You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@activemq.apache.org by "Network (JIRA)" <ji...@apache.org> on 2018/10/01 23:01:00 UTC

[jira] [Created] (AMQ-7064) ActiveMQ cluster blocks indefinitely in the presence of partial network partition

Network created AMQ-7064:
----------------------------

             Summary: ActiveMQ cluster blocks indefinitely in the presence of partial network partition
                 Key: AMQ-7064
                 URL: https://issues.apache.org/jira/browse/AMQ-7064
             Project: ActiveMQ
          Issue Type: Bug
          Components: activemq-leveldb-store
    Affects Versions: 5.15.3
            Reporter: Network


We set up a cluster of 3 ActiveMQ brokers (Replicated LevelDB Store). other 3 machines were used to run Zookeeper cluster. We configured ActiveMQ as the following:
 * Replicas: 3
 * sync:  quorum_disk

when a partial network partition isolates the master broker from the other brokers while ZooKeeper can communicate with both sides of the network partition, the cluster will block indefinitely and clients messages will not processed. The following are the steps to reproduce the behavior:  
 # partial network partition such that the master broker will not be able to communicate with other brokers but can communicate with Zookeeper nodes (ZooKeeper nodes are reachable for all brokers)
 # a client sends a message. However, it does not receive a response. it seems that the cluster stopped processing client request. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)