Increase of Functions timeout
Timeout is set to 90 seconds. This value is a fixed setting and is too low for extensive ingestion functions.
A default value of 600 seconds would be helpful.
Even better would be the ability to set an individual timeout value per application in the settings.
Bastian Vennemann
shared this idea
-
Mansoor
commented
The timeout is currently set to 240 seconds.
Please see the article below to stay up to date with the latest on Function Constraints.
https://www.mongodb.com/docs/atlas/app-services/functions/#constraints -
Simon
commented
I notice the timeout recently went up from 90 to 120 seconds. I have some long-running intensive statistical functions that I have had to break up into manageable parts, saving state along the way. If the function times out I have a timer that calls it again (every 2 minutes) and picks up from where it left off.
-
Anton
commented
We definitely need the ability to extend the execution timeout or something else to run long-living tasks.
-
Karolis Scerbiakas
commented
90 secs as a default is good, but ability to adjust it as required would be super helpful. It would save a lot of time working around this limitation
-
Sachinta
commented
overriding of timeout for each application/trigger will be so helpful
-
Eve Ragins
commented
I personally appreciate the lower default so as to not consume resources unnecessarily. However, I fully support being able to override the default for executions which are known to potentially take a long time.
-
Peter
commented
This is especially helpful for running long-running batch jobs based on various Triggers.
As an example, we run weekly batch jobs to generate reports, which can sometimes take longer than 90 seconds to execute and/or to orchestrate the entire batch. We've mostly worked around the issue by breaking the functionality into multiple functions and relying on asynchronous processing which are able to kick off many individual processes, but there is a limit to how much we can do this before having to move to a service like AWS Batch.