Proxmox 8.0.3 random reboots

[lordtux]

Hi all, have proxmox installed on my mini pc with Ryzen 7840HS.
Facing random reboots.
I disabled apt automatic updates.
Disabled secure boot.
Disabled TPM and even so continue facing the strang reboots.
Checkd journalctl but not a single error, just a reboot message.
Saw that someone else "resolved" this downgrading for proxmox 8.4.
Running this kernel version
Linux 6.2.16-3-pve #1 SMP PREEMPT_DYNAMIC PVE 6.2.16-3 (2023-06-17T05:58Z)

Tried to check if there is a old kernel version but not find anything.
At this same machine I have windows installed in another SSD and not faced any type of issue.
All on the hw is on the latest version.
Had someone a clue how to fix this without having to downgrade?
Is there a way to install old proxmox kernel on the 8.0.3?
I can only see this listed:

root@promox-bee:~# proxmox-boot-tool kernel list
Manually selected kernels:
None.

Automatically selected kernels:
6.2.16-3-pve
root@promox-bee:~#

Here a log frament:

Aug 05 14:06:09 promox-bee pveproxy[1458]: worker 19641 finished
Aug 05 14:06:09 promox-bee pveproxy[19641]: worker exit
Aug 05 14:05:22 promox-bee pvedaemon[28390]: <root@pam> successful auth for user 'root@pam'
Aug 05 14:02:05 promox-bee pveproxy[1458]: worker 30368 started
Aug 05 14:02:05 promox-bee pveproxy[1458]: starting 1 worker(s)
Aug 05 14:02:05 promox-bee pveproxy[1458]: worker 22023 finished
Aug 05 14:02:05 promox-bee pveproxy[22023]: worker exit
Aug 05 13:56:28 promox-bee pvedaemon[1448]: worker 28390 started
Aug 05 13:56:28 promox-bee pvedaemon[1448]: starting 1 worker(s)
Aug 05 18:46:38 promox-bee kernel: x86/fpu: Supporting XSAVE feature 0x040: 'AVX-512 Hi256'
Aug 05 18:46:38 promox-bee kernel: x86/fpu: Supporting XSAVE feature 0x020: 'AVX-512 opmask'
Aug 05 18:46:38 promox-bee kernel: x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers'
Aug 05 18:46:38 promox-bee kernel: x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers'
Aug 05 18:46:38 promox-bee kernel: x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers'
Aug 05 18:46:38 promox-bee kernel:   zhaoxin   Shanghai
Aug 05 18:46:38 promox-bee kernel:   Centaur CentaurHauls
Aug 05 18:46:38 promox-bee kernel:   Hygon HygonGenuine
Aug 05 18:46:38 promox-bee kernel:   AMD AuthenticAMD
Aug 05 18:46:38 promox-bee kernel:   Intel GenuineIntel
Aug 05 18:46:38 promox-bee kernel: KERNEL supported cpus:
Aug 05 18:46:38 promox-bee kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.2.16-3-pve root=/dev/mapper/pve-root ro quiet
Aug 05 18:46:38 promox-bee kernel: Linux version 6.2.16-3-pve (tom@sbuild) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC PVE 6.2.16-3>
-- Boot 95d0d4562cc84435838fe772e430e567 --
Aug 05 18:42:02 promox-bee systemd[1]: Finished apt-daily.service - Daily apt download activities.
Aug 05 18:42:02 promox-bee systemd[1]: apt-daily.service: Deactivated successfully.
Aug 05 18:42:02 promox-bee systemd[1]: Starting apt-daily.service - Daily apt download activities...
Aug 05 18:17:01 promox-bee CRON[118785]: pam_unix(cron:session): session closed for user root
Aug 05 18:17:01 promox-bee CRON[118786]: (root) CMD (cd / && run-parts --report /etc/cron.hourly)
Aug 05 18:17:01 promox-bee CRON[118785]: pam_unix(cron:session): session opened for user root(uid=0) by (uid=0)
Aug 05 17:17:01 promox-bee CRON[98016]: pam_unix(cron:session): session closed for user root
Aug 05 17:17:01 promox-bee CRON[98017]: (root) CMD (cd / && run-parts --report /etc/cron.hourly)
Aug 05 17:17:01 promox-bee CRON[98016]: pam_unix(cron:session): session opened for user root(uid=0) by (uid=0)
Aug 05 16:56:45 promox-bee pvedaemon[25468]: authentication failure; rhost=::ffff:10.0.4.82 user=root@pam msg=Authentication failure
Aug 05 16:56:43 promox-bee IPCC.xs[25468]: pam_unix(proxmox-ve-auth:auth): authentication failure; logname= uid=0 euid=0 tty= ruser= rhost=::ffff:10.0.4.82  user=root

 

[Moayad]

Hi,

Based on the syslog entries you provided, there is nothing related. In this case, I would check before the reboot what happened check (monitor, syslog, CPU/RAM usage).

 

[lordtux]

Everything is normal, from the logs, from the CPU, memory, nothing wrong. And this is exactly the problem.
I removed the SDN created and will wait to see if this will fix anything.
Another thing very strang on the system, if you run the last command, you can see the latest reboots.
Question is, why is showing that reboot was performed by system. So, dont seems to be a hardware failure. Seems the system is rebooting itself with control of the OS, tried checking cron but not find anything.

 

[lordtux]

ok, even with sdn created removed, still rebooting without any clue.

 

[lordtux]

Ok, documenting here my tests, maybe can help someone else.
As proxmox is running in a minipc in my desk, I noticed when I touched it that as a little warmer, the cooler was running fast pushing air outside a little hot.
So instantly I remember the same issue due CPU temperature in my old windows laptop.
I noticed that system was stucked for some time.
Now I put in a better place with some air flow and on htop configured to show the CPU temperature, works if you install lm-sensors.
Installed sysstat as well.
I will keep monitoring.
Also, dont know yet if this is a thing of linux kernel pushing high temp on this CPU.

 

[lordtux]

Ok, nothing good until now.
On morning, rebooted once, after that no reboot until evening.
Looking on the BIOs I disabled a option for sleep but no success, system keep rebooting.
Installed lm-sensors and temperatures really normal, around 40 wiht max 50 degrees Celsius.
One thing that I notice, at the time of the reboot, the fan increased speed a lot to dissipate heat. and system was hanging.
I booted the PC on windows to see if I see some strange behavior but nothing until now.

 

[lordtux]

Ok, issue solved, after all the root cause was a bad design of bee link
the base plate dont have holes for air flow.
Is a know issue and they will send me a new base plate. Mean while I removed the base plate and so no issues with overheating.
Everything running perfect on promox now.