brooklyn-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bostko <...@git.apache.org>
Subject [GitHub] incubator-brooklyn pull request: Riak - Create Bucket type effecto...
Date Tue, 07 Apr 2015 22:51:09 GMT
Github user bostko commented on a diff in the pull request:

    https://github.com/apache/incubator-brooklyn/pull/581#discussion_r27929036
  
    --- Diff: software/nosql/src/main/java/brooklyn/entity/nosql/riak/RiakNodeSshDriver.java
---
    @@ -395,12 +390,12 @@ public void joinCluster(String nodeName) {
                 if (!hasJoinedCluster()) {
                     ScriptHelper joinClusterScript = newScript("joinCluster")
                             .body.append(sudo(format("%s cluster join %s", getRiakAdminCmd(),
nodeName)))
    +                        .body.append(sudo(format("%s cluster plan", getRiakAdminCmd())))
    +                        .body.append(sudo(format("%s cluster commit", getRiakAdminCmd())))
    --- End diff --
    
    Actually I found why it is implemented that way.
    I think the initial idea was to support cluster queue operations. It is correct not to
commit just after the cluster join command. Other than that it was executed for one of the
members.
    Instead of that commit was executed after the joinings are done and then in the RiakCluster.start
at the end it was executed clusterCommit()
    The bug here is that the commit is not executed properly because every time joining was
executed against random member of the ring instead of using one, so commiting should be done
against the nodes or the node which triggered cluster join. This is the actual bug. And the
other bug is may be that leaveCluster had plan and commit command which is unnecessary.
    
    My point is that it could support queue cluster operations but the task requires bigger
effort.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message