Open Bug 1989283 Opened 1 month ago Updated 1 month ago

TB 143 keeps crashing sometimes with Raptor Lake CPU

Categories

(Thunderbird :: Mail Window Front End, defect)

Product:
Thunderbird
Component:
Mail Window Front End
Version:
Thunderbird 142
Type:
defect

Tracking

(Not tracked)

People

(Reporter: submarino, Unassigned)

References

(Blocks 1 open bug)

Details

Crash Data

Attachments

(2 files)

1758197963018.642a9009-f714-418e-9690-993f207668f9.main.jsonlz4
4.26 KB, application/octet-stream
Details
The error (in brazilian portuguese)
34.22 KB, image/png
Details

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/140.0.0.0 Safari/537.36

Steps to reproduce:

Windows 10, when I interact with the interface (drag and drop, click on folder, keyboard arrows to change the message on the list), TB 143 (and TB 144 Daily) freezes. Several times per day. Happened with TB 142 too before update. No problem on previous versions.

Errors like:

BuildID: 20250910231831
CrashTime: 1758198092
InstallTime: 1758110775
ModuleSignatureInfo: {"Mozilla Corporation":["thunderbird.exe","xul.dll","rnp.dll","gkcodecs.dll","nss3.dll","freebl3.dll","softokn3.dll","mozglue.dll","lgpllibs.dll"],"Microsoft Windows":["wsock32.dll","NapiNSP.dll","pnrpnsp.dll","winrnr.dll","wscapi.dll","ktmw32.dll","usp10.dll","TaskFlowDataEngine.dll","explorerframe.dll","ntshrui.dll","thumbcache.dll","dataexchange.dll","directmanipulation.dll","Windows.UI.Immersive.dll","twinapi.dll","Windows.Security.Integrity.dll","Windows.Globalization.dll","wshbth.dll","Bcp47Langs.dll","InputHost.dll","TextInputFramework.dll","Windows.UI.dll","DWrite.dll","BCP47mrm.dll","OneCoreCommonProxyStub.dll","dbgcore.dll","cdp.dll","comctl32.dll","cscapi.dll","urlmon.dll","iertutil.dll","srvcli.dll","linkinfo.dll","winmm.dll","FWPUCLNT.DLL","OneCoreUAPCommonProxyStub.dll","rasadhlp.dll","npmproxy.dll","ColorAdapterClient.dll","mscms.dll","twinapi.appcore.dll","WindowManagementAPI.dll","avrt.dll","version.dll","dhcpcsvc.dll","dhcpcsvc6.DLL","winnsi.dll","MMDevAPI.dll","AudioSes.dll","policymanager.dll","propsys.dll","wtsapi32.dll","dsreg.dll","WinTypes.dll","d2d1.dll","d3d11.dll","netprofm.dll","nlaapi.dll","dcomp.dll","CoreUIComponents.dll","CoreMessaging.dll","WindowsCodecs.dll","dbghelp.dll","uxtheme.dll","dwmapi.dll","kernel.appcore.dll","windows.storage.dll","dxgi.dll","winsta.dll","msvcp110_win.dll","ntmarta.dll","IPHLPAPI.DLL","dnsapi.dll","netutils.dll","mswsock.dll","CRYPTBASE.dll","cryptsp.dll","umpdc.dll","powrprof.dll","wldp.dll","ntasn1.dll","ncrypt.dll","msasn1.dll","devobj.dll","userenv.dll","sspicli.dll","profapi.dll","gdi32full.dll","wintrust.dll","bcrypt.dll","bcryptPrimitives.dll","ucrtbase.dll","crypt32.dll","KERNELBASE.dll","msvcp_win.dll","win32u.dll","cfgmgr32.dll","SHCore.dll","psapi.dll","user32.dll","advapi32.dll","msctf.dll","clbcatq.dll","kernel32.dll","shell32.dll","ole32.dll","combase.dll","shlwapi.dll","setupapi.dll","ws2_32.dll","oleaut32.dll","sechost.dll","gdi32.dll","rpcrt4.dll","imm32.dll","msvcrt.dll","nsi.dll","ntdll.dll","KBDBR.DLL","resourcepolicyclient.dll"],"Microsoft Windows Software Compatibility Publisher":["msvcp140.dll","VCRUNTIME140.dll","VCRUNTIME140_1.dll"]}
ProductID: {3550f703-e582-4d05-9a08-453d09bdfdc6}
ProductName: Thunderbird
ReleaseChannel: release
StartupTime: 1758198002
SubmittedFrom: Client
Throttleable: 1
UptimeTS: 90.0
Vendor: Mozilla
Version: 143.0
WindowsErrorReporting: 1

Actual results:

Software closed.

Tested in "security mode" and all extensions disabled. No difference.
Tested in a different Windows 10 (very basic install) running in a VM (Virtualbox). Same problem.

From Event Viewer:

Faulting application name: thunderbird.exe, version: 143.0.0.0, timestamp: 0x68c213bb
Faulting module name: xul.dll, version: 143.0.0.231, timestamp: 0x68c213cf
Exception code: 0xc0000409
Fault offset: 0x0000000007dca33d
Faulting process ID: 0xb6c
Faulting application start time: 0x01dc289866d80459
Faulting application path: C:\Program Files\Mozilla Thunderbird\thunderbird.exe
Faulting module path: C:\Program Files\Mozilla Thunderbird\xul.dll
Report ID: b6ccee76-f140-4946-8040-b4fdccce51ce
Full name of the failing package:
Application ID for the failing package:

Do you have a crash ID? I'm not seeing any crashes on https://crash-stats.mozilla.org for b6ccee76-f140-4946-8040-b4fdccce51ce

Version: Thunderbird 143 → Thunderbird 142

(In reply to Corey Bryant from comment #3)

Do you have a crash ID? I'm not seeing any crashes on https://crash-stats.mozilla.org for b6ccee76-f140-4946-8040-b4fdccce51ce

How can I get it?

I found this in C:\Users\user\AppData\Roaming\thunderbird\Crash Reports\submitted

https://crash-stats.mozilla.org/report/index/4cfebf59-aae3-4dfe-afcd-0c72e0250916
https://crash-stats.mozilla.org/report/index/c0dcd51e-61ee-4418-a7e7-917ba0250916

and

https://crash-stats.mozilla.org/report/index/539f1e57-81b1-4669-9fcc-128cd0250829

From \crashes reading inside the "store.json.mozlz4" file...

But I cannot find files for today (TB 143)...

You can go to "Help" > "Troubleshooting Information", and search for "Crash Reports"

(In reply to Corey Bryant from comment #7)

You can go to "Help" > "Troubleshooting Information", and search for "Crash Reports"

Found it, but the last one is:

https://crash-stats.mozilla.org/report/index/4cfebf59-aae3-4dfe-afcd-0c72e0250916

No report for today, but I submitted all of them. Probably there's a limit that I can send.

Anyway, the report above is related to the problem...

These crashes happened on a Raptor Lake (family 6 model 183 stepping 1) CPU which is known for its problems, especially in hotter environments, see also meta bug 1975808. Almost 90% of all reported mozilla::detail::nsDequeBase::PopFront crashes are from that CPU. For the PtrInfo::WasTraversed signature on the other hand, see bug 500105 comment 106.

See Also: → cpu-raptor-lake-bugs, 500105

It's an Intel i9 14900K, but it's almost idle, less than 5% usage and temperature below 45 Celsius.

Now I have a new report, specific to TB 143:

https://crash-stats.mozilla.org/report/index/0f7c0c21-0014-4c38-af5c-bcdcd0250918

Any way, hope that will be fixed. That's a top-line CPU.

I read the other topics. I just cannot understand. Until TB 140 there was no problem. Started on 142. Looks like a SW problem, not HW.

Other point: the CPU is only 4 months in use, low usage, day-by-day usage, there's no degradation. And no other software crashes, and I have several, including some games. Even Firefox 143 runs fine, no crashes.

Though I'm not crashing per se, I am noticing the firefox.exe process locking up (website stop loading/working, I shut down FF, processes is still running) and I am having to end task on it.

(In reply to Hartmut Welpmann [:welpy-cw] from comment #9)

These crashes happened on a Raptor Lake (family 6 model 183 stepping 1) CPU which is known for its problems, especially in hotter environments, see also meta bug 1975808. Almost 90% of all reported mozilla::detail::nsDequeBase::PopFront crashes are from that CPU. For the PtrInfo::WasTraversed signature on the other hand, see bug 500105 comment 106.

In fact, my CPU microcode was revision 1801 (2024-11). There was an update, revision 2001 (2025-07). For Asus ROG Z70 Dark Hero.

I've applied the update. And... TB 143 appears to be ok now. I did several tests, trying to crash it. Appears to be stable (still monitoring). 🙌

Well, I'm not an expert in hardware, but it's the first time I saw something like this...

(In reply to Rogerio Arvy from comment #13)

(In reply to Hartmut Welpmann [:welpy-cw] from comment #9)

These crashes happened on a Raptor Lake (family 6 model 183 stepping 1) CPU which is known for its problems, especially in hotter environments, see also meta bug 1975808. Almost 90% of all reported mozilla::detail::nsDequeBase::PopFront crashes are from that CPU. For the PtrInfo::WasTraversed signature on the other hand, see bug 500105 comment 106.

In fact, my CPU microcode was revision 1801 (2024-11). There was an update, revision 2001 (2025-07). For Asus ROG Z70 Dark Hero.

I've applied the update. And... TB 143 appears to be ok now. I did several tests, trying to crash it. Appears to be stable (still monitoring). 🙌

Well, I'm not an expert in hardware, but it's the first time I saw something like this...

That is interesting indeed. So the crashes could possibly be due to trying to exploit an un-patched CPU issue and a BIOS update carrying the microcode update fixes it? Seems too good to be true.

Yes, I've been trying to crash TB 143 for 30 minutes and I can't, clicking everything I can. Before, if I did what I'm doing, it wouldn't last 10 seconds.

I've been using computers since 1986 and consider myself an advanced computer expert (even though I don't have any knowledge of the "electrical" or hardware side, I don't overclock or anything like that). But this is the first time I've actually seen a BIOS update make a difference. I always update when I find out there's an update available. But this is the first time I've actually seen something effective.

Perhaps some of the users experiencing this issue are using Asus ROG Z790 motherboards. And the solution would simply be to "update your BIOS/UEFI," even from other brands.

I'm still confused about the solution. Really impressed.

(In reply to Rogerio Arvy from comment #8)

(In reply to Corey Bryant from comment #7)

You can go to "Help" > "Troubleshooting Information", and search for "Crash Reports"

Found it, but the last one is:

https://crash-stats.mozilla.org/report/index/4cfebf59-aae3-4dfe-afcd-0c72e0250916

Fantastic info. Thanks for seeing it through.

The standard being used in bug 1975808 is to set the crashing bug reports as blocking.

(In reply to Rogerio Arvy from comment #11)

I read the other topics. I just cannot understand. Until TB 140 there was no problem. Started on 142. Looks like a SW problem, not HW.

I don't have an explanation for that. However, this crash signature has a significant crash rate prior to 142, as do most of the other crash signatures associated with this CPU.

Blocks: cpu-raptor-lake-bugs
Status: UNCONFIRMED → NEW
Crash Signature: [@ mozilla::detail::nsDequeBase::PopFront ]
Ever confirmed: true
See Also: cpu-raptor-lake-bugs → https://support.mozilla.org/en-US/questions/1535991
Summary: TB 143 keeps crashing everytime → TB 143 keeps crashing everytime with Raptor Lake CPU
Summary: TB 143 keeps crashing everytime with Raptor Lake CPU → TB 143 keeps crashing everytime with Raptor Lake CPU (but updating BIOS/UEFI to latest microcode fixes it!)

BAD NEWS! 🤨

Problem is happening again... stable for several hours. And now crasing but only sometimes.

Summary: TB 143 keeps crashing everytime with Raptor Lake CPU (but updating BIOS/UEFI to latest microcode fixes it!) → TB 143 keeps crashing everytime with Raptor Lake CPU
Summary: TB 143 keeps crashing everytime with Raptor Lake CPU → TB 143 keeps crashing sometimes with Raptor Lake CPU

(In reply to Rogerio Arvy from comment #17)

BAD NEWS! 🤨

Problem is happening again... stable for several hours. And now crasing but only sometimes.

Wondering if the crash info has changed since the BIOS update and now it's something new? Seems to me that with 142 and prior version you would have been seeing this crashing if it was a CPU microcode issue at fault. Do you have a newer crash report to point to for comparison? Type about:crashes into URL bar and see if there's a newer one post-BIOS Update and paste it to the bug report.

And I see 143.0.1 just got pushed. You don't have trend Micro antivirus by chance?

And I see 143.0.1 just got pushed. You don't have trend Micro antivirus by chance?

Sorry, disregard this. That was for Firefox not TB.

(In reply to Arthur K. (he/him) from comment #18)

(In reply to Rogerio Arvy from comment #17)

BAD NEWS! 🤨

Problem is happening again... stable for several hours. And now crasing but only sometimes.

Wondering if the crash info has changed since the BIOS update and now it's something new? Seems to me that with 142 and prior version you would have been seeing this crashing if it was a CPU microcode issue at fault. Do you have a newer crash report to point to for comparison? Type about:crashes into URL bar and see if there's a newer one post-BIOS Update and paste it to the bug report.

That's the problem, there is no report... there are 2 "crash" boxes: one with one button (to close TB) and other with 2 buttons (restart and close). The crash report is only generated when appears the box with 2 buttons, that happened only twice for me... most is this: https://bugzilla.mozilla.org/show_bug.cgi?id=1989283#c5

I'm working hard to figure out the problem. After updating the BIOS, TB became more stable. Now it's only crashing in certain situations, like when searching or accessing an personal folder (not all). There's no specific reason.

I remembered my motherboard was running in Asus's "AI Overclock Mode." I wondered if that could be the cause. To test, I reset the BIOS to factory settings. So far, there hasn't been any crashing. Since I'm not an expert in hardware, and I honestly still don't understand the relationship between hardware and TB, I'm testing it. One thing I've noticed is that, in addition to TB, I use some virtual machines in VirtualBox, and after the reset, they seem to be faster. I'm not sure. But right now, my CPU is running at the BIOS default. Now I'm wondering if this "AI Overclock" was reliable or not, and if it could be causing something unexpected.

(In reply to Rogerio Arvy from comment #22)

I'm working hard to figure out the problem. After updating the BIOS, TB became more stable. Now it's only crashing in certain situations, like when searching or accessing an personal folder (not all). There's no specific reason.

I remembered my motherboard was running in Asus's "AI Overclock Mode." I wondered if that could be the cause. To test, I reset the BIOS to factory settings. So far, there hasn't been any crashing. Since I'm not an expert in hardware, and I honestly still don't understand the relationship between hardware and TB, I'm testing it. One thing I've noticed is that, in addition to TB, I use some virtual machines in VirtualBox, and after the reset, they seem to be faster. I'm not sure. But right now, my CPU is running at the BIOS default. Now I'm wondering if this "AI Overclock" was reliable or not, and if it could be causing something unexpected.

This is not surprising as far as seeing improvement after factory resetting the BIOS after an update. Often, when going from a minor update (1.00 > 1.01) you don't need to factory reset the BIOS. But often, a Mobo maker recommends that if going from 1.00 > 1.5 or 1.00 > 2.00, a full factory reset is required when there is such a huge version delta. That may the case for you as well. Any improvement or less crashes since factory reset?

(In reply to Arthur K. (he/him) from comment #23)

This is not surprising as far as seeing improvement after factory resetting the BIOS after an update. Often, when going from a minor update (1.00 > 1.01) you don't need to factory reset the BIOS. But often, a Mobo maker recommends that if going from 1.00 > 1.5 or 1.00 > 2.00, a full factory reset is required when there is such a huge version delta. That may the case for you as well. Any improvement or less crashes since factory reset?

Yes, since the factory reset and the AI ​​overclocking feature were disabled, no more crashes. Perhaps a combination of both caused the issue. After 2 days, the TB 143 is working normally, no crashes. Still monitoring but I think that the BIOS update was the main solution to my CPU.

I just noticed that there are two "X" on that crash submission dialog. The second one with the scroll bar is the so-called "crash details" or comment?

So are you saying sometimes you do not see restart and close button choices?

Flags: needinfo?(submarino)

(In reply to Wayne Mery (:wsmwk) from comment #25)

I just noticed that there are two "X" on that crash submission dialog. The second one with the scroll bar is the so-called "crash details" or comment?

Yes, the second screen is the "details" button.

So are you saying sometimes you do not see restart and close button choices?

Exactly. When there's no crash report (about:crashes), the first box only shows the "Close" button. When there is, the "Restart" and "Close" buttons appear. But even in the first case, I could send the report; it says it was sent, but there's no report anywhere. I don't know if I was "blocked" or something, because on the first day I sent all the reports (+20).

Flags: needinfo?(submarino)

per https://support.mozilla.org/en-US/kb/mozilla-crash-reporter-tb#thunderbird:win10:tb128, visit %APPDATA%\Thunderbird\Crash Reports\, open submit.log - post the contents here.

Flags: needinfo?(submarino)

(In reply to Wayne Mery (:wsmwk) from comment #27)

per https://support.mozilla.org/en-US/kb/mozilla-crash-reporter-tb#thunderbird:win10:tb128, visit %APPDATA%\Thunderbird\Crash Reports\, open submit.log - post the contents here.

It's an empty file (0 Kb). But I have a file on "submitted" folder (bp-0f7c0c21-0014-4c38-af5c-bcdcd0250918)`.

Flags: needinfo?(submarino)

I've learned the "restart" choice is not given because "Those are likely crashes caught by Windows Error Reporting. We don't support restarting the application in those cases because the environment could have been poisoned by an attacker, and that might leverage the crash reporter to launch an application that is not Firefox/Thunderbird. That's why there's no restart button." per gsvelto

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

Attachment

General

Creator:
Created:
Updated:
Size: