Proxmox reboots after "Purging GPU memory"

[Demian98]

Hi,

My Proxmox server is sometimes rebooting in the night. The message "pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available." is then thousands of times in the log within a timespan of ~5 minutes.

Can someone help me to fix this?

Thank you a lot

Here are two examples:

Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:31 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Feb 23 01:03:35 pve systemd-journald[2413804]: Journal stopped
-- Reboot --
Feb 23 01:06:34 pve kernel: Linux version 6.5.11-7-pve (build@proxmox) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC PMX 6.5.11-7 (2023-12-05T09:44Z) ()
Feb 23 01:06:34 pve kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.5.11-7-pve root=/dev/mapper/pve-root ro consoleblank=60 quiet intel_iommu=on i915.enable_gvt=1
Feb 23 01:06:34 pve kernel: KERNEL supported cpus:
Feb 23 01:06:34 pve kernel:   Intel GenuineIntel
Feb 23 01:06:34 pve kernel:   AMD AuthenticAMD
Feb 23 01:06:34 pve kernel:   Hygon HygonGenuine
Feb 23 01:06:34 pve kernel:   Centaur CentaurHauls
Feb 23 01:06:34 pve kernel:   zhaoxin   Shanghai

Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
Jan 09 02:09:29 pve kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
-- Reboot --
Jan 09 02:10:17 pve kernel: Linux version 6.5.11-4-pve (fgruenbichler@yuna) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC PMX 6.5.11-4 (2023-11-20T10:19Z) ()
Jan 09 02:10:17 pve kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.5.11-4-pve root=/dev/mapper/pve-root ro consoleblank=60 quiet intel_iommu=on i915.enable_gvt=1
Jan 09 02:10:17 pve kernel: KERNEL supported cpus:
Jan 09 02:10:17 pve kernel:   Intel GenuineIntel
Jan 09 02:10:17 pve kernel:   AMD AuthenticAMD
Jan 09 02:10:17 pve kernel:   Hygon HygonGenuine
Jan 09 02:10:17 pve kernel:   Centaur CentaurHauls
Jan 09 02:10:17 pve kernel:   zhaoxin   Shanghai

 

[mauzeris]

hello,
I have the same problem, it was rock solid before, but after 8.2.2 update (or not), it started rebooting. Anyone knows whats the problem?



May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
May 03 09:38:58 test kernel: Purging GPU memory, 0 pages freed, 0 pages still pinned, 1 pages left available.
-- Boot 637dd94a45c24aa69e10614e92d90fad --
May 03 09:40:23 test kernel: Linux version 6.8.4-2-pve (build@proxmox) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYN>
May 03 09:40:23 test kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.8.4-2-pve root=/dev/mapper/pve-root ro quiet
May 03 09:40:23 test kernel: KERNEL supported cpus:
May 03 09:40:23 test kernel: Intel GenuineIntel
May 03 09:40:23 test kernel: AMD AuthenticAMD
May 03 09:40:23 test kernel: Hygon HygonGenuine
May 03 09:40:23 test kernel: Centaur CentaurHauls
May 03 09:40:23 test kernel: zhaoxin Shanghai
May 03 09:40:23 test kernel: BIOS-provided physical RAM map:

 

[leesteken]

Which GPU(s)? I have encountered at least one issue with Proxmox kernel 6.8 and its amdgpu driver on RX570.

 

[mauzeris]

its integrated VGA compatible controller: Intel Corporation CoffeeLake-S GT2 [UHD Graphics 630] (rev 02)

but.. after some digging I found that SSD which is Proxmox installation have ECC_Uncorr_Error_Count 2. I will change SSD today and look if it helps..

 

[Demian98]

For me it is "i915-GVTg_V5_4" from an Intel i5-7440HQ

 

[mauzeris]

Hello, I have replaced SSD but it didnt helped, still thousants of same lines and reboots...

 

[bymiy]

Hello, I had the same problem with the last kernel update. If you revert to one of the older kernel versions, your problem will be solved. They will probably fix the problem with subsequent kernel updates.

You can get help here.

1 - uname -r
2 - proxmox-boot-tool kernel list
3 - proxmox-boot-tool kernel pin X.XX.XX-X-pve (old version)
4 - init 6

 

[Demian98]

@bymiy Thank you for these tips. Which kernel version is working for you?

The following versions are available to me:

 

[bymiy]

It works very healthily.