Server Crashes During Reporting on my Virtual Machine

Jonas Asum5 days ago

I am currently stress testing my server, and whenever I request heaps of Reports (the last month of positions on one account and on another the same, and then another API request with that), my server is just overloading and crashing, and nobody can use it anymore. I currently have it running on an Azure VM with this pricing: Standard B1s (1 vCPU, 1 GiB memory). I just have 40 devices on there yet, so I don't need too much, but is it crashing because my size is too small or because there is some configuration error? Usually, my CPU is hovering at around 2-3%, but then I have these spikes of CPU usage.

Screenshot 2024-12-18 at 6.17.29 PM.png

Screenshot 2024-12-18 at 6.19.50 PM.png

Screenshot 2024-12-18 at 6.20.28 PM.png

Screenshot 2024-12-18 at 6.18.45 PM.png

Screenshot 2024-12-18 at 6.17.39 PM.png

Screenshot 2024-12-18 at 6.20.46 PM.png

Track-trace5 days ago

Both cpu and memory is very low. Why dont you test is with more cpu and memory. Thats the whole point of cloud computing, that you can scale up if needed.

P.s.
Logicly this has not much to do with the amount of devices, but the amount of data that is stored in the database for the specific timeframe. So it will also depend on the devices upload frequency.

What made you think you can do a lot with 1 vcpu and 1 gb ram?

Anton Tananaev5 days ago

Memory is definitely not enough. I recommend at least 2GB or ideally even more.

Jonas Asum5 days ago

Ok Thanks, I just thought it would have been a waste, cause in the "normal state" I am using so little, appreciate it!