No, but I have one running Blue iris with 16 cameras ranging from full HD (2K) to 4K. I have never seen temperatures like these. They are actually far below the stress test temperatures. zoneminder should not be stressing the CPU.
To be clear, there should never ever be any scenario of CPU utilization which should ever force you to disable turbo boost. These high stress situations are exactly what the boost was designed for. You bought a CPU which is supposed to run at a boosted speed for short spurs of time to address peak computational requirements. Disabling the feature should never be a solution to anything. For the specific system of the NUC of OP. The hottest version reviewed here shows that under the most stressful situation, it should never exceed 80C as the CPU throttles back but these are under sustained and abnormal test use cases :
I had zoneminder on a stock i5 nuc raised the temperature a ton. With 4 cameras (5mp, not 4k) running i would sit around 65-70 ºC in the summer. Without zoneminder I sit around 37ºC.
This is with no modifications.
Side bar, how do you like blue iris? How is it on resources? Everytime I try to go look at it, the community drives me out because they just hate the cameras I purchased (didn’t know what I was doing, but not about to buy more).
It is probably the best for flexibility and inter compatibility though I don’t like windows… (That’s the reason why I did not get HomeSeer.) What camera did you get that does not work with it? Though I admit I tend to want to avoid anything proprietary and stay as generic as possible in my choices… (ONVIF and RTSP) I though Blue iris covers a lot of others too. I have to admit there is also some learning curve… Eventually I think it makes more sense to get either a dedicated NVR or a NAS for this.
I have reolink cameras, they work fine with it from what I can tell. I just get the “you’re a dipshit for buying those vibe” from the community.
Lol… ok I did not join that community. Hope I did not give you that vibe. I am all amcrest/Dahua except for the doorbell which is RCA/Hiksvision and contributed a few fixes to the amcrest component here…
Wait this thing is doing avx steaming ? That uses a different part of the cpu, not the main core ?
Yea not sure that would show in htop ?!
So i did some testing and found out the culprit. Running my NUC normally this morning it was sitting at around 80C CPU temp. I then disabled the MotionEye add-on and within minutes the temp was down to 33C. I guess it just doesn’t like processing the 4K cameras much.
With MotionEye running CPU usage is around 17%, without it drops to 2%. So that 15% CPU usage is generating some serious heat. Are these just not efficient at video processing?
As you can see from @frits1980 feedback above, who also uses motionyeye on a much older and weaker nuc, this seems completely abnormal. I don’t use this component so I can’t help much but you may have a setup issues with it.
I also have my cpu loaded between 15% and 40% all the time and never reach past 50C. Only stress testing gets me beyond but it may vary depending on the type of load for you.
One difference is that he is only running 1x 720P camera, I have 2x 4K cameras plus 2x 1080P cameras.
It is odd though that the temp goes up so much even though the CPU load is still low. I don’t know what else to do
Take this with a grain of salt, as you can find anything on Google (even the wrong answer to a question)
https://stackoverflow.com/questions/60104698/how-do-i-monitor-the-amount-of-simd-instruction-usage
Looks like top/htop don’t show avx or other accelerated functions. So your CPU is probably heating up from avx media decoding/transcoding and not showing on top/htop
The simplest way to test it all is one by one stopping the docker containers and putting them back up. Then you can see which one is putting your CPU to that much stress. Although I can imagine that transcoding / decoding all those streams without hardware acceleration is going to be the one causing this.
if you are using docker with many containers i suggest you this: https://github.com/bcicen/ctop
so you can see who is using CPU among all containers.
BTW i have a NUC8i5 and is at 55 °C when used ( i use it as a regular pc ) and near 40 °C over night with only containers usage.
Tamb at 22 °C.
This is much more like what should be expected…
@sparkydave, you may have some setup issues with motioneye, not making use of the GPU or the hardware CODEC of the cpu for transcoding. I don’t use motioneye but I transcode 16 cams 24/7 and never exceed 56C even when doing a bunch of other things and loading my CPU upwards of 50%. I really need to abuse it with overclocking stress tools/benchmarks to get anywhere close to 85C. Either that or you have failed hardware…
Thanks. I’ll play with some MotionEye settings and see what difference it makes
This only with debian, zoneminder container stopped:
sensors
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +45.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +44.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +44.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +44.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +44.0°C (high = +100.0°C, crit = +100.0°C)
acpitz-acpi-0
Adapter: ACPI interface
temp1: -263.2°C
temp2: +27.8°C (crit = +119.0°C)
iwlwifi-virtual-0
Adapter: Virtual device
temp1: N/A
pch_cannonlake-virtual-0
Adapter: Virtual device
temp1: +44.0°C
top
command, maybe to much frontend use?
09:06:06 up 58 min, 1 user, load average: 2,11, 1,48, 1,26
Tasks: 322 total, 2 running, 320 sleeping, 0 stopped, 0 zombie
%Cpu(s): 7,5 us, 7,6 sy, 0,0 ni, 84,8 id, 0,0 wa, 0,0 hi, 0,1 si, 0,0 st
MiB Mem : 32034,9 total, 24205,7 free, 3662,3 used, 4166,9 buff/cache
MiB Swap: 16252,0 total, 16252,0 free, 0,0 used. 26979,4 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
10758 root 20 0 36624 23804 7920 R 40,2 0,1 23:22.14 frontend
10787 root 20 0 2388 1476 1344 S 9,6 0,0 5:25.40 update-se+
475 root 20 0 153004 64804 62996 S 8,6 0,2 4:54.98 systemd-j+
938 root 20 0 225824 4564 3056 S 6,0 0,0 3:11.05 rsyslogd
2027 t 20 0 3112224 224384 125248 S 3,0 0,7 0:26.11 gnome-she+
955 root 20 0 242568 9088 8044 S 2,0 0,0 0:49.80 accounts-+
1565 t 20 0 380440 97936 71000 S 1,7 0,3 0:11.56 Xorg
31006 t 20 0 616608 45884 34116 S 1,0 0,1 0:00.85 gnome-ter+
1368 root 20 0 1481900 53216 22216 S 0,7 0,2 0:05.92 containerd
13196 t 20 0 5849664 223844 88340 S 0,7 0,7 1:04.89 chromium
10 root 20 0 0 0 0 I 0,3 0,0 0:07.53 rcu_sched
68 root 20 0 0 0 0 I 0,3 0,0 0:01.03 kworker/7+
723 root 20 0 0 0 0 I 0,3 0,0 0:01.72 kworker/4+
2191 t 9 -11 1698840 25536 19908 S 0,3 0,1 0:04.12 pulseaudio
2493 root 20 0 95572 84304 10780 S 0,3 0,3 0:04.92 python2
3192 openvpn+ 20 0 71536 62820 10800 S 0,3 0,2 0:02.80 python2
3434 t 20 0 644720 140052 98428 S 0,3 0,4 0:09.47 Discord
Here is mine, looks odd that Motion is using over 100%… multiple cores I guess
top - 08:08:32 up 6 days, 21:00, 2 users, load average: 1.62, 1.29, 1.28
Tasks: 454 total, 1 running, 381 sleeping, 0 stopped, 0 zombie
%Cpu(s): 14.3 us, 1.6 sy, 0.0 ni, 83.0 id, 0.1 wa, 0.0 hi, 1.1 si, 0.0 st
KiB Mem : 16263428 total, 1316712 free, 3977280 used, 10969436 buff/cache
KiB Swap: 2097148 total, 1699580 free, 397568 used. 11965880 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
4984 root 20 0 1458808 1.272g 7444 S 117.9 8.2 7602:21 motion
9144 root 20 0 470568 276732 8676 S 5.3 1.7 692:17.01 python3
9759 root 20 0 18508 11804 4404 S 1.3 0.1 0:00.04 nmap
9404 sparkyh+ 20 0 51552 4352 3348 R 0.7 0.0 0:00.32 top
12198 root 20 0 1401796 68612 8352 S 0.7 0.4 63:44.43 node-red
1066 root 20 0 3196052 12372 2988 S 0.3 0.1 14:45.29 containerd
3170 root 20 0 346640 115548 9028 S 0.3 0.7 44:17.52 python3
4537 root 20 0 747028 19428 5868 S 0.3 0.1 9:53.18 coredns
6239 root 20 0 6885208 481264 13904 S 0.3 3.0 10:48.80 java
6314 root 20 0 1060300 38164 8204 S 0.3 0.2 19:54.31 mongod
11194 root 20 0 64756 38392 4120 S 0.3 0.2 18:12.89 python3
1 root 20 0 226432 8364 5412 S 0.0 0.1 0:10.53 systemd
2 root 20 0 0 0 0 S 0.0 0.0 0:00.15 kthreadd
3 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 rcu_gp
4 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 rcu_par_gp
6 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 kworker/0:+
9 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 mm_percpu_+
@rafale77 @sparkydave I have googled what is this frontend
process, and found few threads that report to disable secure boot. Had disabled, and result are great with enabled turbo boost, when zoneminder is detecting objects, maybe some short spike to 85C, when detecting and recording on motion:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
5487 www-data 20 0 2053736 1,6g 1,4g R 93,0 5,0 3:35.50 zma
7433 www-data 20 0 622312 76580 37440 R 30,9 0,2 0:00.93 zm_detect.py
5431 www-data 20 0 1973020 1,5g 1,4g S 21,3 4,7 1:54.85 zmc
5427 www-data 20 0 722888 304472 282404 S 10,0 0,9 1:07.57 zmc
1228 t 20 0 457044 134248 101988 S 8,0 0,4 0:25.48 Xorg
5418 www-data 20 0 1293652 874936 854600 S 5,3 2,7 0:33.73 zmc
5462 www-data 20 0 1319452 904000 848524 R 5,3 2,8 0:33.80 zma
5457 www-data 20 0 743268 328372 276504 S 5,0 1,0 0:32.29 zma
5402 www-data 20 0 720872 304128 282196 S 4,3 0,9 0:32.90 zmc
6292 t 20 0 1574348 731636 593604 S 3,0 2,2 0:24.56 zmninjapro
6542 t 20 0 510380 57992 42924 S 2,7 0,2 0:07.68 gnome-system-mo
1413 t 20 0 4002724 256976 129112 S 1,3 0,8 0:24.21 gnome-shell
5619 t 20 0 617336 49112 34156 S 1,0 0,1 0:01.94 gnome-terminal-
6243 t 20 0 1864252 132164 94708 S 1,0 0,4 0:06.18 zmninjapro
1686 t 20 0 3981696 854816 316476 S 0,7 2,6 0:22.75 chromium
5053 systemd+ 20 0 771500 207644 16820 S 0,7 0,6 0:03.32 mysqld
few seconds load with turbo, before was it on 100C, now only fan maybe spins more for few seconds sometimes:
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +85.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +68.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +85.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +66.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +66.0°C (high = +100.0°C, crit = +100.0°C)
acpitz-acpi-0
Adapter: ACPI interface
temp1: -263.2°C
temp2: +27.8°C (crit = +119.0°C)
iwlwifi-virtual-0
Adapter: Virtual device
temp1: N/A
pch_cannonlake-virtual-0
Adapter: Virtual device
temp1: +59.0°C
Idle with turbo:
sensors
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +51.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +50.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +50.0°C (high = +100.0°C, crit = +100.0°C)
acpitz-acpi-0
Adapter: ACPI interface
temp1: -263.2°C
temp2: +27.8°C (crit = +119.0°C)
iwlwifi-virtual-0
Adapter: Virtual device
temp1: N/A
pch_cannonlake-virtual-0
Adapter: Virtual device
temp1: +51.0°C
That is a little more reassuring though I don’t really know why secure boot would cause zone minder to do this. Hmmm interesting… Thanks for reporting.
Thank you for explanantion, because this pushed me to investigate further! I think NUC runs now as it should be with turbo, around 50C or 60C on big loads.