Proposal to bump persistent_timeout to 65 seconds

[shayonj]

Description

I noticed that if Puma's persistent_timeout is shorter than the load balancer's timeout, occasional minor service disruptions can occur. For instance, consider the AWS Application Load Balancer (ALB) with a default idle timeout of 60 seconds. If ALB forwards a request to Puma just as Puma is closing the socket, it may result in a broken TCP handshake, leading to a TCP RST/FIN being sent back to ALB. ALB would then handle that RST/FIN and return a 502 error to the client, which is not ideal.

Since most load balancers, like AWS ALB/ELB and Heroku, have default idle timeouts around 60 seconds and 55 seconds respectively, I am proposing that we increase Puma's persistent_timeout to 65 seconds so that the upstream negotiates and closes the connection before Puma does.

Your checklist for this pull request

• I have reviewed the guidelines for contributing to this repository.
• I have added (or updated) appropriate tests if this PR fixes a bug or adds a feature.
• My pull request is 100 lines added/removed or less so that it can be easily reviewed.
• If this PR doesn't need tests (docs change), I added [ci skip] to the title of the PR.
• If this closes any issues, I have added "Closes #issue" to the PR description or my commit messages.
• I have updated the documentation accordingly.
• All new and existing tests passed, including Rubocop.