AWS ECS & ALB & CloudFront. sometimes happed status code 502

[eretica]

I'm using Nextjs in a configuration called ECS ALB CloudFront using AWS

My production environment may return 502 several times a day when returning NextAPIRoute or a file

I checked the situation in Athena and it looks like a screenshot.

I'm guessing that keepalive is the cause, but is there a solution?

https://shuheikagawa.com/blog/2019/04/25/keep-alive-timeout/
Like this article.

server.keepAliveTimeout = 61 * 1000;
server.headersTimeout = 65 * 1000; // This should be bigger than `keepAliveTimeout + your server's expected response time`

We think we can improve this by setting up

Or, if you have any other suggestions, please let me know.

[jordie23]

Heya the only solution I was able to find was to modify the IdleTimeout on the AWS ELB to be 4 seconds, i.e 1 second less than node's default 5 seconds.

[tabuchid]

I tried making a feature request but it got converted to a discussion #28642
Maybe we'll get some answers there.

[schorfES]

This PR (#35827) introduces a change to the CLI that allows to set keepAliveTimeout:

next start --keepAliveTimeout 65000