CPU Usage too high

General Discussions
1

Anyone else at 100% usage and just running a few apps?

Lightning node is sitting around 55-65% usage and system at ~30%

Hoping there is a work around/fix for this soon, cant be running at 100% for long and over heating my node (even in a cold basement with a fan)

Running UmbrelOS 1.1

2

CPU usage may not be 100% accurate on the dashboard right now- we’re working on improving this.

For a more detailed insight, you can use these steps:

  1. Navigate to the Settings (http:umbrel.local/settings) > Advanced Settings > Terminal > umbrelOS.
  2. Run command top you may need to scroll up a little
3

I installed umbrelOS 1.1.2 on a miniPC N4000/4GB/7TB from a bootable stick as advised at GitHub - getumbrel/umbrel: A beautiful home server OS for self-hosting with an app store. Buy a pre-built Umbrel Home with umbrelOS, or install on a Raspberry Pi 4, Pi 5, any Ubuntu/Debian system, or a VPS. without problem.
After installing bitcoin node it run pretty well with internet speed 500 MB. being happy I added Seafile next day and tried to open: the system became extremely slow and inresponsive. Dashboard stopped, after closing we page using local link and its reopening - endless circling. No access from the terminal - it does not accept the password, no response … What options exist in such situation?
Wait days? Power off/on? Reinstall?
Any advise from Umbrel team or folks?

4

I actually ended up restarting, it pushed an update and now usage has dropped to ~4% which seems right lol

5

After migrating from Umbrel 0.5.4 to 1.1.2 I get 100% CPU load most of the time.

top - 13:51:12 up  4:50,  2 users,  load average: 1.61, 0.93, 0.77
Tasks: 265 total,   3 running, 262 sleeping,   0 stopped,   0 zombie
%Cpu(s): 44.5 us, 25.6 sy,  0.0 ni, 28.6 id,  0.0 wa,  0.0 hi,  1.3 si,  0.0 st
MiB Mem :   7874.9 total,    298.4 free,   5374.1 used,   2511.9 buff/cache
MiB Swap:      0.0 total,      0.0 free,      0.0 used.   2500.8 avail Mem

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   4139 root      20   0   61.1g 135428  17972 R  87.5   1.7   8:43.53 node
   9965 umbrel    20   0 3642148 969564 127244 S  11.2  12.0  19:54.61 bitcoind
   9818 umbrel    20   0 1615440 572996   2552 S   8.3   7.1  18:29.71 electrs
   9208 umbrel    20   0   63212  37512   3988 S   3.6   0.5   2:06.49 tor
    466 root      20   0 3002084 104796  23300 S   3.3   1.3   3:19.55 dockerd
   9188 umbrel    20   0 2220796 510696 161184 S   2.0   6.3   3:38.07 lnd

Notice the 61.1 GB virtmem! Seems there is a memory leak.
This comes from the node.io module but no idea what it actually does:

root@umbrel:/home/umbrel# ps -ef | grep 4139
root        4139    3926  3 09:01 ?        00:08:53 /usr/bin/node --require /usr/local/lib/node_modules/umbreld/node_modules/tsx/dist/preflight.cjs --import file:///usr/local/lib/node_modules/umbreld/node_modules/tsx/dist/loader.mjs /usr/local/lib/node_modules/umbreld/./source/cli.ts --data-directory=/home/umbrel/umbrel --log-level verbose
root        4179    4139  0 09:01 ?        00:00:11 /usr/local/lib/node_modules/umbreld/node_modules/@esbuild/linux-x64/bin/esbuild --service=0.20.2 --ping

After further investigation it seems to be related to the 3-in-1 “Live usage” CPU/Mem/Storage widget. If I disable it from the start page, CPU goes down to normal values.

6

on my pi 4, I was running close to 100% on v1.1 all the time. I just upgraded to v1.2 and it’s fluctuating from 40 ish up to 100. It’s better but something seems off still. mem usage went down with the update though.

PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND

3863 umbrel 20 0 4012076 911164 209448 S 92.4 11.4 34:24.89 bitcoind
11705 umbrel 20 0 2553924 840404 10624 R 56.1 10.5 8:28.46 node

umbrel 11705 11602 18 04:23 ? 00:08:11 node /backend/package/index.js

7

I have the same issue here… node process is eating one core on RPI4 more less all the time. I am on latest version 1.2.2