*FREQUENT* BSODs with Palemoon!

[Rekrul]

I leave my system on 24/7. Previously it was up for about a month. In fact, I don't think I rebooted it at all in December. During that time I've used my old version of Firefox heavily, downloaded tons of files, watched videos, played games, run emulators and I had no problems at all. I rebooted earlier today because I noticed my system was getting a little sluggish.

Tonight, I loaded Palemoon (latest portable version) to access a Tumblr page with images. I was using it for a grand total of about three minutes and opened maybe 5-6 of the images in their own tabs. BSOD!

In fact, virtually every Palmoon session that lasts more than a couple minutes or where I open more than 1-2 tabs usually ends in a BSOD. Of course when I want to make it happen to try and pin down the cause, I can open 20+ tabs and nothing happens. Other times I load it up to access a web site that gives me problems in Firefox, thinking "I'll be OK as long as I don't open too many tabs." and I get a BSOD.

I have no extensions installed in Palemoon. In fact I haven't even bookmarked any sites or changed the default homepage. All I did was unpack the archive and make a shortcut in the Start menu. The plan was to test it out and once I was comfortable with it, I'd install the normal version and make it my default browser. Unfortunately, using Palemoon is like playing Russian Roulette. Sure you might get lucky a couple times, but play long enough and you're guaranteed to lose.

I had this system for 2-3 years before I ever even heard of Palemoon and I think I averaged about 1-2 BSODs a year. Since I first downloaded Palemoon less than a year ago, I've had a couple dozen BSODs, all while using Palemoon.

Is there anything I can do to track down why Palemoon keeps crashing my system? It would be nice to have a modern browser, but as it is now, Palemoon is too unstable to use.

2.4Ghz Core-2-Duo
4GB RAM
Windows XP Pro, SP3

[megaman]

That must be a heavy use, there.
BSODs usually happen due to graphics. You can disable Hardware Acceleration in the Options > Advanced menu.

Can you try a test with a new Profile, since the current one must've gotten corrupted?

Restarting is less due to XP not getting more attention from Microsoft.

I can't give further help, since I use Windows 10, but others that still Windows XP might try the site and see how they fare.

[satrow]

Please attach the zipped minidumps or link to an upload site where we can check the kernel or full dump that you have uploaded.

Alt., read and carefully follow the instructions at Sysnative: http://www.sysnative.com/forums/bsod-cr ... vista.html and then post the required answers and attach the zipped files back here for checking.

I'm almost certain that these BSODs are not the fault of Pale Moon, perhaps some bad driver/update or a piece of malware - but not Pale Moon.


EDIT: I find your comments regarding using Pale Moon like playing Russian roulette and it not being a 'modern browser' to be very insulting. You want help with your problem, don't set off on the wrong foot.

[Rekrul]

That must be a heavy use, there.
BSODs usually happen due to graphics. You can disable Hardware Acceleration in the Options > Advanced menu.

It really wasn't that graphics heavy though. I went to a Tumblr page full of images, went to the "Archive" version, where it shows multiple thumbnails on the same page, then middle-clicked on 5-6 of them to open them in new tabs. Each of those tabs would have had a larger version of the image, maybe some small, incidental graphics (buttons and such) and then a bunch of text comments. Before my old version of Firefox decided it no longer wanted to work on Tumblr archive pages, I'd open a few dozen images in tabs from the same site and the worst thing that would happen is that Firefox would eventually crash.

Can you try a test with a new Profile, since the current one must've gotten corrupted?

OK. I'll have to pick a time when I can risk crashing my system without losing anything.

Please attach the zipped minidumps or link to an upload site where we can check the kernel or full dump that you have uploaded.

OK, attached is the minidump for the crash I had a few days ago. I will enable Kernel memory dump and attempt to generate another crash while using Palemoon.

Alt., read and carefully follow the instructions at Sysnative

Unfortunately, that doesn't work under XP.

I'm almost certain that these BSODs are not the fault of Pale Moon, perhaps some bad driver/update or a piece of malware - but not Pale Moon.

I'll keep an open mind, but as it stands, Palemoon is the only piece of software that my system consistently crashes while using.

I've crashed Firefox 13.0.1 many times by opening too many tabs, or just going to a web site that it didn't like, but I don't recall it ever causing a BSOD.

EDIT: I find your comments regarding using Pale Moon like playing Russian roulette and it not being a 'modern browser' to be very insulting. You want help with your problem, don't set off on the wrong foot.

I never said Palemoon wasn't a modern browser. In fact, I'd like to be able to use Palemoon reliably because it's a modern browser. I'm still using an old version of Firefox (see below) that gives me problems on various sites.

I posted the comment about it being like Russian Roulette out of frustration. Please look at it from my point of view; I use probably a dozen different programs on a daily basis and get no BSOD. I overload Firefox to the point of crashing it, no BSOD. I play an admittedly older game for literally hours at a time, no BSOD. I load Palemoon, open 5-6 tabs (way less than I've successfully opened in Firefox) and I get a BSOD. Even if it's not a bug in Palemoon, it certainly appears that it's doing something different that my system doesn't like.

Everyone reading this is probably wondering why I'm still using Firefox 13...

I started with Firefox 2(?) and dutifully upgraded to each new version. As the version numbers got higher, I noticed that they kept removing features I liked (like the status bar, the old find box, etc). At one point I tried to upgrade from 13 to a higher version, such as 16 or 19, but things didn't go so well. The toolbar came out screwed up, a bunch of extensions didn't work properly and several web sites didn't work properly. So I switched back to FF13 and decided not to try upgrading further.

Unfortunately now, I have problems on a bunch of sites, even ones that used to work previously. Posting new YouTube comments no longer works under my default profile (although I can reply, and it works with a new profile), Tumblr archive pages no longer display the thumbnails or let me select what month to view and quite a few sites refuse to even let me view the site if I'm using an older browser.

It seems as if the more time passes, the worse Firefox works. I keep putting off re-installing it because of how I'll need to find and backup things like the saved passwords, saved cookies, bookmarks, all the custom tweaks I've made to it, etc.

Which is why I'd like to be able to use Palemoon.

[satrow]

I don't have debugging tools loaded here (fresh W7 install - buggy, might wipe/reload again) and XP is a bear to debug anyway! Luckily this is an IRQL_NOT_LESS_OR_EQUAL, almost always caused by a bad 3rd party driver.

This crash happened in the networking subsystem, from the Stack text it looks like the culprit is a Sunbelt?/Kerio firewall driver, fwdrv.sys. I've no idea what version of SB/Kerio it is from but the date is 15/04/2002 - update Kerio FW or completely uninstall it and use a later XP compatible FW (please, not ZA, Comodo might also over complicate things).

There's also an ancient (1998!) Belarc? driver, BANTExt.sys, that looks like it needs uninstalling as well.

Private Firewall might be suitable and it still runs on XP.

[Moonchild]

I've fired up a debugger and loaded WinXP symbols.
Confirmed it's fwdrv that is causing the BSOD.

Code: Select all

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

IRQL_NOT_LESS_OR_EQUAL (a)
An attempt was made to access a pageable (or completely invalid) address at an
interrupt request level (IRQL) that is too high.  This is usually
caused by drivers using improper addresses.
If a kernel debugger is available get the stack backtrace.
Arguments:
Arg1: 00000004, memory referenced
Arg2: 0000001c, IRQL
Arg3: 00000000, bitfield :
	bit 0 : value 0 = read operation, 1 = write operation
	bit 3 : value 0 = not an execute operation, 1 = execute operation (only on chips which support this level of status)
Arg4: 804ffa88, address which referenced memory

Debugging Details:
------------------


READ_ADDRESS:  00000004 

CURRENT_IRQL:  1c

FAULTING_IP: 
nt!KiInsertQueueApc+84
804ffa88 8b7704          mov     esi,dword ptr [edi+4]

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  DRIVER_FAULT

BUGCHECK_STR:  0xA

PROCESS_NAME:  Idle

LAST_CONTROL_TRANSFER:  from 804fc53b to 804ffa88

STACK_TEXT:  
80550dd8 804fc53b 8a4fb048 8a4fb008 00000000 nt!KiInsertQueueApc+0x84
80550df8 804f1860 8a4fb048 00000000 00000000 nt!KeInsertQueueApc+0x51
80550e2c b430f942 80550e9c 89ebf3f8 80550e9c nt!IopfCompleteRequest+0x1d8
80550e44 b4312be6 8a4fb008 00000000 000001bf tcpip!TCPDataRequestComplete+0xa6
80550e64 b430f45f 80550e9c 8aaeabf8 89e7806e tcpip!CompleteSends+0x2b
80550ee0 b430cef5 8ad79728 2cc05543 1403c5d1 tcpip!TCPRcv+0x1be
80550f40 b430cb19 00000020 8ad79728 b430f0b6 tcpip!DeliverToUser+0x18e
80550fbc b430c836 b434c8f0 8ad79728 89e7806e tcpip!DeliverToUserEx+0x95e
80551074 b430b928 8ad79728 89e78082 0000001a tcpip!IPRcvPacket+0x6cb
805510b4 b43106ef 00000000 89e78000 89e78060 tcpip!ARPRcvIndicationNew+0x149
805510e4 b43844a6 8ad71298 89e78000 89e78060 tcpip!ARPRcv+0x42
WARNING: Stack unwind information not available. Following frames may be wrong.
8055111c b43846ae 8ad71298 8aaeeba0 00000000 fwdrv+0xd4a6
8055114c b7ce4ef2 8ad71298 8ae3e8f0 8af2900a fwdrv+0xd6ae
805511b4 b705d01d 00000000 8ae338c8 00000002 NDIS!ethFilterDprIndicateReceivePacket+0x1ff
805511c8 b705d1b4 8b207470 8ae338c8 00000002 psched!PsFlushReceiveQueue+0x15
805511ec b705d5f9 8ae33978 00000000 8b207470 psched!PsEnqueueReceivePacket+0xda
80551204 b7ce4c40 8ae33970 805512b8 00000002 psched!ClReceiveComplete+0x13
80551254 b71441f8 00f373d0 805512b8 00000002 NDIS!ethFilterDprIndicateReceivePacket+0x5a4
80551270 b714e35c 8b200128 805512b8 00000002 e1e5132+0x21f8
80551290 b7150b3d 8b0a08d8 00000000 805512b8 e1e5132+0xc35c
805513c8 b714a341 0109e000 00000002 00000000 e1e5132+0xeb3d
80551404 b7143371 0009e000 80551428 b7cdae99 e1e5132+0x8341
80551410 b7cdae99 8b200128 8055c0c0 ffdff9c0 e1e5132+0x1371
80551428 80545f9f 8b09eb24 8b09eb10 00000000 NDIS!ndisMDpcX+0x21
80551440 8055be60 ffdffc50 00000000 8055be60 nt!KiRetireDpcList+0x61
80551450 80545e84 00000000 0000000e 00000000 nt!KiIdleThread0
80551454 00000000 0000000e 00000000 00000000 nt!KiIdleLoop+0x28


STACK_COMMAND:  kb

FOLLOWUP_IP: 
fwdrv+d4a6
b43844a6 ??              ???

SYMBOL_STACK_INDEX:  b

SYMBOL_NAME:  fwdrv+d4a6

FOLLOWUP_NAME:  MachineOwner

MODULE_NAME: fwdrv

IMAGE_NAME:  fwdrv.sys

DEBUG_FLR_IMAGE_TIMESTAMP:  3cbaab4f

FAILURE_BUCKET_ID:  0xA_fwdrv+d4a6

BUCKET_ID:  0xA_fwdrv+d4a6

Followup: MachineOwner

Driver details:

Code: Select all

0: kd> lmvm fwdrv
start    end        module name
b4377000 b4395000   fwdrv    T (no symbols)           
    Loaded symbol image file: fwdrv.sys
    Image path: fwdrv.sys
    Image name: fwdrv.sys
    Timestamp:        Mon Apr 15 12:28:31 2002 (3CBAAB4F)
    CheckSum:         00020E53
    ImageSize:        0001E000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4

Satrow is correct that this is a critical fault in the driver and you should either stop using it or (if available) update to something newer than a 2002 version ;P

@Satrow: by the way, you can use this guide to debug any minidump this way, including loading of other OS' symbols:
http://www.techrepublic.com/blog/windows-and-office/how-do-i-use-windbg-debugger-to-troubleshoot-a-blue-screen-of-death/

[satrow]

Thanks for the confirmation of the cause, MC.

I usually have the tools loaded but with this buggy new install, I really didn't want to 'waste' 30 minutes getting it setup

- maybe I should though, I have 29 process dumps of my own to look at

[megaman]

Off-topic:
This is unrelated because it's Windows 10 and Palemoon 26 Beta 6, but I just got the same error. I'll check to see what it is.
I was working on Yahoo! Mail when it happened, and I had a few tabs open(Maybe 8) but they were unloaded.
I'll edit with the info.

Edit: Since it's unusual and it would stay at 0%, I felt that it must've been frozen and just forced off the PC. In the end, no minidump.(sigh)
I got another BSOD trying checking on the problem. "Driver_IRQL_Not_Less_Or_Equal(storport.sys)
I'll have a report of that, it's going up on %, so at least some progress.

It's a driver with AMD.
I'll check for updates.

My issue was due to Speccy. Palemoon is not at fault.
We can disregard this message.

[Ultor]

Me too! Frequent BSOD... but with Palemoon?
I can view with Bluescreenview errors with:
hal.dll
NDIS.sys
nlndis.sys
nltdi.sys
ntoskrnl.exe
tcpip.sys

[dark_moon]

nlndis.sys and nltdi.sys is from NetLimiter driver.
Try first without that tool.

[Ultor]

I can't find a simple alternative to Netlimiter!

[Moonchild]

I can't find a simple alternative to Netlimiter!

Considering NetLimiter is a commercial tool, you may want to ask them for support here for your product that's causing you grief.

[Ultor]

They do not respond!
Another problem!
Windows7FirewallControl.sys

[Moonchild]

Well at least it's all focused around networking, so you know where to look.
Netlimiter is a prime suspect. And them not responding is pretty bad. I did use it for a while myself in the past and had similar issues with stability, so I stopped using it.
As a browser developer, we can't do anything about this though, since it's not our code. Pale Moon may trigger it by tickling sensitive code in whatever broken driver is bluescreening your system, but isn't the cause.

[Ultor]

it would be nice a addon which limit the bandwidth of Pale Moon.

[dark_moon]

Why someone need this?
You have problems with your software not with Pale Moon. Use another software or use your system without your buggy software.

Anyway limit the bandwidth have nothing todo with your problems.

[RJARRRPCGP]

0x0000000A means you're probably overclocking your processor and you need more Vcore or you need to turn down the core speed my at least about 50 Mhz. I used to get that one...

[Moonchild]

0x0000000A means you're probably overclocking your processor and you need more Vcore or you need to turn down the core speed my at least about 50 Mhz. I used to get that one...

Although overclocking CAN cause this stop error, it doesn't automatically mean that this stop error is caused by overclocking.
It's also clearly centered around the networking layer, and considering the error happens in a 14 year old driver... well... overclocking instability would definitely not be very probable here as a cause.

[satrow]

0x0000000A means you're probably overclocking your processor and you need more Vcore or you need to turn down the core speed my at least about 50 Mhz. I used to get that one...

No sign of any overclock, it was running at stock speed at the time of the bugcheck:

[Processor Information (Type 4) - Length 40 - Handle 0005h]
Socket Designation XU1 PROCESSOR
Processor Type Central Processor
Processor Family b2h - Pentium IV Processor
Processor Manufacturer Intel
Processor ID f6060000fffbebbf
Processor Version Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
Processor Voltage 8ch - 1.2V
External Clock 1066MHz
Max Speed 6000MHz
Current Speed 2400MHz

STOP 0x0000000A: IRQL_NOT_LESS_OR_EQUAL
Usual causes: Kernel mode driver, System Service, BIOS, Windows, Virus scanner, Backup tool, compatibility

Probably best not to assume that an overclock was in place, esp. in more general tech fora, overclocking being something of a minority activity.

[RJARRRPCGP]

0x0000000A means you're probably overclocking your processor and you need more Vcore or you need to turn down the core speed my at least about 50 Mhz. I used to get that one...

No sign of any overclock, it was running at stock speed at the time of the bugcheck:

[Processor Information (Type 4) - Length 40 - Handle 0005h]
Socket Designation XU1 PROCESSOR
Processor Type Central Processor
Processor Family b2h - Pentium IV Processor
Processor Manufacturer Intel
Processor ID f6060000fffbebbf
Processor Version Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
Processor Voltage 8ch - 1.2V
External Clock 1066MHz
Max Speed 6000MHz
Current Speed 2400MHz

STOP 0x0000000A: IRQL_NOT_LESS_OR_EQUAL
Usual causes: Kernel mode driver, System Service, BIOS, Windows, Virus scanner, Backup tool, compatibility

Probably best not to assume that an overclock was in place, esp. in more general tech fora, overclocking being something of a minority activity.

I bet there's data corruption that's causing programs to do strange stuff in the traces... Bad caps are a possibility, even though a lot of caps are of the poly type.