Through my testing, I’ve noticed that I am regularly encountering a situation in my bot where I stop receiving new events on a stream while it still keeps itself connected (in ESTABLISHED state) without any traffic coming through. This usually happens not out of the gate, but after several reconnection attempts caused by planned connection disruption (see T242767: EventStreams drops the connection after 15 minutes, which makes it unreliable / T248736: ats-tls ran out of FDs on cp1089). I am wondering whether this is the same problem as the one in the other tasks or is this something different.
I am trying to code different workarounds around existing issues of EventStreams, since I don’t like to restart the bot virtually every day, but this issue is hindering my ability to detect that the things are wrong in the first place.
Off the top of my head, similar thing has already been described in T179986: Investigate why current es2r daemon is randomly hanging, but I would like for someone to confirm whether this is the same issue.