Just upgraded to 6.9. Usage like this has been pretty normal since I upgraded, but none of my dockers are showing any utilization (like 0.x %) with an occasional spike from plex. Normal utilzation before the upgrade was less than 10%.

 

Is there any way to know what processes are using CPU to watch in real time, and to sort by utilization % (like windows task manager for instance)? I clicked the "processes" button but it's so long and not sortable, and not real time ....  Thoughts please??

 

cheers

forgot to mention, the dashboard now takes a long time to load.. way longer than it ever did before.

It is always a good idea to upload your diagnostics when asking for help (Tools -> Diagnostics)

 

Without that... have you used the Tips and Tweaks plugin to change the CPU Scaling Governor? (or manually changed the CPU Scaling Governor?) If so, change it to On Demand or Performance.

Hi tiwing,

Something is definitely happening on your machine. In your initial screenshot, note the network traffic? You may be able to gather more information about what is running on your machine by simply going into the terminal and typing: 

top

press q to exit, it should show you what is using your cpu.

Diagnostics are always helpful if this doesn't reveal the problem.

 

Hope this helps,

Del

 

thanks! "top" is exactly what I was looking for - not as "pretty" as task manager lol but gives exactly the right info. So I'm interpreting right - if the value shows 100% in the CPU column, does that mean it's using 100% of ONE thread? So in my case with 32 threads the max possible usage is 3200% ?

 

IF that's the case, then what I see on the screen doesn't match what's showing in "top" command. what I see in "top" is what I'd expect to see, but something fishy is going on somewhere. As for the network traffic, I was copying a file at the time, and my wife was direct playing from plex, neither of which I've ever seen spike CPU like that.  

 

don't have the tips and tweaks plugin. I did mess with the IPMI plugin but removed it and that's the first time I actually remember the dashboard loading slowly ... I wonder if there is some residual stuff I need to clean out

 

For giggles, my diagnostics are attached but I'll monitor for a few more days and see if I can spot something. thanks for your help!

 

tiwing

 

kscs-fvm2-diagnostics-20210304-2144.zip

Edited March 5, 20215 yr by tiwing

Turns out tdarr was stuck in a processing loop on a certain file. I had set cpus=16 on the docker which is why I only saw half the processors spiking. So totally unrelated to upgrade 6.9, just a coincidence in timing. 

 

Cheers