ManageIQ
added support for keep alive timeouts
See #512: Fix in response to watch pods timing out and not recognizing any new notices after about 30 minutes of no new watch events.
Hi, @cben, can you please review this PR and helping on merging this PR as @khchau7 blocked on this for her project?
Is this really controlling what #512 describes? https://github.com/httprb/http/blob/v4.4.1/lib/http/connection.rb#L184-L188 https://github.com/httprb/http/blob/v4.4.1/lib/http/client.rb#L113-L116 seems it affects Ruby-level logic closing connection on inactivity, not TCP-level keepalive?
Generally, I'm not against exposing more options :+1:. We can merge this first and let people experiment with optimal settings.
Note however that sooner or later, we'd like to replace http with faraday: #488 — does this setting have a parallel in faraday?
If raising the timeout helps but the problem is specific to http gem, then perhaps we should just set it higher but not expose as new knob.
If you believe this is correct approach, please add a mention of the new option in README and add a CHANGELOG entry (in "Unreleased" section, which is currently horribly incomplete).
