Cloud Run crashes after 121 seconds

Question

I'm triggering a long running scraping Cloud Run function with a PubSub topic and subscription trigger. Everytime I run it it does crash after 121.8 seconds but I don't get why.

POST 503 556B 121.8s APIs-Google; (+https://developers.google.com/webmasters/APIs-Google.html) https://????.a.run.app/

The request failed because either the HTTP response was malformed or connection to the instance had an error.

I've got a built-in timeout trigger and when I set it at 1 minute the functions runs without any problems but when I set at 2 minutes the above error gets triggered so it must be something with the Cloud Run or Subscription timeout settings but I've tried to increase those (read more below).

Things involved

1 x Cloud Run 1 x SubPub subscription 1 x SubPub topic

These are the things I've checked

• The timeout of the Cloud Run instance (900 sec)
• The timeout of the Pubsub subscription (Acknowledgement deadline - 600 sec & Message retention duration - 10 minutes)
• I've increased the memory to 4GB and that is way above what it's needed.

Anyone who can point me in the right direction?

Answer 1

This is almost certainly due to Node.js' default server timeout of 120secs.

Try server.setTimeout(0) to remove this timeout.