Hello list,
Have a next question about `autorestart` option for `program` sections. I have program (some worker which is consuming messages from rabbitmq queue) with `autorestart: true` option. Imagine next situation - some problems with rabbitmq server and it went down. Worker got critical error and exited with some exit code. Supervisord will catch this event and will do restart attempt, it will do this attempts up to 5 times (note: I don't have `startretries` parameter in my configs). But due to fact that rabbitmq is down, worker will fail again and again immediately; and after 5 attempts supervisor will say `entered FATAL state, too many start retries too quickly*` *and that's all. I'd prefer not to make immediate attempts to restart program (worker) but rather wait some time before this attempt. Can I implenent such workflow with supervisord or should I put that responsibility on my worker (prevent worker from exiting on some bad things)?
_______________________________________________ Supervisor-users mailing list Supervisor-users@lists.supervisord.org https://lists.supervisord.org/mailman/listinfo/supervisor-users
