964351 - Spike in Aurora 28.0a2 as of 2014-01-26 with TppTimerpExecuteCallback

Status: RESOLVED WORKSFORME

(Core :: General, defect)

Description

[u279076]

This bug was filed from the Socorro interface and is 
report bp-4a0bc901-1fdb-4654-9954-c61dc2140127.
=============================================================
0 		@0x65fa2140 	
1 	ntdll.dll 	TppTimerpExecuteCallback 	
2 	ntdll.dll 	ntdll.dll@0x1f6c0 	
3 	kernel32.dll 	BaseThreadInitThunk 	
4 	ntdll.dll 	__RtlUserThreadStart 	
5 	ntdll.dll 	_RtlUserThreadStart 	

More reports:
https://crash-stats.mozilla.com/report/list?product=Firefox&signature=TppTimerpExecuteCallback

Current overall ranking:
 * Firefox 29: #143 @ 0.07% (-0.03%)
 * Firefox 28: #25 @ 0.49% (+0.14%)
 * Firefox 27: N/A (volume too low for topcrash reports)
 * Firefox 26: N/A (volume too low for topcrash reports)

Pushlog:
http://hg.mozilla.org/releases/mozilla-aurora/pushloghtml?fromchange=4b8d14a78596&tochange=c40099a42c1f
=============================================================

This signature has spiked on Aurora 28.0a2 starting on 2014-01-26 with 408 crashes per 1MM ADU (it's been in the 20-78 range over the previous week). It's not at a volume that we'd traditionally give topcrash designation to but it is a startup crash and the spike on Aurora seems suspect. Unfortunately I do not see any useful URLs, correlations, or comments.

Comment 1

[(Away)]

KaiRo, do the top crash rankings count duplicate reports? There is one Aurora user who sent about 30 duplicates. This might not actually be a true spike...

Comment 2

[Robert Kaiser]

(In reply to David Major [:dmajor] from comment #1)
> KaiRo, do the top crash rankings count duplicate reports? There is one
> Aurora user who sent about 30 duplicates. This might not actually be a true
> spike...

Yes, we do not exclude duplicates in any reports as we don't completely trust the mechanism that marks them as dupes at this point. The long-term plan is to detect *real* dupes in a different way and exclude them, and then remove this dupe-marking completely.

Comment 3

[(Away)]

(In reply to Robert Kaiser (:[e-mail]) from comment #2)
Ok, thanks. Sounds like we shouldn't worry too much about this one.

Comment 4

[Lukas Blakk [:lsblakk] use ?needinfo]

not tracking as per comment 3

Comment 5

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

In 47.0b1 so far this is the top crash that happens with install age < 120 (seconds, I think).  So it seems most frequently hit right after installing.  (The question is whether it happens once, or it prevents users from coming back.)

It seems to be all different users in 47.0b1.


Looking at all the crashes with this signature since September, the platform distribution seems a bit unusual:

Rank 	Platform pretty version 	Count 	%
1 	Windows 8.1 	1171 	59.17 %
2 	Windows 7 	304 	15.36 %
3 	Windows 10 	178 	8.99 %
(other versions much smaller numbers)

Comment 6

[Ritu Kothari (:ritu) (Inactive, please n-i to RyanVM, jcristau, or pascal)]

Tracked for 47+ as it's a top crasher.

Comment 7

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Though another note is that the table of CPUs affected for all of the TppTimerpExecuteCallback crashes so far this year is:

Rank 	Cpu info 	Count 	%
1 	AuthenticAMD family 20 model 2 stepping 0 | 2 	40 	93.02 %
2 	AuthenticAMD family 20 model 1 stepping 0 | 2 	2 	4.65 %
3 	AuthenticAMD family 20 model 2 stepping 0 | 1 	1 	2.33 %

which makes me think it's related to bug 772330.

Comment 8

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Er, I had a filter on that search and was looking at a subset -- and now I'm not sure which subset.  That doesn't hold true for the larger sample.

Comment 9

[[:philipp]]

all the crashes on 47.0b seem to happen with ATI gpus so far (adaptor vendor id 0x1002).

Comment 10

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

The issue in Firefox 47 is bug 1270686.  Let's keep this bug about whatever the older issue was.

Comment 11

[BugBot [:suhaib / :marco/ :calixte]]

Crash volume for signature 'TppTimerpExecuteCallback':
 - nightly (version 51): 0 crashes from 2016-08-01.
 - aurora  (version 50): 0 crashes from 2016-08-01.
 - beta    (version 49): 75 crashes from 2016-08-02.
 - release (version 48): 18 crashes from 2016-07-25.
 - esr     (version 45): 6 crashes from 2016-05-02.

Crash volume on the last weeks (Week N is from 08-22 to 08-28):
            W. N-1  W. N-2  W. N-3
 - nightly       0       0       0
 - aurora        0       0       0
 - beta         22      29      13
 - release       5       4       3
 - esr           1       1       0

Affected platform: Windows

Crash rank on the last 7 days:
           Browser   Content     Plugin
 - nightly
 - aurora
 - beta    #910      #1841
 - release #1989
 - esr

Comment 12

[BugBot [:suhaib / :marco/ :calixte]]

Closing because no crashes reported for 12 weeks.