You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Greg DeAngelis (JIRA)" <ji...@apache.org> on 2013/08/16 04:30:48 UTC

[jira] [Commented] (CASSANDRA-4952) Add blocking force compaction (and anything else) calls to NodeProbe

    [ https://issues.apache.org/jira/browse/CASSANDRA-4952?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13741820#comment-13741820 ] 

Greg DeAngelis commented on CASSANDRA-4952:
-------------------------------------------

Hi I'm new to the project and would like to start contributing. Before I try to work on this I just wanted to validate what I think needs to be done.

As I understand it, what is being asked for is a blocking version of forceUserDefinedCompaction. This looks like it involves following what is done in forceUserDefinedCompaction for creating the list of keyspace/column family pairs, using the work done to construct the runnable in submitUserDefined (minus the executor.submit) to create a list of runnables then calling executor.invokeAll with some default timeout. Does this sound about right?
                
> Add blocking force compaction (and anything else) calls to NodeProbe
> --------------------------------------------------------------------
>
>                 Key: CASSANDRA-4952
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-4952
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 1.1.6
>            Reporter: Michael Harris
>            Priority: Minor
>              Labels: lhf
>
> There are times when I'd like to get feedback about when compactions complete.  For example, if I'm deleting data from cassandra and want to know when it is 100% removed from cassandra (tombstones collected and all).  This is completely trivial to implement based on the existing code (the method called by the non-blocking version returns a future, so you could just wait on that, potentially with a timeout).

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira