[ 
https://issues.apache.org/jira/browse/CASSANDRA-1546?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12923374#action_12923374
 ] 

Sylvain Lebresne commented on CASSANDRA-1546:
---------------------------------------------

I'm not sure I fully understand your scenario. But thinking about it more too, 
I retract the fact that we should force regenerating the node id during 
bootstrap.
If you decommission a node and boostrap it back, then as long you haven't done 
one of 1) deleting its sstable and not its system tables or 2) deleting its 
system tables (at least some of them) and not its sstables, then you are fine. 
You are fine because when bootstrapping the node back it'll reuse it's old node 
id saved in the system table.

I'm not sure yet what is the best way to prevent 1) or 2), but I'm pretty sure 
the answer doesn't lie in qualifying streaming operations. My opinion is that 
it's probably fine to as people to not do 1) or 2), at least for starter. Maybe 
could we try to detect those case at startup (even though I'm not sure how to 
detect 1), but that's a fairly weird situation anyway) and refuse to start. In 
any case, I'm hopeful we can improve that later. 

Now maybe I haven't understand your point or I'm missing something, but I don't 
see problem with the scenario you've described above.

> (Yet another) approach to counting
> ----------------------------------
>
>                 Key: CASSANDRA-1546
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1546
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: Sylvain Lebresne
>            Assignee: Sylvain Lebresne
>             Fix For: 0.7.1
>
>         Attachments: 0001-v2-Remove-IClock-from-internals.patch, 
> 0001-v3-Remove-IClock-from-internals.txt, 0001-v4-Counters.patch, 
> 0002-v2-Counters.patch, 0002-v3-Counters.txt, 0002-v4-thrift-changes.patch, 
> 0003-v2-Thrift-changes.patch, 0003-v3-Thrift-changes.txt, marker_idea.txt
>
>
> This could be described as a mix between CASSANDRA-1072 without clocks and 
> CASSANDRA-1421.
> More details in the comment below.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to