[
https://issues.apache.org/jira/browse/CASSANDRA-5015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13608761#comment-13608761
]
Sylvain Lebresne commented on CASSANDRA-5015:
---------------------------------------------
I believe Jonathan's referring to how in SSTableReader.load we recompute the
index summary if we can't read it from disk. In fact, SSTableReader.load
already takes a boolean to decide if it should rebuild the bloom filter, so it
seems we only need to pass true if we detect the bloom_filter_fp_chance has
changed. Though that later detection may require that we save the bffc in the
SSTable metadata.
> move bloom_filter_fp_chance to compaction options
> -------------------------------------------------
>
> Key: CASSANDRA-5015
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5015
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.0.7
> Reporter: Jonathan Ellis
> Assignee: Aleksey Yeschenko
> Priority: Minor
> Fix For: 2.0
>
>
> This setting doesn't take affect until data is recompacted, so should be
> moved into compaction options.
> Alternatively, we could do what index_interval does, and rebuild it on
> startup if changed.
--
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
