Proxmox Shutdowns after daily activities

[lpereira]

Hi, newbie here,

Every day since I installed proxmox my laptop, it always shutdowns without any interaction. This happens only at night arround 1-3 pm.
I can only see the following logs

Nov 22 02:13:52 pve smartd[640]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed>
Nov 22 02:11:38 pve kernel: asus_wmi: Unknown key 6d pressed
Nov 22 01:55:56 pve systemd[1]: pve-daily-update.service: Consumed 2.089s CPU time.
Nov 22 01:55:56 pve systemd[1]: Finished Daily PVE download activities.
Nov 22 01:55:56 pve systemd[1]: pve-daily-update.service: Succeeded.
Nov 22 01:55:56 pve pveupdate[193518]: command 'apt-get update' failed: exit code 100
Nov 22 01:55:56 pve pveupdate[193518]: <root@pam> end task UPID:pve:0002F3F4:00613D7F:637C2C29:aptupdate::root@pam:>
Nov 22 01:55:56 pve pveupdate[193524]: command 'apt-get update' failed: exit code 100
Nov 22 01:55:53 pve pveupdate[193518]: <root@pam> starting task UPID:pve:0002F3F4:00613D7F:637C2C29:aptupdate::root>
Nov 22 01:55:49 pve systemd[1]: Starting Daily PVE download activities...

After this there is no more logs until I power up the machine again

Nov 22 08:52:13 pve kernel: Linux version 5.15.30-2-pve (build@proxmox) (gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU>
-- Boot c4b823265202417ca4bbc13398d0f09d --

What am I missing here?

 

[shanreich]

This Issue describes something similar to what you are experiencing: [1]. Are you closing the lid of your laptop, which might cause it to overheat? Have you checked your temperatures after having closed the lid for awhile?

This entry also gives me reason to believe this might have something to do with overheating:

Nov 22 02:13:52 pve smartd[640]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed>

[1] https://bugzilla.kernel.org/show_bug.cgi?id=208867

 

[lpereira]

I will check that, I can see memory is always at 90% in pve, but my 2 vms only consumes like 10% overall, which is weird

 

[lpereira]

Today happened again, however, I don't think it is the airflow_Temperature, because I saw higher values before and the PC did not shut down, here are the logs again before shutting down:

Nov 23 00:22:15 pve smartd[664]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed from 70 to 71
Nov 23 01:17:01 pve CRON[159180]: pam_unix(cron:session): session opened for user root(uid=0) by (uid=0)
Nov 23 01:17:01 pve CRON[159181]: (root) CMD (   cd / && run-parts --report /etc/cron.hourly)
Nov 23 01:17:01 pve CRON[159180]: pam_unix(cron:session): session closed for user root
Nov 23 02:16:05 pve kernel: asus_wmi: Unknown key 6d pressed
Nov 23 02:17:01 pve CRON[168743]: pam_unix(cron:session): session opened for user root(uid=0) by (uid=0)
Nov 23 02:17:01 pve CRON[168744]: (root) CMD (   cd / && run-parts --report /etc/cron.hourly)
Nov 23 02:17:01 pve CRON[168743]: pam_unix(cron:session): session closed for user root
-- Reboot --
Nov 23 09:06:21 pve kernel: Linux version 5.15.30-2-pve (build@proxmox) (gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2) #1 SMP PVE 5.15.30-3 (Fri, 22 Apr 2022 18:08:27 +0200) ()

Here is the airflow temperature values randomly across the day without shutting down:

Nov 22 21:22:15 pve smartd[664]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed from 70 to 71

Nov 23 09:06:23 pve smartd[660]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed from 71 to 80

Nov 22 08:52:15 pve smartd[664]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed from 68 to 80

Nov 22 09:22:15 pve smartd[664]: Device: /dev/sda [SAT], SMART Usage Attribute: 190 Airflow_Temperature_Cel changed from 80 to 74

I removed the battery like on the post you described, and will test it again today

 

[shanreich]

It is not the Airflow_Temperature that is the cause for your shutdowns, but rather I used it as an indicator that your laptop is probably getting very hot. 80 Degrees Celsius is a very high value for a SSD/HDD, which is an indicator that it probably gets heated up by the surrounding hardware. Usually the laptop shuts down because the CPU (or GPU) get too hot, not the Disks. The SMART attribute is merely an indicator that your laptop is probably getting too hot overall, causing the laptop to shutdown. I would monitor the CPU (and GPU if you have one) temperature and check that, since the laptop shuts down when the CPU/GPU gets too hot.

 

[lpereira]

It seems to be an issue that only happens when the lid is closed. When I open the lid this never occurs. It has nothing to do with the temperature, as the temperature is the same whether the lid is closed or not.
Maybe some configuration on /etc/systemd/logind.conf
Only have

HandleLidSwitch=ignore
HandleLidSwitchDocked=ignore

Am I missing something there?

 

[BobhWasatch]

Most laptops run at least a little warmer with the lid shut. Some run a lot warmer. I put mine on an elevated stand to increase airflow when running with the lid closed because otherwise the fans run a lot more. Since your SSD is much warmer than it should be it is not impossible that @shanreich is on the right track here.

 

[lpereira]

For now I believe you are right, I open the lid, turned the screen off, I never saw the PC showing down again.Thanks