I get an error in TimeGen1 deployment: The model execution took longer than the timeout supplied in request_timeout_ms under request_settings of your model deployment config.

Question

The error started appearing one week ago even though I did not change anything to my deployment

Answer 1

Hi @maxime ,

Check the Nixtla client for request timeout parameter adjustment.

Here, you can set the timeout to None to disable or you required timeout.

Answer 2

Hi maxime!

Below Error suggests on model is taking longer times then "request_timeout_ms" declared in your model deployment.

The model execution took longer than the timeout supplied in request_timeout_ms under request_settings of your model deployment config.

You can do the following steps to address above issue.

1. Adjust timeouts param from your endpoint (score_timeout, timeouts in custom liveness and readyness prob)
2. You can adjust autoscalling option in your endpoints to scale automatically on metrics like CPU_utlization, latency, request per minut etc if it is taking more time /load on existing endpoint
3. Use Higher SKU for faster processing
4. Optimized your code to predict smaller batches of data to lower inference time.Please don’t forget to Accept Answer and Yes for "was this answer helpful" wherever the information provided helps you, this can be beneficial to other community members.Thank You.

Thank you.