gfmio opened a new pull request, #337: URL: https://github.com/apache/flink-statefun/pull/337
This PR adds a `maxRetries` option that limits the amount of times a request will be attempted before the request is dropped. When set to the default value of -1, batches will be retried indefinitely which matches the current behaviour. This provides a simple means to drop failed messages after a configurable number of attempts when such behaviour is desired, which is currently not very easy to implement. This also addresses an issue with Flink Statefun where a pipeline will get stuck if a remote function fails deterministically and the issue cannot be caught inside the handler. Normally, such issues can and should be caught inside the handler code, but this is not possible if the server deterministically crashes, e.g. with `log.Fatalw` in golang. Setting a maxRetries limit prevents such behaviour from blocking pipelines indefinitely. -- 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]
