How to inspect write queue

I'm on 6.8... my write queues are filling up and I don't really know, how to inspect their contents. I have even stopped Logstashes, but something is constantly writing. Does recovery use the write queues?

It's not consistent across all data nodes, I have some that are almost idle.

Any help would be appreciated!

In the end I used the tasks API and the hot_threads API to see what's going on.

Recoveries indeed contributed to the queues filling up, stopping it then later limiting the recovery speed help alleviating the problem.

End of thread.

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