[ 
https://issues.apache.org/jira/browse/TRAFODION-2940?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16388999#comment-16388999
 ] 

ASF GitHub Bot commented on TRAFODION-2940:
-------------------------------------------

Github user kevinxu021 commented on a diff in the pull request:

    https://github.com/apache/trafodion/pull/1427#discussion_r172736968
  
    --- Diff: dcs/src/main/java/org/trafodion/dcs/zookeeper/ZkClient.java ---
    @@ -151,8 +163,28 @@ public void connect() throws IOException, 
InterruptedException {
                                this.zk=null;
                                throw new IOException("Cannot connect to 
Zookeeper");
                        }
    -                   
    -                   LOG.debug("Zookeeper.State=" + this.zk.getState());
    +
    +            // Solve the forcible reconnection
    +            // When zk reconn, the backup-master may take over the master,
    +            // so current master should restart, and queues in 
/dcs/master/leader
    +            if (LOG.isDebugEnabled()) {
    +                LOG.debug("force = [" + force + "]. checkPath = [" + 
checkPath + "]");
    +            }
    +            if (force && checkPath != null) {
    +                try {
    +                    Stat stat = zk.exists(checkPath, false);
    +                    if (LOG.isDebugEnabled()) {
    +                        LOG.debug("stat = [" + stat + "].");
    +                    }
    +                    if (stat == null) {
    +                        // this means master has change.
    +                        setSessionRecoverSuccessful(false);
    +                    }
    +                } catch (KeeperException e) {
    +                    e.printStackTrace();
    --- End diff --
    
    Log messages in file.


> In HA env, one node lose network, when recover, trafci can't use
> ----------------------------------------------------------------
>
>                 Key: TRAFODION-2940
>                 URL: https://issues.apache.org/jira/browse/TRAFODION-2940
>             Project: Apache Trafodion
>          Issue Type: Bug
>    Affects Versions: any
>            Reporter: mashengchen
>            Assignee: mashengchen
>            Priority: Major
>             Fix For: 2.3
>
>
> In HA env, if one node lose network for a long time , once network recover, 
> there will have two floating ip, two working dcs master, and trafci can't be 
> use.



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

Reply via email to