One Core is always at 90-100%

[heimdall-one]

Hey there,

 

I just built a new Unraid server.

 

Components: 

Celeron N5105 

16GB RAM

 

4 HDDs 2 SSDs

 

The HDDs are in a regular array with 2 Parity drives currently (maybe I reduce that to 1 at some time) and the SSDs are in a pool as cache.

 

 

Only 2 Docker Containers are running, MariaDB and Nextcloud

 

 

Now the CPU1 Core always hops between 90 and 99% (confirmed by htop) and I have no clue why.

 

 

Maybe someone also had this recently

[heimdall-one]

Even when all Docker containers are stopped CPU1 wont come back to sanity.

 

So the whole system stays at 30W on idle (yes, not thaaaat much, but I took the low power CPU for a reason :D)

 

According to Powertop there is no idling for any CPU Cores.

Edited January 2 by heimdall-one

[morytox]

i seem to have a similar issue after update to 6.12.6 =(. Did you get any help or resolution in the meanwhile ?

 

[heimdall-one]

13 minutes ago, morytox said:

i seem to have a similar issue after update to 6.12.6 =(. Did you get any help or resolution in the meanwhile ?

 

Unfortunately not.

 

Its also hard to determine for me, what the problem could be. 

 

Its also happening when ALL containers are shut down and all disks are spun down.

[morytox]

exactly same with me. 

It's not even similar hw. 

 

(Ryzen 2700, x470d4u)

 

With your system it was right after a fresh install ? 

[heimdall-one]

4 minutes ago, morytox said:

With your system it was right after a fresh install ? 

yes exactly. 

 

[morytox]

can you go to command line and enter top 

 

my highes cpu utilization does go to kworker and ksoftirqd

 

  293 root      20   0       0      0      0 R  61.1   0.0  47:08.86 kworker/3:1+usb_hub_wq                                        
   35 root      20   0       0      0      0 R  38.9   0.0  30:05.82 ksoftirqd/3 

 

Do you have a similiar issue ?

 

[heimdall-one]

Yes.

 

2 Cores at 100% currently.

Edited January 7 by heimdall-one

[morytox]

looks more like this thread, maybe it helps.

 

https://askubuntu.com/questions/7858/why-is-ksoftirqd-0-process-using-all-of-my-cpu

[heimdall-one]

Well the Problem is persisting, even when every container is shut down.

[JorgeB]

Try this:

https://forums.unraid.net/topic/129514-high-cpu-load-caused-by-kworker-and-ksoftirqd-usb-auto-suspend/?do=findComment&comment=1179867

 

[heimdall-one]

7 hours ago, JorgeB said:

Try this:

https://forums.unraid.net/topic/129514-high-cpu-load-caused-by-kworker-and-ksoftirqd-usb-auto-suspend/?do=findComment&comment=1179867

 

Tried that, rebooted, waited about an hour. 

 

Did nothing =( it looks like Core 1 is even higher now.

[RoTalk]

Shouldn't this be caught in the syslogs, if enabled and to write to array + usb? 

I might be having the same issue as well, replaced, usb, mem test for 2 days all good since day 1, I also noticed I cannot boot in GUI mode, and always used web.