You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jaakko Laine (JIRA)" <ji...@apache.org> on 2010/01/15 13:47:54 UTC
[jira] Created: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Cluster-wide mutex (ZooKeeper integration)
------------------------------------------
Key: CASSANDRA-704
URL: https://issues.apache.org/jira/browse/CASSANDRA-704
Project: Cassandra
Issue Type: New Feature
Reporter: Jaakko Laine
Assignee: Jaakko Laine
Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jaakko Laine updated CASSANDRA-704:
-----------------------------------
Attachment: (was: 704-cluster-mutex.patch)
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12834881#action_12834881 ]
Hudson commented on CASSANDRA-704:
----------------------------------
Integrated in Cassandra #357 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/357/])
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12803683#action_12803683 ]
Hudson commented on CASSANDRA-704:
----------------------------------
Integrated in Cassandra #331 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/331/])
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jaakko Laine resolved CASSANDRA-704.
------------------------------------
Resolution: Fixed
Committed to trunk/contrib.
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jaakko Laine updated CASSANDRA-704:
-----------------------------------
Attachment: 704-cluster-mutex.patch
new version. sometimes zookeeper.create fails even though return code indicates a success. this can be handled on server side, so no need to report an error to client in this case.
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12802699#action_12802699 ]
Jonathan Ellis commented on CASSANDRA-704:
------------------------------------------
I would rather see this mutex in contrib, and I don't think CASSANDRA-721 belongs in our repo at all (CASSANDRA-580 will allow incr in a way that fits Cassandra's design better, as I think you agree).
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12802703#action_12802703 ]
Jaakko Laine commented on CASSANDRA-704:
----------------------------------------
I'll move this to contrib.
Yeah, you're of course right about 721. I merely posted that patch so that if somebody wants to use counters before 580 is finished, a patch is available for that. Obviously there is no need for ZK for this purpose as soon as 580 is finished and the patch does not need to go anywhere near Cassandra repo, it's just a temporary thing anyway (should have stated that clearly in the patch)
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12802588#action_12802588 ]
Jaakko Laine commented on CASSANDRA-704:
----------------------------------------
Nobody is using it at the moment :) Just added #721 atomic increment/decrement operations, which uses this. It is not suitable to expose this to clients, as it would make it very difficult to avoid deadlocks. If clients need ZK for external locking, they can use it directly. This is just for internal operations.
This can be put to contrib as well. Just placed it in service because thougth of it as "locking service for the whole server".
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jaakko Laine (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jaakko Laine updated CASSANDRA-704:
-----------------------------------
Attachment: zookeeper-3.2.2.jar
704-cluster-mutex.patch
patch attached for mutex. zookeeper jar is needed as well.
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-704) Cluster-wide mutex (ZooKeeper
integration)
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12801847#action_12801847 ]
Jonathan Ellis commented on CASSANDRA-704:
------------------------------------------
Who is calling CM.lock()? If it's intended to be exposed to thrift I don't see that here, but I don't see any other uses either.
Is there a compelling reason to put this in core instead of contrib/?
> Cluster-wide mutex (ZooKeeper integration)
> ------------------------------------------
>
> Key: CASSANDRA-704
> URL: https://issues.apache.org/jira/browse/CASSANDRA-704
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jaakko Laine
> Assignee: Jaakko Laine
> Attachments: 704-cluster-mutex.patch, zookeeper-3.2.2.jar
>
>
> Provide a cluster-wide mutex (zookeeper integration). Not strictly needed at the moment, but since it's done now, I'll post it here for community to review & decide if it should be included. This would allow us to have atomic increment/decrement operations before vector clocks, and might come handy in automatic load balancing and other future tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.