Cluster going RED after mapper-size plugin installed

elasticsearch version 7.6.2

I have installed mapper-size plugin on 3 nodes out of 5 I have in the cluster.
Restarting each node after the installation.
I have noticed that one index is going into RED status all the time sine installing the plugin.
So I have deleted the plugin from node-1 and everything got back to normal.
Also needed to close/open the indices manually, as some of them stayed in YELLOW status

Any known issues?
Any explanations?
What can I do to avoid this situation? Or what to do when cluster health goes RED?


What do the logs on that node show?

Only error I see is

[2020-07-13T16:16:31,505][ERROR][i.n.u.c.D.rejectedExecution] [NODE1] Failed to submit a listener notification task. Event loop shut down?
java.util.concurrent.RejectedExecutionException: event executor terminated

Not sure if related.

It's more than just errors though, seeing what's happening more generally is benefical.

I would recommend installing the plugin on all nodes in the cluster.

1 Like

That was my goal.
However, I am doing it gradually, one node after the other.
But when issues has started, I rolled back.

Anything specific to look for?
It's a huge log file...

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.