Proxmox Node freezes and requires a reboot

P

phanos

Renowned Member
Oct 23, 2015
33
2
73
Hi,

I have two nodes running proxmox pve. The one is running version 4.4 for some years now and is rock solid. Never freeze and always responding. I tried upgrading it last year to version 7.1 but I immediately notice that after the upgrade I was getting very frequently freezes so I reverted back to version 4.4. The old version (4.4) is very solid (even if it is unsupported at this point).

After my failed attempt to upgrade my node above I decide to setup another one with the latest proxmox (version 7.2) to see how stable it is. Since then I regularly upgrade the second node and reboot when needed so I am running the latest version. Right now I am running version promox 7.4.3 and kernel 5.15.104-1-pve.

The second node started with only 12 GB of ram so I only ran one VM at this node (HomeAssistance Linux OS) which uses up to 8GB of ram. At first it was running just fine but after some time I notice that in irregular intervals this node will freeze and need a restart.

At first I thought it was just random and maybe caused by power surge but after looking into to it more I notice that before the freeze the node will seem to be running out of memory. The command "cat /proc/meminfo" under "MemFree:" it will report (right before the freeze) memory below 250MB.

Since I also run backups for this VM every night at 4:00 (consolidated the last 8 versions of the backup) I simply thought that the memory of the server was not enough so two months ago I installed another 4GB or ram (now total 16GB). I still only run the one VM I mentioned above with the same memory requirements (8GB).

At first it was running just fine (for a couple of days) but then it started freezing again. Again I notice that before any freeze the free memory will go below 250MB. For example last night the MemFree was arround 210MB for at least 30 minutes before it freezed at 2:35 in the morning.

So i would like to ask you. Are there any issues with version 7.xx that still need to be fixed? If not can someone help me troubleshoot the problem more so I can find some solution?

I will also like to upgrade my first node to the latest version but I am worried that it will become unstable. In the first node I have 32GB ram and ran 12 VMs (Linux) and the memory utilization is below 24GB. I do not run any automatic backups on this node though, I only manual run backups when I need to. Does this matter?

Phanos
 
Hi,

after 1-2 days I notice that again the free memory is very low 220MB. After checking the server this is what I get about memory on the server right now

ps -eo pid,ppid,cmd,comm,%mem,%cpu --sort=-%mem | head -25 PID PPID CMD COMMAND %MEM %CPU 3824 1 /usr/bin/kvm -id 100 -name kvm 50.5 26.8 3791 1 pveproxy pveproxy 0.9 0.0 817 1 /lib/systemd/systemd-journa systemd-journal 0.9 0.0 31304 3791 pveproxy worker pveproxy worker 0.8 0.0 31303 3791 pveproxy worker pveproxy worker 0.8 0.0 31305 3791 pveproxy worker pveproxy worker 0.8 0.0 3697 3694 pvedaemon worker pvedaemon worke 0.8 0.0 3696 3694 pvedaemon worker pvedaemon worke 0.8 0.0 3695 3694 pvedaemon worker pvedaemon worke 0.8 0.0 3694 1 pvedaemon pvedaemon 0.7 0.0 3780 3733 /usr/bin/gnome-shell gnome-shell 0.6 0.0 3810 1 pve-ha-lrm pve-ha-lrm 0.6 0.0 3723 1 pve-ha-crm pve-ha-crm 0.6 0.0 3940 1 pvescheduler pvescheduler 0.6 0.0 3480 1 pve-firewall pve-firewall 0.5 0.2 3494 1 pvestatd pvestatd 0.5 0.3 2943 1 /usr/bin/pmxcfs pmxcfs 0.3 0.0 3808 1 spiceproxy spiceproxy 0.3 0.0 31302 3808 spiceproxy worker spiceproxy work 0.3 0.0 2620 1 /usr/bin/tvheadend -f -p /r tvheadend 0.2 0.1 3647 3645 /usr/lib/xorg/Xorg vt1 -dis Xorg 0.2 0.0 2207 1 /usr/lib/snapd/snapd snapd 0.2 0.0 2996 1 /usr/sbin/veeamservice --pi veeamservice 0.2 0.0 3473 3050 /usr/libexec/goa-daemon goa-daemon 0.1 0.0

Is the memory consumption by proxmox normal?
 
You must log in or register to reply here.
Top Bottom