MongoDB service kill itself abruptly with no reason
Yesterday at night I started seeing timeouts in my streamlit app. When I checked the mongodb logs, I saw the following sequence of logs that I attach in the image.
Checking the metrics, both CPU and Memory were completely normal (low usage actually).
Is it possible to ask for more internal logs to check how this happened?
Thank you so much!
11 Replies
Project ID:
24229b64-e963-4d48-8c97-2f4befd0cf5d
24229b64-e963-4d48-8c97-2f4befd0cf5d
those are the only logs railway would have, but those are json logs, you can click the little plus icon on the left of a log line when hovering
you said cpu and mem look fine, and discord is telling me you're on the hobby plan, so my first thought would be you ran out of disk space on the 5gb volume
Hi brody!
Thanks for your quick anwser. The app is not lauched yet, so the disk space used is also super low (less than 1%). This mongodb instance has been having little traffic so far.
Yes, the first log I see is this (image attached). But the detail gives me little info. I look it up on the internet, and it says its usually a problem of RAM or CPU usage.
Not my case tho 😦
there are more logs than just that one to look into
Before that all logs were the same, the usual ones of mongo ops.
That log, the "signal was sent by kill", is the one that triggered the shutdown of the service.
I will try and check if it happens again
is the database online right now?
Yes, a simple restart did the trick.
very odd
Yes. The service itself has been up for a couple of months (on and off) with no problem whatsoever. Always with the same low traffic/usage.
Will keep an eye on it, to see if the error comes back.
Thanks!
to be honest there isn't too much railway can do about this though, the databases are all completely unmanaged