R7800 100% CPU Spikes

  • ATTENTION! As of November 1, 2020, you are not able to reply to threads 6 months after the thread is opened if there are more than 500 posts in the thread.
    Threads will not be locked, so posts may still be edited by their authors.
    Just start a new thread on the topic to post if you get an error message when trying to reply to a thread.

Warlord1981

Occasional Visitor
Hello everyone! Can you please help me understand & troubleshoot below 100% CPU spikes (mostly system, user & softirq) that happen repeatedly every 7 mins? Just installed Netdata and noticed this.

All network devices are idle during all this time, except very light browsing on one PC.

I have latest Voxel fw + Kamoj Addon latest version + Dncrypt v2 + Entware & Aegis on USB. No VPN.

cpu.PNG
 
Last edited:

ducdeptrai1234

Occasional Visitor
You are more lucky than me, I got the spike every 2 minutes. Moreover, I only used voxel + aegis + entware :)
 

Attachments

  • Screen Shot 2021-02-09 at 4.05.26 PM.png
    Screen Shot 2021-02-09 at 4.05.26 PM.png
    361.5 KB · Views: 12

Warlord1981

Occasional Visitor
Every 2 minute might be if you have band width monitoring on.
Check with top2 command to se process name!
I just checked with top2 command via SSH and actually recorded a video during this 100% CPU spike that happens every 7 mins.

It lasts about 20sec and during this time the two commands that I notice on top2 with higher than normally cpu usage (NOT 100% but about 3% each) are haveged and addon_info_upda

I read about haveged (it's in entware library) that it Feeds the kernel entropy pool by timing CPU loops, but don't really get what it means. Regarding addon_info_upda it must indicate that some addon tries to update? I have all addons disabled except netdata and there are no update Cron Jobs. Any thoughts?

Please note that Aegis, Dnscrypt & Bandwidth Monitor are all disabled during this test.
 
Last edited:

ducdeptrai1234

Occasional Visitor
Every 2 minute might be if you have band width monitoring on.
Check with top2 command to se process name!
Thanks kamoj, disable bandwidth monitoring in Advanced -> Kamoj Menu -> Setiings menu helps alot.
 

Attachments

  • Screen Shot 2021-02-10 at 10.23.05 AM.png
    Screen Shot 2021-02-10 at 10.23.05 AM.png
    219 KB · Views: 13

kamoj

Very Senior Member
Is it 100% or 3%, I'm confused now.
If 3% every 7 minute it might be "that you have enabled "Show idle devices in device lists" in settings.
There is a sub-routine pinging all addresses every 7 minute to keep the ARP table updated.

(When you talk CPU load, remember that this CPU has 2 kernels, and really difficult to load both cores to 100%.
Also most of these background processes are running with lowest possible nice priority, meaning that they
would be interrupted by other processes when needed.)

Another note is that netdata itself is a "big" cpu consumer...
It's like quantum physics, you can't measure without disturbing/changing the real values ;-)
I just checked with top2 command via SSH and actually recorded a video during this 100% CPU spike that happens every 7 mins.

It lasts about 20sec and during this time the two commands that I notice on top2 with higher than normally cpu usage (NOT 100% but about 3% each) are haveged and addon_info_upda

I read about haveged (it's in entware library) that it Feeds the kernel entropy pool by timing CPU loops, but don't really get what it means. Regarding addon_info_upda it must indicate that some addon tries to update? I have all addons disabled except netdata and there are no update Cron Jobs. Any thoughts?

Please note that Aegis, Dnscrypt & Bandwidth Monitor are all disabled during this test.
 

Warlord1981

Occasional Visitor
Is it 100% or 3%, I'm confused now.
If 3% every 7 minute it might be "that you have enabled "Show idle devices in device lists" in settings.
There is a sub-routine pinging all addresses every 7 minute to keep the ARP table updated.

(When you talk CPU load, remember that this CPU has 2 kernels, and really difficult to load both cores to 100%.
Also most of these background processes are running with lowest possible nice priority, meaning that they
would be interrupted by other processes when needed.)

Another note is that netdata itself is a "big" cpu consumer...
It's like quantum physics, you can't measure without disturbing/changing the real values ;-)

Many thanks you were right! Disabled "Show idle devices in device lists" in settings, Rebooted and now no spikes! :)

Netdata was showing 100% but top2 3%.
 

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top