Electron CPU usage

mattn

First off, apologies for the necro! But I had to reinstall MM this weekend and came across this same issue. After a fair bit of time reinstalling/googling, I managed to resolve it by adding the following to /etc/environment

ELECTRON_FORCE_WINDOW_MENU_BAR=1

So thought I’d post it here in case anyone else experiences this.
I’ve had the default install running for a while since adding the fix and my load avg has dropped from 6+ to “0.00, 0.02, 0.02” and electron seems to peak at around 7% every ~30s for about 2s. The rest of the time it isn’t even visible in top’s list.

credit to reddit: https://www.reddit.com/r/kde/comments/5x9t7j/electron_app_high_cpu_usage/
Edit: After adding my usual modules, electron’s CPU usage has gone up to a fairly constant ~9% across 3 processes.

top - 00:04:33 up 25 min,  3 users,  load average: 0.06, 0.13, 0.13
Tasks: 156 total,   2 running, 154 sleeping,   0 stopped,   0 zombie
%Cpu(s):  2.0 us,  0.4 sy,  0.0 ni, 97.6 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem:    947732 total,   486072 used,   461660 free,    26476 buffers
KiB Swap:   102396 total,        0 used,   102396 free.   240288 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 1387 pi        20   0  300260  98384  55840 S   4.9 10.4   0:45.48 electron
 1394 pi        20   0  286500  68700  49420 S   2.0  7.2   0:23.08 electron
 1358 pi        20   0  436408  78528  57656 S   1.0  8.3   0:19.25 electron

Edit2: Left it running over night, and it’s stable:

top - 07:51:22 up  8:12,  3 users,  load average: 0.24, 0.23, 0.17
Tasks: 153 total,   1 running, 152 sleeping,   0 stopped,   0 zombie
%Cpu(s):  2.1 us,  0.3 sy,  0.0 ni, 97.5 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem:    947732 total,   543280 used,   404452 free,    34664 buffers
KiB Swap:   102396 total,        0 used,   102396 free.   271040 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 1798 pi        20   0  301180  99652  56580 S   5.3 10.5  22:54.84 electron
 1805 pi        20   0  321060  79368  50868 S   2.6  8.4  11:36.49 electron
 1770 pi        20   0  480972  82744  58648 S   1.0  8.7   7:20.91 electron

Doum

@mattn I am pretty new to linux/raspbian… I do not have any file/folder under /etc with the name environment?

Should i create a file and add this line in? add it somewhere else?

Thanks

nparab

Thanks v.much to everyone on this forum. I just downgraded Electron to 1.4.15, without downgrading either MagicMirror2 or my OS, and my CPU usage and temperature are significantly down. But I’m facing yet another problem because of this downgrading of Electron. My MMM-PIR-Sensor module is giving this error:

App threw an error during load
Error: Module version mismatch. Expected 50, got 54.
    at Error (native)
    at process.module.(anonymous function) [as dlopen] (ELECTRON_ASAR.js:173:20)
    at Object.Module._extensions..node (module.js:583:18)
    at Object.module.(anonymous function) [as .node] (ELECTRON_ASAR.js:173:20)
    at Module.load (module.js:473:32)
    at tryModuleLoad (module.js:432:12)
    at Function.Module._load (module.js:424:3)
    at Module.require (module.js:483:17)
    at require (internal/module.js:20:19)
    at bindings (/home/pi/MagicMirror/modules/MMM-PIR-Sensor/node_modules/bindings/bindings.js:81:44)
Whoops! There was an uncaught exception...
Error: Module version mismatch. Expected 50, got 54.
    at Error (native)
    at process.module.(anonymous function) [as dlopen] (ELECTRON_ASAR.js:173:20)
    at Object.Module._extensions..node (module.js:583:18)
    at Object.module.(anonymous function) [as .node] (ELECTRON_ASAR.js:173:20)
    at Module.load (module.js:473:32)
    at tryModuleLoad (module.js:432:12)
    at Function.Module._load (module.js:424:3)
    at Module.require (module.js:483:17)
    at require (internal/module.js:20:19)
    at bindings (/home/pi/MagicMirror/modules/MMM-PIR-Sensor/node_modules/bindings/bindings.js:81:44)

Any help on fixing this would be appreciated.

nparab

Fixed it! I just went to the modules/MMM-PIR-Sensor directory and did

npm install

That fixed it.

mattn

@Doum Yeah, just create the file, paste that line, save and reboot.

martinkooij

@mattn What version of electron are you running? I downgraded to 1.4.15 electron, have v6.12.3 nodejs and 2.3.1Magic Mirror. I set the environment variable as you indicated. CPU usage goes down remarkably, although not as much as in your case.

Did you also got rid of a memory leak? I start with 43% free and this falls to 7% after a day. Stays between 5%-7% for a few days and then after 3-5 days my mirror halts completely with E_NOMEM panic in the log files. Any reader knows of handy memory leak analysistools for nodejs? It seems you keep on having 50% used / 50% free… Not me ;(.

top - 20:03:02 up 21:14,  3 users,  load average: 0,30, 0,33, 0,41
Tasks: 165 total,   2 running, 163 sleeping,   0 stopped,   0 zombie
%Cpu(s):  8,6 us,  4,5 sy,  0,0 ni, 86,9 id,  0,0 wa,  0,0 hi,  0,0 si,  0,0 st
KiB Mem:    949580 total,   907688 used,    41892 free,    60200 buffers
KiB Swap:   102396 total,     5172 used,    97224 free.   444532 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 1178 pi        20   0  828888 244624  65840 R  22,5 25,8 327:41.98 electron
 1216 pi        20   0  603272 349144 290184 S  13,6 36,8 132:11.73 electron
  666 root      20   0  200848  58756  28796 S  12,2  6,2 131:39.07 Xorg
24453 pi        20   0    5224   2580   2148 R   1,0  0,3   0:02.03 top

mattn

@martinkooij TBH, I have no clue how to check my electron version, but I used the MM install script and haven’t downgraded anything. So I’m running the version supplied by the script as of approximately two weeks ago.

I’ve not noticed any mem issues. My mirror’s been up 6 days mem usage is ~10% per process:

top - 19:19:50 up 6 days,  8:37,  3 users,  load average: 0.13, 0.24, 0.20
Tasks: 155 total,   1 running, 153 sleeping,   0 stopped,   1 zombie
%Cpu(s): 32.7 us,  4.8 sy,  0.0 ni, 62.5 id,  0.1 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem:    947732 total,   674684 used,   273048 free,    96984 buffers
KiB Swap:   102396 total,        0 used,   102396 free.   288436 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 1099 pi        20   0  310396  98548  55500 S 104.4 10.4 765:21.28 electron
 1111 pi        20   0  366396 102884  51368 S  21.5 10.9 297:36.05 electron
 1071 pi        20   0  474832  84892  58020 S  12.3  9.0 244:20.94 electron

Could it be a module causing it? Or try reinstalling MM with the script and see what happens with the default install/modules

As far as my modules go, I’m only using the date and weather mods from standard and then 4 custom mods(technically 2, as 3 are separate instances of the same mod for bus/train times).

Regards,

Matt

martinkooij

@mattn thanks! I love my modules, so difficult to lose one ;).

It however seems you have a slow leak too, growing memory usage of ca 200Mbyte in 6 days. Much less than me, but still more than it should be, I guess. Just let us know how you fare! Thanks for sharing.

martinkooij

@mattn and others.
I have a stable situation with the following:

electron 1.4.15 (withELECTRON_FORCE_WINDOW_MENU_BAR=1).
higher elektron version gives black screen, the environment variable lowers the cpu-usage by elektron considerably).
Magic Mirror 2.3.1 (actually I was quick to install MM2.3.0 but manually downgraded electron, but I am quite sure that amounts to the same as installing MM 2.3.1).
nodejs version v9.11.1. (I used to have a 1.6.x nodejs, but that one caused memory leaks. )

Is now running in this configuration almost 7 days without interrupt. And no signs of stopping soon. Thanks you all in this forum for the help.

(Rasberry pi 3, Jessie, CPU Hardware: BCM2835, Revision: a02082)

cdelaorden

Hi guys

Im new to the forum and this project but I have some small experience with the Pis and having the same issue

I think that its related to the video driver, using ‘sudo raspi-config’ and changing the video driver to one of the GL drivers resulted in a much lower cpu load in my case.

Doesn’t matter if you choose full KMS or fake KMS as long as you dont use the legacy non-GL driver.

It took my 4 CPU from full 100% to idle bellow 35%

Im sure someone else with this problem can confirm this