Chromium/Firefox rebooting system upon startup after updates - how to fix it?

My system is crashing when I try to start Chromium/Iridium/Firefox after recently updating it. It was working just fine before these updates I did in the past day or two.

Iridium did start for like 2 mins but system automatically crashed after that - Chrome/FF instant crash .... right now able to write from midori browser which seems ok till now.

I tried looking into /var/log/messages but it doesn't seem to even show anything related to a crash - any tips on where/how to fix this?

I also tried scrubbing the zfs system on my single hard drive - it ran fine without any errors. Also tried to change to an earlier boot environment but somehow same issues.
 
Checklist, probably incomplete:
  • What does "crashing" mean? Immediate reboot? "Frozen" system? something else?
  • Which version of FreeBSD?
  • What's your machine (architecture, specs, ...)?
  • Official binary packages or built yourself? If the latter, any custom options, anything in make.conf?
  • Which versions of the browsers?
(and, maybe, with all of that a PR would make sense....)
 
Which version of FreeBSD?
13.1-RELEASE-p6 FreeBSD 13.1-RELEASE-p6 GENERIC amd64
What does "crashing" mean? Immediate reboot? "Frozen" system? something else?
Reboot
What's your machine (architecture, specs, ...)?
amd64
Official binary packages or built yourself? If the latter, any custom options, anything in make.conf?
official binary packages
Which versions of the browsers?
chromium seemed like latest version from packages or round about - difficult to open and check


wondering if it is something with the latest updates 🤔
 
It's somewhat strange - it keeps rebooting whenever I try to open something that's somewhat memory intensive (even for Midori browser it keeps rebooting when opening a few pages, that's the impression i got)

I checked htop and sysctl they seem to be reporting RAM as is on the system.

I do remember hearing a beep sound all of a sudden earlier maybe indicative of some kind of hardware failure?

Does anyone know how to proceed?

How do I even determine the cause of reboots?


RAM values seem fine, harddisk scrub didn't throw any errors. What else?
 
Tried to stress test cores for 60 seconds using this: Post in thread 'CPU cooling info FYI' https://forums.freebsd.org/threads/cpu-cooling-info-fyi.60214/post-346079

The cores heat up, but no shutdown/reboot as with FF/Chome

So I guess that rules out a faulty CPU?

Also tried to see if battery was the issue - seems to work without power supply (I always keep it plugged in) for last 10 mins.

So I guess that rules out faulty power supply issues?

Already ruled out harddisk with Zfs scrub and RAM with values from sysctl - what could be the root of these crashes/reboots? 🤔
 
You might miss messages on the console.

Try running the browser in a remote X11 over ssh, with the offending machine sitting in the text console.
 
Seems like a rather lengthy process. Was hoping for some easier way out. Is there some other way to find out what the issue is? (I tried to look into /var/log/messages but couldn't find anything there, some other locations?)
You might miss messages on the console.

Try running the browser in a remote X11 over ssh, with the offending machine sitting in the text console.
I actually only have this one machine currently, unfortunately. I might have to go buy another laptop to do this 😢 (running out of time with some work I need to complete). I do an extra HDD though that's a partial copy of the current SSD from a couple of months back. Maybe I'll try running that and see if it still crashes when I run these memory intensive programs.

Update: I tried to run it with the older HDD I had, it crashed, just like with SSD, on the very first time I tried to run it. I am starting to suspect it's something physically wrong with the machine and not a software issue. My main suspects are 1) CPU overheating causing shutdown 2) some power issue (my charger did get yanked a bit yesterday)
 
Wondering if that one random beep sound I heard all of a sudden earlier might be indicative of some kind of hardware failure 🤔

I went through this link https://kb.iu.edu/d/afzy - not conclusive (i also don't remember the nature of the sound except it was longish) . How do I check if video adapter is working fine? Or some other basic checks I could do?
 
You can try to cleanup the config file from your user profile and start the browser without hardware acceleration (--disable-gpu flag)
 
You can try to cleanup the config file from your user profile and start the browser without hardware acceleration (--disable-gpu flag)
Thanks.

So I tried doing this
1) first time it rebooted simply upon logging in.
2) second time I tried to open chrome using "chrome --disable-gpu" from cli. Rebooted again
3) third time tried "chrome --disbale-gpu-vsync". Rebooted again.

Config file: do you mean I should try booting into single user mode? I did try that earlier as root was logged in and didn't reboot, but didn't run any applications from the cli.

Sometimes when I only have a terminal on it doesn't reboot, but sometimes when trying to open applications it does ( as stated above). At this point I'm not sure what's wrong with the system at all, or if the same problem is getting worse ☹️
 
Update: So when I go into boot screen and enable safe mode from boot options it seems to not crash when I open Firefox 💡

Guess that means it's not a hardware issue?

Although it doesn't boot anymore with a Ubuntu stick and says something like: not syncing.... kernel panic... Stack corrupted...

So weird - now I am confused - but this safe mode seems to be giving some hint?
 
if you read what safe mode does, it disables all extensions and such that you have enabled.
a quick test:
reboot system so you know you are starting fresh
cd $HOME
mv .mozilla .mozilla.ORIG
start firefox and see what happens

If you do not get a reboot then it is likely something in your default profile, a setting, an extension, whatever.

Although it doesn't boot anymore with a Ubuntu stick and says something like: not syncing.... kernel panic... Stack corrupted...

So booting from a "live system" type of media, it has problems? Then you have hardware issues some where. First thing I would try is dust. Get rid of any dust in the system. It stops fans which make things over heat which causes issues like sporadic reboots.

What type of hardware and how old is it?
 
if you read what safe mode does, it disables all extensions and such that you have enabled.
I see, thanks. I thought it would be something like that - to be clear I didn't change any settings or enable extensions recently, except for an update and then changed Boot Environment to latest one - after which problems _seem to have_ started (difficult to pinpoint exactly). Reverted BE to earliest available for now.
a quick test:
reboot system so you know you are starting fresh
cd $HOME
mv .mozilla .mozilla.ORIG
start firefox and see what happens
So this reboot issue also happens with Chrome as well, as well as iridium, does it still make sense to rename FF config files?
So booting from a "live system" type of media, it has problems? Then you have hardware issues some where. First thing I would try is dust. Get rid of any dust in the system. It stops fans which make things over heat which causes issues like sporadic reboots.
I do suspect that the thermal paste might have dried up - that's a hunch. Fans seem to be working fine, temp used to shoot up earlier too depending upon workload. Maybe applying thermal paste might fix it? Although it wouldn't really work in safe mode with FF if that the issue, right?🤔

Or else could be a hardware issue ( in which case it might be cheaper to get another laptop instead of replacing the motherboard parts)
What type of hardware and how old is it?
Almost a 10 yr old laptop. 0 th gen machine with i7 - but was working perfectly fine till a couple of days back until, about, I installed updates

I think you could try it
Set dumpdev="AUTO" to /etc/rc.conf and check after rebooting /var/crash folder
I actually have dumpdev set to auto in rc.conf . However /var/crash doesn't seem to have any records other than a file called minfree with only the number 2048 as it's content.
 
Whatever might be wrong with some browser profile (sure, I've seen very weird issues caused there), it should never crash the OS/kernel. In my experience, -RELEASE kernels don't crash. This doesn't mean it's "unthinkable" of course, it just means some hardware issue is much more likely. But in case it is NOT a hardware issue, some information (console output, crash dump) is definitely needed to debug it 🤷‍♂️
 
Whatever might be wrong with some browser profile (sure, I've seen very weird issues caused there), it should never crash the OS/kernel. In my experience, -RELEASE kernels don't crash
Yes, i suspected the same earlier that it's unlikely to be a software issue - reason: reboot issue across multiple hard drives and with different browsers. Hard drive not showing errors. RAM test not showing errors either.

For now i entered latest Boot Environment (in safe mode) and updating to 13.1-p7 .... Maybe something with last update/changing BEs might have broken something?

Update: It did not get updated when I tried to reboot. Uname -a still showing 13.1-p6 somehow. And when I tried to update fetch in normal mode it just reboots again 🤔
 
But in case it is NOT a hardware issue, some information (console output, crash dump) is definitely needed to debug it 🤷‍♂️

Aye. Until it's sorted out, it couldn't hurt to crank up the logging to dmesg and so to /var/log/messages, by adding to /boot/loader.conf:

verbose_loading="YES"
boot_verbose="YES"

and scanning for clue(s).
 
Aye. Until it's sorted out, it couldn't hurt to crank up the logging to dmesg and so to /var/log/messages, by adding to /boot/loader.conf:

verbose_loading="YES"
boot_verbose="YES"

and scanning for clue(s).
Ok will try this and post dmesg from safe mode (since normal crashes).

Also - in safe mode Chrome and FF both open fine (chrome has about 2k tabs and FF a few hundred) - so I'm guessing if it was a hardware issue it wouldn't work even in safe mode? Or maybe some kind of acceleration?

Temperature also seems fine in safe mode - around 80 C , earlier in normal mode used to even touch 100 C depending on workload. Loads also seem to be 3-4ish - nothing crazy.

Will post dmesg soon below with verbose enabled
 
Ok will try this and post dmesg from safe mode (since normal crashes).

I was thinking more that you might get some clues from /v/l/messages after it did crash ...

Temperature also seems fine in safe mode - around 80 C , earlier in normal mode used to even touch 100 C depending on workload. Loads also seem to be 3-4ish - nothing crazy.

80C is no way fine, and 100C is critical. Offhand, I'd say that's likely your problem right there, especially when initialising hundreds of browser tabs.

Fan/s, dust, heatsink paste; try getting normal operating temperature down to more like 50-60C, 80C max on big compilations, long rendering jobs etc.
 
I was thinking more that you might get some clues from /v/l/messages after it did crash ...
The thing is /var/log/messages doesn't seem to be picking up any obvious errors, but will paste it here in below dmesg. Here is the log for dmesg (had to use pastebin since it was too big)

Here is the (partial) output (from bottom) of /var/log/messages (had to use pastebin since it was too big)

80C is no way fine, and 100C is critical. Offhand, I'd say that's likely your problem right there, especially when initialising hundreds of browser tabs.

Fan/s, dust, heatsink paste; try getting normal operating temperature down to more like 50-60C, 80C max on big compilations, long rendering jobs etc
I thought the same initially - but then when FF/Chrome worked in safe mode - it lead me to believe that cpu temp couldn't be the reason - although I do have to get the thermal paste reapplied anyways. Would it still have worked in safe mode if thermal paste was the issue?
 
Back
Top