Latest version uses a lot of resources

[sevensolutions]

Describe the bug
I'am trying to host the latest version but it's using a lot of resources, so my machine becomes unresponsible.
The machine has 4 cores and 8GB RAM hosted on Hetzner Cloud.

To Reproduce
Steps to reproduce the behavior:

1. Setup using docker compose as mentioned here: https://oneuptime.com/docs/installation/docker-compose
2. After starting the stack, CPU usage is at 100% continuously and even disk IO is at 1-1,5 GBps

Expected behavior
It is ok to use more resources when starting up while migrations are running etc.
But i'am expecting CPU usage to go down again after a few minutes.
I'am also running an older version (7.0.3040) on a similar server which runs at 3-5% CPU.

Am I doing anything wrong?

Screenshots

Especially the nodejs processes need a lot CPU.

Deployment Type
Self Hosted, Docker compose.
Ubuntu 24.04

Additional Notes
I think that even the mentioned 20 GB Disk for a homelab aren't enough, because with all the images pulled, I'am already at 37GB.

[sevensolutions]

Ok, got it working on a machine with 6 cores and 16GB RAM.
But shouldn't there be two global probes connected?
At least the containers are running but they don't show up in the admin dashboard.
How can i connect them?

The probe container outputs:

Error in fetching monitor list
BadDataException [Error]: Probe ID not found
    at Function.getProbeId (/usr/src/app/Utils/Probe.ts:11:13)
    at Function.getDefaultRequestBody (/usr/src/app/Utils/ProbeAPIRequest.ts:9:26)
    at FetchListAndProbe.fetchListAndProbe (/usr/src/app/Jobs/Monitor/FetchList.ts:79:32)
    at FetchListAndProbe.run (/usr/src/app/Jobs/Monitor/FetchList.ts:43:20) {
  _code: 400
}

EDIT: Found the problem:
If i change the ONEUPTIME_HTTP_PORT from 80 to lets say 8080, which i need in order to free the port 80 for my own proxy, then the probe URLs are wrong, because they point to http://localhost (port 80).
So i changed them to 8080 too and now the probes are connected.