[ https://issues.apache.org/jira/browse/CASSANDRA-8624?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14336402#comment-14336402 ]
Prajakta Bhosale commented on CASSANDRA-8624: --------------------------------------------- I tried to reproduce the same bug using apache cassandra-2.0.8 which dse-4.5.1 uses but couldn't reproduce. Tried the stress test with @6lakh records, still things are working as expected. The node is shown as down if there is network issue. I also tried with apache cassandra-2.1.2 and the same happened with it too. > Cassandra Cluster's Status Inconsistency Strangely > -------------------------------------------------- > > Key: CASSANDRA-8624 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8624 > Project: Cassandra > Issue Type: Bug > Components: Tools > Environment: Cassandra 1.2.11 > Reporter: ZhongYu > Priority: Minor > Attachments: QQ截图20150115125254.png > > > We found a strange phenomenon about Cassandra Cluster's status that all the > nodes in the cluster found other node's status inconsistency. Especially, the > inconsistency has an interesting patten. See the following example: > There are 5 nodes (pc17, pc19, pc21, pc23, pc25) in the cluster. Their seeds > configuration are all "pc17, pc19, pc21, pc23, pc25". In a moment, > pc17 found others UP; > pc19 found pc17 DN, others UP; > pc21 found pc17, pc19 DN, others UP; > pc23 found pc17, pc19, pc21 DN, others UP; > pc25 found pc17, pc19, pc21, pc23 DN, only self UP; > See attachments as screen's snapshot. -- This message was sent by Atlassian JIRA (v6.3.4#6332)