rawlinp commented on PR #7007: URL: https://github.com/apache/trafficcontrol/pull/7007#issuecomment-1206720102
So, exponential backoff retrying is certainly better than non-exponential, but it's really important for there to be jitter in the sleep intervals as well. Otherwise, clients will all retry at the same time still. For example, if a particular sleep window is 1 minute long, the client should sleep some random about between 0 and 1 minute before retrying. This prevents all retries from happening right at that 1 minute mark. Also, since we're running t3c every 1 - 2 minutes already without retries, I think it would make sense for the longest sleep window to be something like 4 or 8 minutes. We should also provide the ability to cap the window to some maximum like 10 minutes. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
