Is process cluster event timeout value configurable and how if yes?

[f02061232]

I got ProcessClusterEventTimeoutException when create/delete index, default value is 30s. Is this configurable and how to config it?

org.elasticsearch.cluster.metadata.ProcessClusterEventTimeoutException: failed to process cluster event (delete-index [[xxx-indexname]]) within 30s
at org.elasticsearch.cluster.service.MasterService$Batcher.lambda$onTimeout$0(MasterService.java:124)
at java.util.ArrayList.forEach(ArrayList.java:1249)
at org.elasticsearch.cluster.service.MasterService$Batcher.lambda$onTimeout$1(MasterService.java:123)
at org.elasticsearch.common.util.concurrent.ThreadContext$ContextPreservingRunnable.run(ThreadContext.java:626)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

[DavidTurner]

Yes, you can pass a different timeout on a request-by-request basis: DELETE /my-index?timeout=60s&master_timeout=60s. However, if it's exceeding the default 30-second timeout then I suspect there's something wrong with your cluster, particularly since deleting an index is an URGENT-priority task which should be happening well within 30 seconds.