Hello again! Bowen, your answer was right. After forcing clock synchronization, the count of MUTATION messages radically dropped down. After sync I made full repair on affected nodes and now I don't have any MUTATION drops in logs. Thanks for your help!
вт, 12 апр. 2022 г. в 12:31, Bowen Song <bo...@bso.ng>: > > First thing to check, do you have NTP client running on all Cassandra > servers? Are their clock in sync? If you answer "yes" to both, check the > server load, does any server have high CPU usage or disk utilization? > Any swapping activity? If not, check the GC logs, and looking for long > GC pauses. > > > On 12/04/2022 10:15, Azamat Hackimov wrote: > > Hello. > > > > I've got two DCs with 8 hardware servers on each. After migration to > > Cassandra 4 I have strange messages in system.log: > > > > INFO [ScheduledTasks:1] 2022-04-12 12:05:58,595 > > MessagingMetrics.java:206 - MUTATION_RSP messages were dropped in last > > 5000 ms: 0 internal and 241 cross node. Mean internal dropped latency: > > 0 ms and Mean cross-node dropped latency: 2395 ms > > INFO [ScheduledTasks:1] 2022-04-12 12:05:58,596 > > MessagingMetrics.java:206 - MUTATION_REQ messages were dropped in last > > 5000 ms: 0 internal and 8 cross node. Mean internal dropped latency: 0 > > ms and Mean cross-node dropped latency: 2395 ms > > INFO [ScheduledTasks:1] 2022-04-12 12:05:58,596 > > MessagingMetrics.java:206 - PAXOS_COMMIT_RSP messages were dropped in > > last 5000 ms: 0 internal and 102 cross node. Mean internal dropped > > latency: 0 ms and Mean cross-node dropped latency: 2395 ms > > > > Can someone interpret these messages? I suspect that messages indicate > > that the cluster has some replication issues. -- >From Siberia with Love!
