Sudden shutdown

[dark.initr0]

Hello!
Over a year I experience a problem with my 10y old computer - sudden shutdown (not reboot - just power turns off in 1 second).
I have FreeBSD 13.2-STABLE + Xorg + XFCE4 on my home computer.

I'm guessing this problem related with overhitting, but I'm not sure.
I don't load my CPU with any heavy usage CPU programs. Mostly I just read web.

Could you recommend me a soft / steps to find root on this problem?
thnx!

 

[Crivens]

Is it a Thinkpad? Some of them are prone to do this when the wlan card develops electrical problems. Remove it, carefully clean the contacts (You do have ESD gear, yes?) and refit. That helped my x230, but I need to do it again soon.

 

[Aminavy]

You can see the cpu temperature:
sysctl hw.acpi.thermal.tz0.temperature

note: there is not 13.2-STABLE. we have 13-STABLE and 13.2-RELEASE (or 13.3)

 

[dark.initr0]

Is it a Thinkpad? Some of them are prone to do this when the wlan card develops electrical problems. Remove it, carefully clean the contacts (You do have ESD gear, yes?) and refit. That helped my x230, but I need to do it again soon.

This is not a laptop.

 

[dark.initr0]

You can see the cpu temperature:
sysctl hw.acpi.thermal.tz0.temperature

note: there is not 13.2-STABLE. we have 13-STABLE and 13.2-RELEASE (or 13.3)

I don't have such sysctl address. Maybe missing some driver?

"13.2-STABLE" I see in my "uname -a"

 

[Crivens]

Then do a "sysctl -a | grep temp", that should return a bit of info

 

[dark.initr0]

Then do a "sysctl -a | grep temp", that should return a bit of info

I have only

vm.pfault_oom_attempts: 3
net.inet6.ip6.use_tempaddr: 0
net.inet6.ip6.temppltime: 86400
net.inet6.ip6.tempvltime: 604800
net.inet6.ip6.prefer_tempaddr: 0
hw.usb.template: -1

 

[nxjoseph]

Then do a "sysctl -a | grep temp", that should return a bit of info

This might need coretemp or amdtemp to be loaded if it isn't already loaded. coretemp is for intel CPUs and amdtemp is for amd CPUs.

 

[drhowarddrfine]

How often does it happen? At the same time?

 

[Cath O'Deray]

"13.2-STABLE" I see in my "uname -a"

Full output of `uname -a`, please, if you can. Bear in mind, it's outdated (snapshots of 13-2-STABLE are no longer offered, and so on).

---

For comparison:

% fetch -o /tmp/revision.txt -q https://download.freebsd.org/snapshots/amd64/13.3-STABLE/REVISION && cat /tmp/revision.txt
54030e8a5df5
%

– that is, <https://cgit.freebsd.org/src/log/?h=stable/13&qt=range&q=54030e8a5df5> (a range ending four days ago).

 

[Jose]

This is not a laptop.

Some server-class computers will shut themselves off after some time if their Lights Out Management port is not connected.

 

[LkderS4KGc2bFd63rM]

Looks like a hardware problem. How often does this happen? Does the problem still occur on one stick of RAM?
If not RAM, then the power supply. And if it's not the memory and the power supply, I would restore the BIOS to factory defaults and take the computer to a service center.

 

[SirDice]

note: there is not 13.2-STABLE. we have 13-STABLE and 13.2-RELEASE (or 13.3)

It's 13.3-STABLE at this time, 13.2-STABLE is entirely possible if stable/13 was built before releng/13.3 was branched off.

TYPE="FreeBSD"
REVISION="13.3"
BRANCH="STABLE"

newvers.sh « conf « sys - src - FreeBSD source tree

cgit.freebsd.org

The version changes from 13.2-STABLE to 13.3-PRERELEASE to 13.3-STABLE.

src - FreeBSD source tree

cgit.freebsd.org

src - FreeBSD source tree

cgit.freebsd.org

 

[dark.initr0]

This might need coretemp or amdtemp to be loaded if it isn't already loaded. coretemp is for intel CPUs and amdtemp is for amd CPUs.

thanks! I have loaded this AMD module and have now such picture (from sysctl):

dev.cpu.5.temperature: 57.0C
dev.cpu.4.temperature: 57.0C
dev.cpu.3.temperature: 57.0C
dev.cpu.2.temperature: 57.0C
dev.cpu.1.temperature: 57.0C
dev.cpu.0.temperature: 57.0C

I will try to make a cron script to log every minute temperature and will analyze results when have a problem next time.

 

[dark.initr0]

How often does it happen? At the same time?

Random time - maybe 2 times a week.

 

[SirDice]

thanks! I have loaded this AMD module and have now such picture (from sysctl):

Is that at idle? That seems excessively hot. If this is an old system you might want to check your CPU fan and cooler for dust bunnies. And redoing the thermal paste will probably help too.

 

[rooksis]

Maybe you have some script that updates large git repos? Some time ago I owned the x220T and it could handle almost everything, except cloning src and ports, it overheated and rebooted.

 

[drhowarddrfine]

dark.initr0 By any chance were you doing any new installations or fiddling with settings?

I posted here a while back about random shutdowns myself. This wasn't too long after doing some things with the linuxulator. I didn't get to finish what I was doing but those settings and some things I installed are still there which makes me wonder if that started the problem.

 

[Crivens]

For what it's worth, my x230 does this also while running other OSes. This can be a hardware thing.

 

[dark.initr0]

Maybe you have some script that updates large git repos? Some time ago I owned the x220T and it could handle almost everything, except cloning src and ports, it overheated and rebooted.

no, I didn't such operations

 

[dark.initr0]

dark.initr0 By any chance were you doing any new installations or fiddling with settings?

I posted here a while back about random shutdowns myself. This wasn't too long after doing some things with the linuxulator. I didn't get to finish what I was doing but those settings and some things I installed are still there which makes me wonder if that started the problem.

I have compiled kernel (just commented some lines). I don't use linuxulator.