Avoid getting stuck on infinite loop

[hasnat]

If we get 400 which could be part of DLQ, plugin here gets stuck in loop indefinitely on retry
e.g. https://discuss.elastic.co/t/logstash-5-6-1-inifinite-loop-encountered-a-retryable-error/101753

Thanks for contributing to Logstash! If you haven't already signed our CLA, here's a handy link: https://www.elastic.co/contributor-agreement/

[jordansissel]

This will cause data loss when queue.type: memory. For these scenarios, you can circumvent the safety here by interrupting logstash two times: the first time it will try to safely shutdown, and a second time should cause it to abandon safety and exit.

The reported error "[2017-09-25T19:54:44,850][ERROR][logstash.outputs.elasticsearch] Encountered a retryable error. Will Retry with exponential backoff {:code=>400, :url=>"http://es-data-int.ex.com:9200/_bulk", :body=>"{"error":{"root_cause":[{"type":"parse_exception","reason":"request body is required"}],"type":"parse_exception","reason":"request body is required"},"status":400}"}" in the discuss forum, "request body is required" is a bug, not a feature, and we should fix that bug.

[jordansissel]

For reasons stated above, Logstash has a default mechanism of safe shutdown. If you wish to abandon this safety, that is a choice you can make, but we will not make this unsafe (abandon the retry) the default behavior.

[hasnat]

thanks for the explaining @jordansissel
I never knew those 400s were due to a different bug, my goal is to pass down all the 400s to dead letter queue. Would there be a way to do this via config instead of sending interrupt signals to logstash?