I recently deployed the Context Chat Backend on my nextcloud host. Ever since I noticed extreme CPU usage a good chunk of the time. Nextcloud itself still ran fine so I wasn’t too bothered but I still did some digging.
The CPU usage is generated by a main.py. This main.py is started by the Context Chat Backend container. I also saw such a process that also generated tons of CPU usage and was started by the Test-deploy container. Can anyone tell me what those scripts do?
Are they supposed to run for days?
Or is the long runtime due to insufficient processing power (Running in a vm with 4 cores of a Ryzen 7 3700x 6GB RAM)
I hope someone can tell me if this is expected behaviour or if there is something wrong and can point me in the right direction.
Context Chat php app would send all the user-accessible files to the backend for indexing in the background
So I assume that is what you are seeing: It is currently chunking and producing vector embeddings of your documents in the background. However, that is just a guess, I am no expert here
No worries. I think you should ask here again if it does not stop after some days.
From my experience, long running things like this sometimes get stuck (depending on how they implemented the error handling) if there are files with strange encodings or corrupted. In that case, it is best to check the logs to see where it hangs.
.