Please provide more verbose description of those failures, logs and your configuration.
Thanks for your reply, We have formed cluster with the three nodes (dl360x4500, dl360x3826, dl360x2868) and we have down the machine (dl360x3826) from the cluster and then we have saw transaction failure for 4-5 seconds across the cluster nodes. Also, I have attached the log file for the same issue.
Requesting you to provide guidance to resolve this.
Those logs contain multiple server restarts with some errors as the server is restarted without killing another instance first - could you isolate the issue? Anyway, I can't find any exception thrown to the user code - what exactly is your issue?
I have attached the updated log files for the same scenario for your review. Kindly do let me know if you need any information on this.
Seems like a message was sent to the killed node, but the mapping from symbolic to physical addresses was lost and some reliable delivery protocol is trying to deliver that without success. No harm done, probably. If you want to find the root cause, we would need to see who sent that message - set on trace logging for org.jgroups, please.