Hello, I have set up a cluster of 18 nodes where all of them can be master in case of a crash. I call it "all in one".
A few days ago I detected that I have "Indexing operations have failed"
I'm not using logstash or filebeats yet, just metricbeats. I use version "8.5.3"
I suspect that the amount of shard primaries and replicas are very few, currently it has 1 Primaries and 1 Replicas.
Any ideas would help me.