[ 
https://issues.apache.org/jira/browse/CASSANDRA-5051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13649471#comment-13649471
 ] 

Jonathan Ellis commented on CASSANDRA-5051:
-------------------------------------------

I wanted to see if I could decouple things a bit and 
https://github.com/jbellis/cassandra/commits/5051-v7 is the result (also 
rebased to current trunk).  I added IRangeProvider to allow tests to inject 
what they want to test directly.  I 

I also switched off of using bootstrapMode in favor of Mode, but either way I 
don't think we're quite in the clear -- when we tell node A to move or relocate 
data to B, it relies on gossip to tell node B about that (the Pending Ranges).  
But if gossip is extra slow for whatever reason -- even just a large cluster 
where everything else is working as planned -- it might take longer than the 
hardcoded ring delay for B to learn about his new responsibility, and 
compaction would delete the data.

I think we might need to add a "tell the recipient about the new ranges he's 
going to have" step to streaming. /cc [~yukim] [~brandon.williams]
                
> Allow automatic cleanup after gc_grace
> --------------------------------------
>
>                 Key: CASSANDRA-5051
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5051
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: Brandon Williams
>            Assignee: Vijay
>              Labels: vnodes
>             Fix For: 2.0
>
>         Attachments: 0001-5051-v4.patch, 0001-5051-v6.patch, 
> 0001-5051-with-test-fixes.patch, 0001-CASSANDRA-5051.patch, 
> 0002-5051-remove-upgradesstable.patch, 
> 0002-5051-remove-upgradesstable-v4.patch, 0004-5051-additional-test-v4.patch, 
> 5051-v2.txt
>
>
> When using vnodes, after adding a new node you have to run cleanup on all the 
> machines, because you don't know which are affected and chances are it was 
> most if not all of them.  As an alternative to this intensive process, we 
> could allow cleanup during compaction if the data is older than gc_grace (or 
> perhaps some other time period since people tend to use gc_grace hacks to get 
> rid of tombstones.)

--
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

Reply via email to