Closed Bug 1243098 Opened 9 years ago Closed 9 years ago

Firefox hangs or crashes on startup

Categories

(Core :: IPC, defect)

Product:
Core
Component:
IPC
Version:
45 Branch
Platform:
x86
Windows 7
Type:
defect
Priority:
Not set
Severity:
critical

Tracking

()

Status:
RESOLVED FIXED
Tracking Flags:
Tracking Status
firefox44 + fixed
firefox45 + fixed
firefox46 + fixed
firefox47 + fixed

People

(Reporter: dsmith, Unassigned)

References

Details

(Keywords: crash, regression, Whiteboard: dom-triaged [44dll])

Crash Data

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Firefox/45.0
Build ID: 20160125004010

Steps to reproduce:

Attempting to start browser, either from scratch, or restarting after an update becomes available, or restarting after enabling/disabling certain addons.

Attempts to correct the issue include:
1) Complete uninstall/reinstall of firefox
2) Refresh of profile
3) Disabling all addons in profile
4) Use a complete, brand-new temp profile

The fewer addons installed, the more likely the program is to succeed on restart, so I typically do updates from a temp profile.

No new device drivers have been added recently.


Actual results:

One of several issues occurred.  Memory noted is Working Set as listed in the Resource Monitor program.

1) Program gets to about 4 MB, then self-quits
2) Program gets to about 15 MB, then self-quits
3) Program gets to about 15 MB, then hangs.  Requires forcibly ending the process.
4) Program gets to somewhere between 20 MB and 60 MB, then hangs.  Requires forcibly ending the process.

A typical restart requires 4 to 6 attempts of the program self-quitting or hanging, before the browser can manage to completely load and run.

When this happens during a daily update, the update is not installed.  The next check of the About box will cause a new download of the update, and prompt for another restart attempt.

This rarely causes crashes, preferring to either hang or quit.  However I did have a period where I was able to get a crash report sent.  Here are the crashes that occurred:

https://crash-stats.mozilla.com/report/index/0022825d-497a-48b7-87d0-82f082160122
https://crash-stats.mozilla.com/report/index/7c2b971b-a67f-4b45-a782-7b3c02160122
https://crash-stats.mozilla.com/report/index/2d0d34b1-05dc-4627-bf72-652db2160122
https://crash-stats.mozilla.com/report/index/e7c6065c-ca31-40f0-9b5e-60a972160122
https://crash-stats.mozilla.com/report/index/9fd85d6f-dadc-47ec-b31e-f96d22160123

They all seem to relate to one of two functions: RtlpLowFragHeapAllocFromContext, and RtlpLowFragHeapFree.

However, for those five crashes, I've probably had a hundred instances of hangs and quits that were never reported.


Some of the original failures occurred with electrolysis enabled, but recently electrolysis has remained off (and the hangs and quits continue), because I can no longer reset the accessibility flags that forcibly disable electrolysis. Resetting and then restarting the browser (even if I can manage an immediate restart) has them back in the state where electrolysis is not allowed.


Expected results:

Firefox should start normally.
Thought it would be automatically included, but: OS/Platform is Windows 7, 32-bit x86.
OS: Unspecified → Windows 7
Hardware: Unspecified → x86
Severity: normal → critical
Crash Signature: [@ RtlpLowFragHeapAllocFromContext | RtlAllocateHeap | _calloc_impl | _calloc_impl | msvcr120.dll@0x21bbf | __getlocaleinfo ] [@ RtlpLowFragHeapFree | RtlFreeHeap | _flsbuf_s ] [@ RtlpLowFragHeapAllocFromContext | RtlAllocateHeap | _calloc_impl | _callo…
Component: Untriaged → General
Keywords: crash
Aaron, do you know (someone else who can tell us) what's going here / where this belongs? From the stack top function name, it seems that maybe it's a low memory situation, but at least for the crashes in comment #0, it seems there's plenty of available physical memory never mind everything else.
Component: General → Untriaged
Flags: needinfo?(aklotz)
Just a note that the problem is still there after updating to version 46.

On memory usage, I generally only have about half of the 12 GB on my system in use (give or take a GB or so).  Most of the remainder is marked as in 'standby' (ie: cached data or inactive code), according to Resource Monitor, but I'm not sure if that has any relevance.

I'll see if I can find a build date for when this started happening.
See Also: → 1243765
From what I can tell, January 11 was the last good build.  Starting January 12, restarting the browser has a chance to have the browser hang or self-quit during startup.  I'm generally testing this by enabling/disabling Classic Theme Restorer, as that allows a quick restart, but also just closing and starting the browser from scratch can cause this.

My own first encounter with the issue was the 1/12 to 1/13 update, where the program continually failed to update on restart.  I couldn't find any bugs on it at the time, though, and didn't see anyone else mentioning it or able to confirm it, so left it alone til after I started getting crashes.

Similar bug: Bug 1243765 - This mentions graphics issues, which are part of the stack trace on some of the crashes.  Possibly related to Bug 1218473 from the 1/12 checkins?
(In reply to David Smith from comment #4)
> From what I can tell, January 11 was the last good build.  Starting January
> 12, restarting the browser has a chance to have the browser hang or
> self-quit during startup.

Thanks, that's super helpful. You're using developer edition (based on the UA string of 45 in comment #0, and having recently updated to 46), right? In which case, these are the only things that landed in that timeframe:

http://hg.mozilla.org/releases/mozilla-aurora/pushloghtml?startdate=2016-01-11&enddate=2016-01-12

Does your machine have nvidia hardware and/or drivers (optimus drivers) and/or network management software installed?
Flags: needinfo?(dsmith)
(In reply to :Gijs Kruitbosch from comment #5)
> (In reply to David Smith from comment #4)
> > From what I can tell, January 11 was the last good build.  Starting January
> > 12, restarting the browser has a chance to have the browser hang or
> > self-quit during startup.
> 
> Thanks, that's super helpful. You're using developer edition (based on the
> UA string of 45 in comment #0, and having recently updated to 46), right? In
> which case, these are the only things that landed in that timeframe:
> 
> http://hg.mozilla.org/releases/mozilla-aurora/pushloghtml?startdate=2016-01-
> 11&enddate=2016-01-12
> 
> Does your machine have nvidia hardware and/or drivers (optimus drivers)
> and/or network management software installed?

Yes, I'm using the Developer Edition.

Yes, I have an nVidia graphics card.
Card: GeForce GTX 660
Driver: 359.06 (December 1, 2015)
I don't recognize 'optimus' as a type of driver name for the graphics card.
I don't know of any nVidia network management drivers.  Network port on the computer is based on Realtek.
Flags: needinfo?(dsmith)
(In reply to David Smith from comment #6)
> (In reply to :Gijs Kruitbosch from comment #5)
> > (In reply to David Smith from comment #4)
> > > From what I can tell, January 11 was the last good build.  Starting January
> > > 12, restarting the browser has a chance to have the browser hang or
> > > self-quit during startup.
> > 
> > Thanks, that's super helpful. You're using developer edition (based on the
> > UA string of 45 in comment #0, and having recently updated to 46), right? In
> > which case, these are the only things that landed in that timeframe:
> > 
> > http://hg.mozilla.org/releases/mozilla-aurora/pushloghtml?startdate=2016-01-
> > 11&enddate=2016-01-12
> > 
> > Does your machine have nvidia hardware and/or drivers (optimus drivers)
> > and/or network management software installed?
> 
> Yes, I'm using the Developer Edition.
> 
> Yes, I have an nVidia graphics card.
> Card: GeForce GTX 660
> Driver: 359.06 (December 1, 2015)
> I don't recognize 'optimus' as a type of driver name for the graphics card.
> I don't know of any nVidia network management drivers.  Network port on the
> computer is based on Realtek.

Hm. Interesting. If you run this build: http://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-aurora-win32/1452542313/firefox-45.0a2.en-US.win32.zip - does that behave correctly ? It excludes the fix from bug 1218473 which seems most likely to have something to do with startup crashes/hangs. If not, what about this build: http://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-aurora-win32/1452523410/ ? (from https://hg.mozilla.org/releases/mozilla-aurora/rev/af3852b5e369 ) 

Just trying to narrow down what triggered this...
Flags: needinfo?(dsmith)
I have not been able to get any crashes or hangs with either of the above two builds.
Flags: needinfo?(dsmith)
Aaron, based on the earlier comments, seems that at least here this is directly tied to the landing of https://hg.mozilla.org/releases/mozilla-aurora/rev/7f6dc6f3589b . Could you take a look, please?
Blocks: 1218473
Status: UNCONFIRMED → NEW
Component: Untriaged → IPC
Ever confirmed: true
Keywords: regression
Product: Firefox → Core
(In reply to David Smith from comment #8)
> I have not been able to get any crashes or hangs with either of the above
> two builds.

(In reply to :Gijs Kruitbosch from comment #9)
> Aaron, based on the earlier comments, seems that at least here this is
> directly tied to the landing of
> https://hg.mozilla.org/releases/mozilla-aurora/rev/7f6dc6f3589b . Could you
> take a look, please?

Here's the thing: backing that change out is not an option. I guarantee you that it would make the situation even worse.

This must be an additional problem with injected nvidia code. This will require further investigation.
Flags: needinfo?(aklotz)
(In reply to Aaron Klotz [:aklotz] (please use needinfo) from comment #10)
> Here's the thing: backing that change out is not an option. I guarantee you
> that it would make the situation even worse.
> 
> This must be an additional problem with injected nvidia code. This will
> require further investigation.

Sure. I didn't intend to suggest "back it out", just that it seems we're not quite done... Is there anything David could do to help with that?
Flags: needinfo?(aklotz)
This crash signature and others like it are pointing to corruption in the Windows heap (as opposed to the jemalloc heap). I think that temporarily enabling the heap validation gflags on the firefox process would be the easiest way for us to find out where the corruption first occurs.

David, if I provided you with instructions, would you be comfortable attempting this?
Flags: needinfo?(aklotz) → needinfo?(dsmith)
(dom-noted just so this doesn't turn up in my triage queries...carry on the excellent work :)
Whiteboard: dom-noted
(In reply to Aaron Klotz [:aklotz] (please use needinfo) from comment #12)
> This crash signature and others like it are pointing to corruption in the
> Windows heap (as opposed to the jemalloc heap). I think that temporarily
> enabling the heap validation gflags on the firefox process would be the
> easiest way for us to find out where the corruption first occurs.
> 
> David, if I provided you with instructions, would you be comfortable
> attempting this?

Yes, that shouldn't be a problem.
Flags: needinfo?(dsmith)
I have posted instructions at https://pastebin.mozilla.org/8858430
Feel free to contact me via this bug or email (see my bugzilla profile) if you're experiencing any problems.
Flags: needinfo?(dsmith)
Whiteboard: dom-noted → dom-triaged
Whiteboard: dom-triaged → dom-triaged [44dll]
Tracking for 45+ since this seems to be a regression from 45. 
Sylvestre, adding n-i for you to make sure you see I tracked this for you.
status-firefox45: --- → affected
status-firefox46: --- → affected
status-firefox47: --- → affected
tracking-firefox45: --- → +
tracking-firefox46: --- → +
tracking-firefox47: --- → +
Flags: needinfo?(sledru)
[Tracking Requested - why for this release]:
(In reply to Liz Henry (:lizzard) (needinfo? me) from comment #18)
> Tracking for 45+ since this seems to be a regression from 45. 
it is already regressing with 44.0
status-firefox44: --- → affected
tracking-firefox44: --- → ?
Please feel free to delete this post if it isn't helpful.

I'm just a computer tech so I can't help with actual coding, but I'd be happy to help with testing.

I'm seeing this problem across a lot of PCs (5 Win7 and 3 Vista) so far.  There is no similarity in hardware across these PCs (only one has NVidia hardware).  The only fix I found is an inplace upgrade/repair install in Windows 7 (don't know if that fixes it in Vista yet, but I'm running some tests).

The symptoms I'm seeing are: 
1)Very slow to browse to pages - in some cases, the page will load in 10 to 15 seconds, in others, the page never loads.
2)The firefox.exe process continues to run for a long time after closing Firefox (doesn't matter how I close Firefox).
3)After 30 seconds to two minutes, the process closes and about 90% of the time I get a crash report.

Things I've tried:
1)Running FF in safe mode.
2)Booting Windows in safe.
3)Updating video, NIC, chipset and SATA drivers.
4)SFC /scannow
5)Different DNS servers.
6)Completing removing FF (Uninstall, then remove folders, registry entries) and reinstalling.

Once the problem occurs, going back to an older FF doesn't fix the problem in the single case I tried.

I'm quite happy to try any suggestions here and pass on reports.  I have three Vista PCs and one Win7 right now with the problem.

Here's one of my threads on the issue - please note I posted it from an un-affected PC:
https://support.mozilla.org/en-US/questions/1107092

Thanks,

T
@Aaron Klotz - Sorry for no response in a while, but I've had a lot of difficulties.  At one point (after the monthly Windows security patches and a reboot) I thought the problem had disappeared, but it came back.  Was also sick for a few days, so couldn't really do much.  And the worst problem is that I simply can't get it to crash.

I hesitated to mention it originally since I wasn't sure exactly what methods you were going to suggest for testing, but as mentioned in the bugs opening post, it's difficult for me to actually get the program to crash.  It will happily hang 3 startups out of 4, but it's only crashed maybe 1 time in 100.  And since the gflags program makes it so that it takes about 5 minutes per restart, it takes a lot of time for me to go through the attempts.

I will note that the program is far less likely to hang when gflags are set.

I will continue to try testing this, but at least wanted to let you know where things stood at the moment.
Flags: needinfo?(dsmith)
Please take a look at 1243910, I could track my problem down to 44.0b9 as the first version that is affected (for me, at least). In my case, Firefox crashes on exit, but only when started from the taskbar. I had all kind of traceability issues as well.

The change set between 44.0b8 (unaffected) and 44.0b9 is rather small, so maybe that will give an idea what has caused this.
Hmm, bug 1218473 landed between b8 and b9 from what I can tell and some people have implied that it may be connected.
I just tried these two builds:
http://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-aurora-win32/1452542313/firefox-45.0a2.en-US.win32.zip 
http://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-aurora-win32/1452523410/

Both crash on Vista the same way the currently released 44 version does.

Here are some more crash logs from various computers:

Vista System 1:
Application Basics
------------------

Name: Firefox
Version: 44.0
Build ID: 20160123151951
Update Channel: release
User Agent: Mozilla/5.0 (Windows NT 6.0; rv:44.0) Gecko/20100101 Firefox/44.0
Multiprocess Windows: 0/2 (default: false)
Safe Mode: false

Crash Reports for the Last 3 Days
---------------------------------

Report ID: bp-e7d5d62a-9c0b-484c-ac71-2e5252160210
Submitted: 5 minutes ago

Report ID: bp-a67b8b3a-c7d4-4084-b3c3-5e6652160210
Submitted: 5 minutes ago

Report ID: bp-74adb3b4-9402-4b6f-a07f-971af2160210
Submitted: 5 minutes ago

Report ID: bp-03c01caa-71a4-4595-821d-3e5b52160210
Submitted: 6 minutes ago

Report ID: bp-5d2f748b-231b-4da4-894f-8f8672160210
Submitted: 6 minutes ago

Report ID: bp-f4279d9b-ba62-4d01-b526-69f072160210
Submitted: 6 minutes ago

Report ID: bp-cece68c8-f024-40f6-93d7-9cd672160210
Submitted: 6 minutes ago

Report ID: bp-49524756-bb14-4769-888c-1d1182160210
Submitted: 6 minutes ago

Report ID: bp-75133a24-b99a-4e28-8e05-777c72160210
Submitted: 1 hour ago

Report ID: bp-42a4bcf8-f3d4-4b77-9cb5-2401f2160209
Submitted: 20 hours ago

Report ID: bp-f1d1139a-8ea4-443e-a2c7-f90162160208
Submitted: 2 days ago

All Crash Reports

Extensions
----------

Graphics
--------

Adapter Description: Mobile Intel(R) 965 Express Chipset Family
Adapter Drivers: igdumdx32 igd10umd32
Adapter RAM: Unknown
Asynchronous Pan/Zoom: none
Device ID: 0x2a02
Direct2D Enabled: Blocked for your graphics driver version.
DirectWrite Enabled: false (7.0.6002.23845)
Driver Date: 2-26-2009
Driver Version: 7.15.10.1666
GPU #2 Active: false
GPU Accelerated Windows: 0/2 Basic (OMTC)
Subsys ID: 011b1025
Supports Hardware H264 Decoding: No; Hardware video decoding disabled or blacklisted
Vendor ID: 0x8086
WebGL Renderer: Google Inc. -- ANGLE (Mobile Intel(R) 965 Express Chipset Family Direct3D9Ex vs_3_0 ps_3_0)
windowLayerManagerRemote: true
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Important Modified Preferences
------------------------------

browser.cache.disk.capacity: 358400
browser.cache.disk.filesystem_reported: 1
browser.cache.disk.smart_size.first_run: false
browser.cache.disk.smart_size.use_old_max: false
browser.cache.frecency_experiment: 1
browser.download.importedFromSqlite: true
browser.places.smartBookmarksVersion: 7
browser.sessionstore.upgradeBackup.latestBuildID: 20160123151951
browser.startup.homepage_override.buildID: 20160123151951
browser.startup.homepage_override.mstone: 44.0
dom.apps.reset-permissions: true
dom.mozApps.used: true
extensions.lastAppVersion: 44.0
gfx.blacklist.direct2d: 3
gfx.crash-guard.d3d11layers.appVersion: 44.0
gfx.crash-guard.d3d11layers.deviceID: 0x2a02
gfx.crash-guard.d3d11layers.driverVersion: 7.15.10.1666
gfx.crash-guard.d3d11layers.feature-d2d: false
gfx.crash-guard.d3d11layers.feature-d3d11: false
gfx.crash-guard.status.d3d11layers: 2
media.gmp-eme-adobe.enabled: false
media.gmp-gmpopenh264.enabled: false
media.gmp-manager.buildID: 20160123151951
media.gmp-manager.lastCheck: 1454958335
media.hardware-video-decoding.failed: true
network.cookie.prefsMigrated: true
network.predictor.cleaned-up: true
places.history.expiration.transient_current_max_pages: 80261
plugin.disable_full_page_plugin_for_types: application/pdf
plugin.importedState: true
plugin.state.nppdf: 0
privacy.sanitize.migrateClearSavedPwdsOnExit: true

Important Locked Preferences
----------------------------

JavaScript
----------

Incremental GC: true

Accessibility
-------------

Activated: false
Prevent Accessibility: 0

Library Versions
----------------

NSPR
Expected minimum version: 4.11
Version in use: 4.11

NSS
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSMIME
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSSL
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSUTIL
Expected minimum version: 3.21
Version in use: 3.21

Experimental Features
---------------------
https://crash-stats.mozilla.com/report/index/bp-e7d5d62a-9c0b-484c-ac71-2e5252160210
https://crash-stats.mozilla.com/report/index/bp-a67b8b3a-c7d4-4084-b3c3-5e6652160210
https://crash-stats.mozilla.com/report/index/bp-74adb3b4-9402-4b6f-a07f-971af2160210
https://crash-stats.mozilla.com/report/index/bp-5d2f748b-231b-4da4-894f-8f8672160210
https://crash-stats.mozilla.com/report/index/bp-f4279d9b-ba62-4d01-b526-69f072160210

Vista system 2:
Application Basics
------------------

Name: Firefox
Version: 45.0a2
Build ID: 20160111064330
Update Channel: default
User Agent: Mozilla/5.0 (Windows NT 6.0; rv:45.0) Gecko/20100101 Firefox/45.0
Multiprocess Windows: 0/1 (default: false)
Safe Mode: false

Crash Reports for the Last 3 Days
---------------------------------

Report ID: bp-87084805-cc66-48c8-ad3c-393b02160210
Submitted: 15 minutes ago

All Crash Reports

Extensions
----------

Name: Firefox Hello
Version: 0.1
Enabled: true
ID: loop@mozilla.org

Graphics
--------

Adapter Description: NVIDIA GeForce G102M
Adapter Drivers: nvd3dum nvwgf2um,nvwgf2um
Adapter RAM: 512
Asynchronous Pan/Zoom: none
Device ID: 0x0873
DirectWrite Enabled: false (7.0.6002.23845)
Driver Date: 12-29-2012
Driver Version: 9.18.13.1090
GPU #2 Active: false
GPU Accelerated Windows: 0/1 Basic (OMTC)
Subsys ID: 19b41043
Supports Hardware H264 Decoding: No; Failed to create H264 decoder
Vendor ID: 0x10de
WebGL Renderer: Google Inc. -- ANGLE (NVIDIA GeForce G102M Direct3D9Ex vs_3_0 ps_3_0)
windowLayerManagerRemote: true
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Important Modified Preferences
------------------------------

browser.cache.disk.capacity: 358400
browser.cache.disk.filesystem_reported: 1
browser.cache.disk.smart_size.first_run: false
browser.cache.frecency_experiment: 2
browser.download.importedFromSqlite: true
browser.places.smartBookmarksVersion: 7
browser.sessionstore.upgradeBackup.latestBuildID: 20160111064330
browser.startup.homepage_override.buildID: 20160111064330
browser.startup.homepage_override.mstone: 45.0a2
browser.tabs.remote.autostart.2: false
dom.apps.reset-permissions: true
dom.mozApps.used: true
extensions.lastAppVersion: 45.0a2
gfx.crash-guard.d3d11layers.appVersion: 45.0a2
gfx.crash-guard.d3d11layers.deviceID: 0x0873
gfx.crash-guard.d3d11layers.driverVersion: 9.18.13.1090
gfx.crash-guard.d3d11layers.feature-d2d: true
gfx.crash-guard.d3d11layers.feature-d3d11: true
gfx.crash-guard.status.d3d11layers: 2
media.gmp-eme-adobe.abi: x86-msvc-x86
media.gmp-eme-adobe.lastUpdate: 1455117685
media.gmp-eme-adobe.version: 15
media.gmp-gmpopenh264.abi: x86-msvc-x86
media.gmp-gmpopenh264.lastUpdate: 1455117728
media.gmp-gmpopenh264.version: 1.5.3
media.gmp-manager.buildID: 20160111064330
media.gmp-manager.lastCheck: 1455117668
media.hardware-video-decoding.failed: true
network.cookie.prefsMigrated: true
network.predictor.cleaned-up: true
places.database.lastMaintenance: 1455117821
places.history.expiration.transient_current_max_pages: 93909
plugin.disable_full_page_plugin_for_types: application/pdf
plugin.importedState: true
privacy.sanitize.migrateClearSavedPwdsOnExit: true
storage.vacuum.last.index: 0
storage.vacuum.last.places.sqlite: 1455117821

Important Locked Preferences
----------------------------

JavaScript
----------

Incremental GC: true

Accessibility
-------------

Activated: false
Prevent Accessibility: 0

Library Versions
----------------

NSPR
Expected minimum version: 4.11
Version in use: 4.11

NSS
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSMIME
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSSL
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSUTIL
Expected minimum version: 3.21
Version in use: 3.21

Experimental Features
---------------------
https://crash-stats.mozilla.com/report/index/bp-87084805-cc66-48c8-ad3c-393b02160210


Windows 7
https://crash-stats.mozilla.com/about/throttling
Application Basics
------------------

Name: Firefox
Version: 45.0a2
Build ID: 20160111064330
Update Channel: default
User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Firefox/45.0
Multiprocess Windows: 1/1 (default: true)
Safe Mode: false

Crash Reports for the Last 3 Days
---------------------------------

All Crash Reports (including 1 pending crash in the given time range)

Extensions
----------

Name: Firefox Hello
Version: 0.1
Enabled: true
ID: loop@mozilla.org

Graphics
--------

Adapter Description: Intel(R) HD Graphics
Adapter Drivers: igdumd64 igd10umd64 igdumdx32 igd10umd32
Adapter RAM: Unknown
Asynchronous Pan/Zoom: none
ClearType Parameters: Gamma: 2200 Pixel Structure: R ClearType Level: 100 Enhanced Contrast: 50
Device ID: 0x0046
Direct2D Enabled: true
DirectWrite Enabled: true (6.2.9200.17568)
Driver Date: 1-30-2013
Driver Version: 8.15.10.2993
GPU #2 Active: false
GPU Accelerated Windows: 1/1 Direct3D 11 (OMTC)
Subsys ID: 00000000
Supports Hardware H264 Decoding: Yes
Vendor ID: 0x8086
WebGL Renderer: Google Inc. -- ANGLE (Intel(R) HD Graphics Direct3D11 vs_4_0 ps_4_0)
windowLayerManagerRemote: true
AzureCanvasBackend: direct2d 1.1
AzureContentBackend: direct2d 1.1
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Important Modified Preferences
------------------------------

browser.cache.disk.capacity: 358400
browser.cache.disk.filesystem_reported: 1
browser.cache.disk.smart_size.first_run: false
browser.cache.disk.smart_size.use_old_max: false
browser.cache.frecency_experiment: 3
browser.download.importedFromSqlite: true
browser.places.smartBookmarksVersion: 7
browser.startup.homepage_override.buildID: 20160111064330
browser.startup.homepage_override.mstone: 45.0a2
dom.apps.reset-permissions: true
dom.mozApps.used: true
extensions.lastAppVersion: 45.0a2
gfx.crash-guard.d3d11layers.appVersion: 45.0a2
gfx.crash-guard.d3d11layers.deviceID: 0x0046
gfx.crash-guard.d3d11layers.driverVersion: 8.15.10.2993
gfx.crash-guard.d3d11layers.feature-d2d: true
gfx.crash-guard.d3d11layers.feature-d3d11: true
gfx.crash-guard.status.d3d11layers: 2
gfx.crash-guard.status.d3d9video: 2
gfx.direct3d.last_used_feature_level_idx: 1
media.gmp-manager.buildID: 20160111064330
media.hardware-video-decoding.failed: false
network.cookie.prefsMigrated: true
network.predictor.cleaned-up: true
places.database.lastMaintenance: 1455117527
places.history.expiration.transient_current_max_pages: 102102
plugin.disable_full_page_plugin_for_types: application/pdf
privacy.sanitize.migrateClearSavedPwdsOnExit: true
storage.vacuum.last.index: 0
storage.vacuum.last.places.sqlite: 1455117527

Important Locked Preferences
----------------------------

JavaScript
----------

Incremental GC: true

Accessibility
-------------

Activated: false
Prevent Accessibility: 0

Library Versions
----------------

NSPR
Expected minimum version: 4.11
Version in use: 4.11

NSS
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSMIME
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSSL
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSUTIL
Expected minimum version: 3.21
Version in use: 3.21

Experimental Features
---------------------

Another Vista computer
Application Basics
------------------

Name: Firefox
Version: 45.0a2
Build ID: 20160111115833
Update Channel: default
User Agent: Mozilla/5.0 (Windows NT 6.0; rv:45.0) Gecko/20100101 Firefox/45.0
Multiprocess Windows: 1/1 (default: true)
Safe Mode: false

Crash Reports for the Last 3 Days
---------------------------------

Report ID: bp-9d7c80e8-c747-47ed-8e50-f2e292160210
Submitted: 1 minute ago

Report ID: bp-824bf1b3-b630-494b-aad6-4d3e62160210
Submitted: 4 hours ago

Report ID: bp-e0bfc66a-7f35-4c1b-9e7e-1a22a2160208
Submitted: 2 days ago

All Crash Reports (including 1 pending crash in the given time range)

Extensions
----------

Name: Firefox Hello
Version: 0.1
Enabled: true
ID: loop@mozilla.org

Graphics
--------

Adapter Description: Intel(R) G33/G31 Express Chipset Family
Adapter Drivers: igdumdx32
Adapter RAM: Unknown
Asynchronous Pan/Zoom: none
Device ID: 0x29c2
Direct2D Enabled: Blocked for your graphics driver version.
DirectWrite Enabled: false (7.0.6002.23845)
Driver Date: 2-26-2009
Driver Version: 7.15.10.1666
GPU #2 Active: false
GPU Accelerated Windows: 0/1 Basic (OMTC)
Subsys ID: 020d1028
Supports Hardware H264 Decoding: No; Failed to create H264 decoder
Vendor ID: 0x8086
windowLayerManagerRemote: true
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Important Modified Preferences
------------------------------

browser.cache.disk.capacity: 358400
browser.cache.disk.filesystem_reported: 1
browser.cache.disk.smart_size.first_run: false
browser.cache.frecency_experiment: 3
browser.download.importedFromSqlite: true
browser.places.smartBookmarksVersion: 7
browser.sessionstore.upgradeBackup.latestBuildID: 20160111115833
browser.startup.homepage_override.buildID: 20160111115833
browser.startup.homepage_override.mstone: 45.0a2
dom.apps.reset-permissions: true
dom.mozApps.used: true
extensions.lastAppVersion: 45.0a2
gfx.blacklist.direct2d: 3
gfx.crash-guard.d3d11layers.appVersion: 45.0a2
gfx.crash-guard.d3d11layers.deviceID: 0x29c2
gfx.crash-guard.d3d11layers.driverVersion: 7.15.10.1666
gfx.crash-guard.d3d11layers.feature-d2d: true
gfx.crash-guard.d3d11layers.feature-d3d11: true
gfx.crash-guard.status.d3d11layers: 2
media.gmp-manager.buildID: 20160111115833
media.hardware-video-decoding.failed: true
network.cookie.prefsMigrated: true
network.predictor.cleaned-up: true
places.history.expiration.transient_current_max_pages: 53385
plugin.disable_full_page_plugin_for_types: application/pdf
plugin.importedState: true
privacy.sanitize.migrateClearSavedPwdsOnExit: true
storage.vacuum.last.index: 0

Important Locked Preferences
----------------------------

JavaScript
----------

Incremental GC: true

Accessibility
-------------

Activated: false
Prevent Accessibility: 0

Library Versions
----------------

NSPR
Expected minimum version: 4.11
Version in use: 4.11

NSS
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSMIME
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSSSL
Expected minimum version: 3.21 Basic ECC
Version in use: 3.21 Basic ECC

NSSUTIL
Expected minimum version: 3.21
Version in use: 3.21

Experimental Features
---------------------

https://crash-stats.mozilla.com/about/throttling
https://crash-stats.mozilla.com/report/index/bp-9d7c80e8-c747-47ed-8e50-f2e292160210
https://crash-stats.mozilla.com/report/index/bp-824bf1b3-b630-494b-aad6-4d3e62160210
I just tried 43.0.3 (both EME and non-EME) on affected computers.  Both have the same problem as 44.  So that's really weird, because it means that once the FF install gets hosed, it doesn't seem to matter what version one uses.
I have just tried removing: Adobe Flash, Adobe Reader, Adobe Air, Java, Silverlight to see if it's one of those, but that seems to make no difference.  Oddly, the Reader plugin still shows up in Firefox (installed after I removed Reader).  I ran FF in Safe Mode with no change.

Basically, pages take minutes to load (if they ever do).  On the same machine, Chrome loads the same page in a second or two (I'm using news.google.ca as a test).  When I close Firefox, the firefox.exe process keeps running for 30 seconds to two minutes, and when it closes, I get a crash report.

At this point, I'm willing to try anything.
Update:  On Win7 and Vista if I disable and stop both DHCP Client (which means setting a static IP) and DNS Client services on affected machines, the browsing speed improves significantly, but I still get crashes on exit about 70% of the time.  So that's an improvement - hopefully that helps.
I'm back to being able to reproduce this (see bug 1243237 comment 41), but only on release, and not with the installed aurora or the special builds in this or from bug 1243237.  About 50/50 on whether it will start, and if it does start, no access to internet.

There are four threads all sitting in QueryDnsForFamily/.../GetThreadPreferredUILanguages/.../WaitForSingleObject if that is out of the ordinary.  And a very rare shutdown crash instead:

https://crash-stats.mozilla.com/report/index/b9d43d44-246b-4294-9d6b-b1d2b2160210
And in local build, I crash on exit when I'm down to a single thread:

msvcp120.dll!std::locale::~locale() Line 440	C++
msvcp120.dll!std::basic_streambuf<wchar_t,std::char_traits<wchar_t> >::~basic_streambuf<wchar_t,std::char_traits<wchar_t> >() Line 80	C++
msvcp120.dll!_CRT_INIT(void * hDllHandle, unsigned long dwReason, void * lpreserved) Line 416	C
msvcp120.dll!__DllMainCRTStartup(void * hDllHandle, unsigned long dwReason, void * lpreserved) Line 522	C
msvcp120.dll!_DllMainCRTStartup(void * hDllHandle, unsigned long dwReason, void * lpreserved) Line 472	C
ntdll.dll!_LdrpCallInitRoutine@16()	Unknown
ntdll.dll!_LdrShutdownProcess@0()	Unknown
ntdll.dll!_RtlExitUserProcess@4()	Unknown
kernel32.dll!_ExitProcessStub@4()	Unknown
firefox.exe!__crtExitProcess(int status) Line 774	C
>firefox.exe!doexit(int code, int quick, int retcaller) Line 666	C
firefox.exe!exit(int code) Line 417	C
firefox.exe!__tmainCRTStartup() Line 264	C
kernel32.dll!@BaseThreadInitThunk@12()	Unknown
ntdll.dll!___RtlUserThreadStart@8()	Unknown
ntdll.dll!__RtlUserThreadStart@8()	Unknown

First-chance exception at 0x7075E437 (msvcp120.dll) in firefox.exe: 0xC0000005: Access violation reading location 0x00746153.
Unhandled exception at 0x7075E437 (msvcp120.dll) in firefox.exe: 0xC0000005: Access violation reading location 0x00746153.
Thane, can you give this build a try and see how it goes? You should probably install it to a different location than your primary install. Thanks!
https://archive.mozilla.org/pub/firefox/try-builds/ryanvm@gmail.com-ac98a6a7a02f71253a82c62cbd7c20279e9da5d8/try-win32/firefox-44.0.2.en-US.win32.installer.exe
Flags: needinfo?(thane)
could you try the test build from comment 31 as well?
Flags: needinfo?(f+bugzilla)
Works on my machine - it installs in another directory, so I had to add it to the taskbar first, but when started from there afterwards, it exits without any crash!
Flags: needinfo?(f+bugzilla)
(In reply to Ryan VanderMeulen [:RyanVM] from comment #31)
> Thane, can you give this build a try and see how it goes? You should
> probably install it to a different location than your primary install.
> Thanks!
> https://archive.mozilla.org/pub/firefox/try-builds/ryanvm@gmail.com-
> ac98a6a7a02f71253a82c62cbd7c20279e9da5d8/try-win32/firefox-44.0.2.en-US.
> win32.installer.exe

On one of the Vista computers, I'm still having problems with this build - crash report here:
https://crash-stats.mozilla.com/report/index/bp-f3aa85ff-f283-41b1-8838-6f6dc2160210

I'm going to test it on the other computers now.

T
Flags: needinfo?(thane)
I forgot to add: Pages don't load at all on that build in Vista.  Windows 7 seems to load pages fine (still not as fast a Chrome, but 3-5 seconds).
@Thane K. Sherrington: based on your crash reports that you have posted in the thread at support.mozilla.org (all shutdownhangs that are dealt with in other bugs), i think your issue is a different one than is discussed in this report. 
so your data points are probably not able to evaluate the effectiveness of the try build from comment 31.
Hi Milan (the verification so far looks promising), would you be able to give the try build a shot and let us know how it goes? Thanks!
Flags: needinfo?(milan)
Ok, on my first Win7 test - pages now load but slowly - comparing to my XP computer, Nightly takes about 4 seconds to load news.google.ca on the Win7 machine and my XP computer loads it in 2.

So far, no crashes on exit.  I've built a script to load Nightly, open two tabs (news.google.ca and www.msn.ca), wait three seconds, close the tabs, the close Nightly, wait until the process ends, then loop.  So far, I've run about eight loops and no crash reporter (crash reporter was coming up at least 70% of the time before).  I have DNS and DHCP running, and the IP address is now set to DHCP.  This was also unstable on 44.0.1.

The only thing that is still problematic (although livable) is the firefox.exe process from nightly is taking 25-30 seconds to close.  This is better than before, but not as good as my XP machine running Nightly, where the process closes about four seconds after Nightly is closed.

I will continue to work on the other 7 computer and three other Vista computers and report back.
(In reply to philipp from comment #36)
> @Thane K. Sherrington: based on your crash reports that you have posted in
> the thread at support.mozilla.org (all shutdownhangs that are dealt with in
> other bugs), i think your issue is a different one than is discussed in this
> report. 
> so your data points are probably not able to evaluate the effectiveness of
> the try build from comment 31.

Sorry, Philipp, I thought I was helping. :)  Can you point me to the thread I should be posting in?

T
(In reply to Ryan VanderMeulen [:RyanVM] from comment #31)
> Thane, can you give this build a try and see how it goes? You should
> probably install it to a different location than your primary install.
> Thanks!
> https://archive.mozilla.org/pub/firefox/try-builds/ryanvm@gmail.com-
> ac98a6a7a02f71253a82c62cbd7c20279e9da5d8/try-win32/firefox-44.0.2.en-US.
> win32.installer.exe

This version is doing the same thing - sitting at "connecting" on the first Vista computer I tried.

Is there a way to tell what it's doing when it's "connecting"?

T
Thane, if you can run into this problem easily, it seems that having HTTP logging file helps people investigating the problems - see instructions https://developer.mozilla.org/en-US/docs/Mozilla/Debugging/HTTP_logging for how to get one of those log.txt files, and then attach it here.
Flags: needinfo?(milan)
I had the same problem, my ff hanged on 35mb.
After uninstalling Avast antivirus (latest version on today), firefox worked fine.
44.0.2 release notes list this bug as "fixed". Bugzilla shows "New". Can someone elaborate? Wrong bug number? Wrong status?
https://www.mozilla.org/en-US/firefox/44.0.2/releasenotes/
(In reply to steve-_- from comment #44)
> 44.0.2 release notes list this bug as "fixed". Bugzilla shows "New". Can
> someone elaborate? Wrong bug number? Wrong status?

yes, this was fixed in 44.0.2 by the backout in https://bugzilla.mozilla.org/show_bug.cgi?id=1218473#c83
so the status here needed updating.
Status: NEW → RESOLVED
Closed: 9 years ago
status-firefox44: affected → fixed
status-firefox45: affected → fixed
status-firefox46: affected → fixed
status-firefox47: affected → fixed
Resolution: --- → FIXED
I saw it, thanks Liz :)
Flags: needinfo?(sledru)
I'm still seeing the same symptom in 42.02 on two computers, a Vista and 7. Not sure if it's the same bug.

Pages sit at "connecting" for between 25-30 seconds (if they load) and forever (they don't load about 50% of the time).

Exiting FF leaves the process running for at least 30 seconds, and generates a crash error when it shuts down.

I can't post the HTTP logging results because they are too long.
Thane, you meant 44.0.2, rather than 42.0.2, right?
Yes, 44.0.02.  Sorry about that.  I've been trying too many versions. :)
I'd be happy to run more tests - I still have two problematic computers (and may run into more this week).  Just let me know what you need and how to send it.

T
Hi! 

I'm having very similar (same?!) problems discussed in this thread, e.g., google.com taking a long time to load (spiking in CPU) and often only for error messages to appear ("The connection was reset"). Hitting "Try again" or refreshing the page then oftentimes loads the page correctly but not always. 

It looks like there has yet to be found a solution as indicated by the post from Thane 3 days ago (though the status says Resolved Fixed in the bottom)?

Windows 7 Home Premium
64-bit OS
Firefox 44.0.2

Thank you!

Nic
Are you running 32-bit Firefox on 64-bit OS, or a 64-bit Firefox?  Do you eventually connect?  How did you update to 44.0.2?
I believe I'm running the 64-bit version. Eventually it connects but I have to refresh or hit Try Again etc. My firefox is set to update automatically.

Name 	Firefox
Version 	44.0.2
Build ID 	20160210153822
Update History 	
Update Channel 	release
User Agent 	Mozilla/5.0 (Windows NT 6.1; WOW64; rv:44.0) Gecko/20100101 Firefox/44.0
Profile Folder 	
Enabled Plugins 	about:plugins
Build Configuration 	about:buildconfig
Memory Use 	about:memory
Registered Service Workers 	about:serviceworkers
Multiprocess Windows 	0/1 (default: false)
Safe Mode 	false
I am running 44.0.2 on Vista 32 bit and Win7 64 bit, which exhibits the same problem.
I started looking for a pattern and it seems like firefox has trouble particularly with loading google.com and gmail.
I have verified if I use https://www.google.com for my default start page, I consistently get a blank page when I close and reopen Firefox.
please open a new bug about this issue - it won't be related to the problem originally fixed in here. thank you!
Try downloading SG TCP optimizer from here:
http://www.speedguide.net/downloads.php

And set it to Windows default.  Then reboot and try again.

T
You need to log in before you can comment on or make changes to this bug.