10

[[github robot]]

Commits pushed to master at https://github.com/mozilla/socorro

https://github.com/mozilla/socorro/commit/006ac2df94e13a1482f419903fa1e66ad6596535
bug 814968 - processed crashes may be missing threads, iterate over what is there instead

https://github.com/mozilla/socorro/commit/a64be72b78cde9f474b7a7a7077485982a533c02
Merge pull request #988 from rhelmer/bug814968-partial-thread-dump

bug 814968 - processed crashes may be missing threads, iterate over what...