Last month, I received an unusually high bill due to excessive memory consumption by one of my services. Upon checking the "Metrics" tab, I noticed a steady increase in memory usage, eventually reaching 1.4 GB. I managed to solve it by triggering a new deployment and since then, the memory usage has stayed at a normal level (~250 MB).

The main issues I'm facing are:

• I can't access HTTP logs or older-than-30-days Metrics' charts, which prevents me from pinpointing when the memory spike began.

• I can't SSH into my container, so:

  • I’m unable to take memory snapshots without implementing a complex solution, like deploying a custom Docker image with a script for taking snapshots and uploading them somewhere.

Any ideas on how to debug this would be greatly appreciated 🙏🏽