Nodes threadpool settings

We run elasticsearch 1.7 and the cluster has around 342 nodes.
I am trying to set the search threadpool size to 14 and the setting is applied correctly, meaning GET /_cluster/settings reflects the new values but the in /_nodes/stats the vaue of "" is defaulted to 13 form majority of the host. Only around a 100 hosts are getting the new value (14)

Any points on how can I debug this issue? seems like nodes are not pooling the latest config from master

What's pending tasks on some of the nodes that haven't applied it look like?

You should really upgrade. 2.2 is nearly EOL!


Hey Mark, I figured out the issue. It was related to the new plugin we integrated for elasticsearch metrics. And yes, Upgrading is the plan !

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