Costant CPU usage

V

vaschthestampede

Active Member
Oct 21, 2020
131
8
38
39
I have a PBS installation that shows constant usage, around 10%, even in the absence of tasks.

1751881958447.png

Package version:
Code: 
proxmox-backup: 3.4.0 (running kernel: 6.8.12-11-pve)
proxmox-backup-server: 3.4.2-1 (running version: 3.4.2)
proxmox-kernel-helper: 8.1.1
proxmox-kernel-6.8.12-11-pve-signed: 6.8.12-11
proxmox-kernel-6.8: 6.8.12-11
proxmox-kernel-6.8.12-9-pve-signed: 6.8.12-9
ifupdown2: 3.2.0-1+pmx11
libjs-extjs: 7.0.0-5
proxmox-backup-docs: 3.4.2-1
proxmox-backup-client: 3.4.2-1
proxmox-mail-forward: 0.3.3
proxmox-mini-journalreader: 1.5
proxmox-offline-mirror-helper: 0.6.7
proxmox-widget-toolkit: 4.3.11
pve-xtermjs: 5.5.0-2
smartmontools: 7.3-pve1
zfsutils-linux: 2.2.7-pve2

Output of top:
1751882091437.png

The problem persists even after reboot.

Please let me know what other information might be helpful.
 
Last edited:
please describe the used server platform, the used drive, cpu choice and the ram availability.
 
CPU: 4 x Intel(R) Celeron(R) CPU J3455 @ 1.50GHz (1 Socket)
RAM: 8GB
Disks: 12 x 16TB (BTRFS raid 6)
 
Last edited:
Please share
Bash: 
proxmox-backup-manager task list
ps aux -w --sort=-%cpu | head -n 15
 
Last edited:
Code: 
root@pbsQNAP:~# proxmox-backup-manager task list
root@pbsQNAP:~# ps aux --sort=-%cpu | head -n 15
USER         PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
root         937 30.9  0.4 808720 37952 ?        Ssl  11:29  20:01 /usr/lib/x86_64-linux-gnu/proxmox-backup/proxmox-backup-api
backup      1003  9.9  1.2 1105516 99348 ?       Ssl  11:29   6:26 /usr/lib/x86_64-linux-gnu/proxmox-backup/proxmox-backup-proxy
root        1325  0.2  0.1  19412 11008 ?        Ss   12:31   0:00 /lib/systemd/systemd --user
root        1356  0.1  0.1  17868 10784 ?        Ss   12:32   0:00 sshd: root@pts/0
root          17  0.1  0.0      0     0 ?        I    11:27   0:06 [rcu_preempt]
root           1  0.0  0.1 102652 12108 ?        Ss   11:27   0:01 /sbin/init
root         288  0.0  0.0      0     0 ?        S    11:27   0:01 [usb-storage]
root        1362  0.0  0.0   7992  4480 pts/0    Ss   12:32   0:00 -bash
root        1218  0.0  0.0      0     0 ?        I    11:54   0:00 [kworker/1:3-mm_percpu_wq]
root          11  0.0  0.0      0     0 ?        I    11:27   0:00 [kworker/u8:0-bnxt_pf_wq]
root         406  0.0  0.3  48960 27264 ?        Ss   11:27   0:00 /lib/systemd/systemd-journald
root        1242  0.0  0.0      0     0 ?        I    11:55   0:00 [kworker/u10:2-events_power_efficient]
root          77  0.0  0.0      0     0 ?        I    11:27   0:00 [kworker/0:2-events]
root        1315  0.0  0.0      0     0 ?        I    12:19   0:00 [kworker/3:0-events]
root@pbsQNAP:~#
 
Thanks. I just wanted to verify that there's really no task running and check the full name of the process as top cut it off.
I always have a little bit (5-10% or so) of CPU usage too. If I disable the PBS data store(s) in PVE and also stop pulse it goes down to pretty much nothing.
So basically the CPU usage seems to be caused by continuously polling it. I'm not saying it should be like that but this seems to be normal behavior.
Maybe something is constantly polling it in your case too. It might just be higher for you because your CPU is very weak.

I'm no expert at this but if you want to debug you can try to strace the process to see what it's doing like this
Bash: 
# Monitor
strace -p 937

# Summary
timeout 30 strace -cp 937
Perhaps try with f too. Also check systemctl status proxmox-backup and journalctl -ru proxmox-backup for any problems.
 
Last edited:
I know BTRFS raid 5 and 6 are not production ready.
This is a test system with little RAM so ZFS is not recommended.
 
  • Like
Reactions: Johannes S
You must log in or register to reply here.
Top Bottom