500 Internal Server Error when sending large requests

[Henrik Jönsson]

Hello!

We are running Google Endpoints using serverless setup with Cloud Run.

When large requests (~15Mb JSON) is sent to any endpoint (existing or non existing) we get 500 Internal Server Error as response - and then (from the logs in Log Viewer) it looks like ESPv2 is restarting.

I created a issue on GitHub (https://github.com/GoogleCloudPlatform/esp-v2/issues/553) - because I thought it was related somehow to our setup/code.

But today I tried creating a new ESP instance in Cloud Run, and followed steps in official tutorial: https://cloud.google.com/endpoints/docs/openapi/get-started-cloud-run. And when I send large JSON requests (for instance to a non existing endpoint that normally should return 404), we get 500 Internal Server Error response instead.

I have tried running ESP with debug flag, but it didn't seem to give any more clues to what the problem is. I have also tried increasing memory limit in Cloud Run, but still same problem.

Hopefully you can reproduce the issue just by following same tutorial as I did. Otherwise, I have a running test Cloud Run instance that I can share the URL for.

Any help in resolving this issue would be greatly appreciated - because this issue is affecting our production environment.