I have noticed that my host CPU is constantly around 25%-35% load due to usr/libexec/udisks2/udisksd :
Systeminfo:
The Proxmox runs on top of a debian bullseye - I know udisks2 isn't installed from proxmox....
I have seen various threads about this and mostly the solution is to mask the service, but as i use the host also as debian workstation, I dont want to mask the service or disable it - nor can i uninstall it due to reverse dependencies.
Interestingly (and for this I didnt see any threads) - the high load happens only when I run a LXC Container.
As soon as I stop the container, the load drops to a sane few percent. This is absolutely reproducible
Of course I tried the various suggestions from the other threads, but none of them solved the problem (unless I stop the service...)
For me the only 'solution' is to use VMs instead of containers, but if possible I'd like to avoid that.
Has anyone a better idea of a workaround or even a solution for this?
PID USER PRI NI VIRT RES SHR S CPU%▽MEM% TIME+ Command
1 root 20 0 160M 10856 7528 S 5.8 0.0 12:00.83 /sbin/init
697 root 20 0 388M 14200 10880 R 27.6 0.0 50:37.18 ├─ /usr/libexec/udisks2/udisksd
813 root 20 0 388M 14200 10880 S 4.5 0.0 9:29.05 │ ├─ /usr/libexec/udisks2/udisksd
803 root 20 0 388M 14200 10880 S 0.6 0.0 0:55.79 │ ├─ /usr/libexec/udisks2/udisksd
727 root 20 0 388M 14200 10880 S 0.0 0.0 0:00.00 │ ├─ /usr/libexec/udisks2/udisksd
748 root 20 0 388M 14200 10880 S 0.0 0.0 0:00.01 │ └─ /usr/libexec/udisks2/udisksd
Systeminfo:
pve-manager/7.2-7/d0dd0e85 (running kernel: 5.15.53-1-pve)
The Proxmox runs on top of a debian bullseye - I know udisks2 isn't installed from proxmox....
I have seen various threads about this and mostly the solution is to mask the service, but as i use the host also as debian workstation, I dont want to mask the service or disable it - nor can i uninstall it due to reverse dependencies.
Interestingly (and for this I didnt see any threads) - the high load happens only when I run a LXC Container.
As soon as I stop the container, the load drops to a sane few percent. This is absolutely reproducible
Of course I tried the various suggestions from the other threads, but none of them solved the problem (unless I stop the service...)
For me the only 'solution' is to use VMs instead of containers, but if possible I'd like to avoid that.
Has anyone a better idea of a workaround or even a solution for this?
Last edited: