Hi, I have created an elasticsearch river module and I have a question
about its lifecycle.
I met a situation where cluster nodes were shut down before sending
shutdown signal to es-plugin.
Is there any possibility to make my plugin more aware about cluster
state and stop it before nodes will shut down?
If the nodes shut down earlier that my plugin it will throw exceptions
(nodes shutdown) during indexing. In that case nothing bad happens
because data will be indexed next time, but system logs doesn't look
Thanks for any helpful advice