[
https://issues.apache.org/jira/browse/CASSANDRA-21034?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18039418#comment-18039418
]
Dmitry Konstantinov commented on CASSANDRA-21034:
-------------------------------------------------
yes, it is actually by design, the idea was to flush fast with a cheap
compression and switch to a better compression in background during a
compaction (CASSANDRA-15379)
> ZstdDictionaryCompressor is not used upon flush
> -----------------------------------------------
>
> Key: CASSANDRA-21034
> URL: https://issues.apache.org/jira/browse/CASSANDRA-21034
> Project: Apache Cassandra
> Issue Type: Bug
> Components: Feature/Compression
> Reporter: Stefan Miklosovic
> Priority: Normal
>
> When I insert data to a table and do {{nodetool flush}}, an SSTable is
> created, but it will be compacted with LZ4 (seen via sstablemetadata).
> Correct compressor should be used upon flush as well so flushed data are
> compressed by a Zstd dictionary too.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
