Based on different opinions on the Internet, the following type of logged messages…
DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2019349100 for /xxx.xxx.xxx.xx1 DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2019428200 for / xxx.xxx.xxx.xx2 DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2019384900 for / xxx.xxx.xxx.xx3 DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2592364600 for / xxx.xxx.xxx.xx3 DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2019322500 for / xxx.xxx.xxx.xx4 DEBUG [GossipStage:1] 2018-07-25 07:03:27,336 FailureDetector.java:456 - Ignoring interval time of 2019394600 for / xxx.xxx.xxx.xx5 … means that the network is flakey or the cluster is under heavy load. What if the network is not flakey and the cluster is not under heavy load? What could cause this? https://github.com/apache/cassandra/blob/8b3a60b9a7dbefeecc06bace617279612ec7092d/src/java/org/apache/cassandra/gms/FailureDetector.java#L450-L474 MAX_INTERVAL_IN_NANO seems to be how this is controlled. Where does the result of getMaxInterval() which sets MAX_INTERVAL_IN_NANO come from? Thanks, David Payne | ̄ ̄| _☆☆☆_ ( ´_⊃`) c. 303-717-0548 dav...@cqg.com<mailto:dav...@cqg.com>