Jump to content

ODROID-XU4 half the cores idle


Will Santana

Recommended Posts

Hi, gotta ask a question

Recently I just reinstalled all my XU4 based server from scratch.
Brand new armbian Bullseye with OMV on top installed via armbian-config
Then I noticed as seem bellow that half the cores (the low power ones) are idling, even when the system is fully under load on the high power cores

 

image.png.3c7f564c49a19e836adae5f10d408b3c.png


The cores are enabled in the /sys/devices/system/cpu/cpuX/online files

So the question is "what's wrong"?
I don't have prints but I remember all the cores working before

Thanks in advance

Will

Link to comment
Share on other sites

Hi,

 

I could not reproduce this behavior on my ODROID-XU4 with current Armbian (Armbian 22.02.1 Bullseye) :

 

top - 14:41:54 up  1:22,  2 users,  load average: 3,15, 0,77, 0,26
Tasks: 138 total,   9 running, 129 sleeping,   0 stopped,   0 zombie
%CPU0  : 19,5 us, 79,8 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU1  : 24,1 us, 75,2 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st
%CPU2  : 16,5 us, 82,8 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU3  : 18,5 us, 81,1 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,3 hi,  0,0 si,  0,0 st
%CPU4  : 32,5 us, 66,9 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU5  : 33,4 us, 65,6 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st
%CPU6  : 35,1 us, 63,9 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st     %CPU7  : 36,8 us, 62,3 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st
MiB Spch:   1990,5 total,   1496,8 free,    130,3 used,    363,3 buff/cache
MiB Swap:    995,2 total,    995,2 free,      0,0 used.   1792,8 avail Spch

 

xxx@odroidxu4:~$ uname -a
Linux odroidxu4 5.4.181-odroidxu4 #22.02.1 SMP PREEMPT Sun Feb 27 08:55:42 UTC 2022 armv7l GNU/Linux
 

I guess your application does not have multiple threads so it does not occupy all cores at once?

 

Link to comment
Share on other sites

On 5/22/2022 at 9:52 AM, blacki2 said:

Hi,

 

I could not reproduce this behavior on my ODROID-XU4 with current Armbian (Armbian 22.02.1 Bullseye) :

 

top - 14:41:54 up  1:22,  2 users,  load average: 3,15, 0,77, 0,26
Tasks: 138 total,   9 running, 129 sleeping,   0 stopped,   0 zombie
%CPU0  : 19,5 us, 79,8 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU1  : 24,1 us, 75,2 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st
%CPU2  : 16,5 us, 82,8 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU3  : 18,5 us, 81,1 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,3 hi,  0,0 si,  0,0 st
%CPU4  : 32,5 us, 66,9 sy,  0,0 ni,  0,0 id,  0,0 wa,  0,7 hi,  0,0 si,  0,0 st     %CPU5  : 33,4 us, 65,6 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st
%CPU6  : 35,1 us, 63,9 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st     %CPU7  : 36,8 us, 62,3 sy,  0,0 ni,  0,0 id,  0,0 wa,  1,0 hi,  0,0 si,  0,0 st
MiB Spch:   1990,5 total,   1496,8 free,    130,3 used,    363,3 buff/cache
MiB Swap:    995,2 total,    995,2 free,      0,0 used.   1792,8 avail Spch

 

xxx@odroidxu4:~$ uname -a
Linux odroidxu4 5.4.181-odroidxu4 #22.02.1 SMP PREEMPT Sun Feb 27 08:55:42 UTC 2022 armv7l GNU/Linux
 

I guess your application does not have multiple threads so it does not occupy all cores at once?

 

 

I'm using OMV ontop of armbian with multiple containers despite the processes (like greyhole and mariadb) running in the host itself
Maybe it's something related to the kernel mods that OMV does to Armbian
I've been using armbian + OMV for a few years now and never noticed this behaviour before. If it wasn't such a job switching back to an old release to test again, I would do it

I moved some containers to a C2 I had for the moment but using "half" the processing power of the XU4 is such a shame

Link to comment
Share on other sites

Just an update

I made a fresh install of the same (armbian image 22.01 and then upgraded all available packages), on another SD card and proceeded to load test again with armbianmonitor
All cores peaked arount 100% as usual. Installed OMV and even greyhole and no change. All kept working
Got back to my installation and thought the only big piece missing was the docker serivce. Stoped docker (and thus it's containers) and all the cores got back to normal. At least for a few minutes. After 1 or 2 tests I realized two cores stuck at 0% again
A few reboots (no changes made) and I was able to use all cores even with docker and some containers running as shown bellow

image.png.ec295293eaa87718b7523f38269a8982.png

 

Unfurnatelly after some load tests, some cores got stuck again

The last thing I'm thinking of is that the new installation runs out of an SSD and a new USB to SATA bridge. The SSD is healthy and fine, but maybe it's causing an overrun. Or maybe a faulty SATA brigde stalling IO operations and locking the cores

I'll be back as soon I can with further tests

Att,
Will

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines