[jira] [Updated] (CASSANDRA-13484) Truncate the list of SSTableReaders logged when there is insufficient space on disk during compaction

["C. Scott Andreas (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-13484?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

C. Scott Andreas updated CASSANDRA-13484:
-----------------------------------------
    Component/s: Observability

> Truncate the list of SSTableReaders logged when there is insufficient space on disk during compaction
> -----------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-13484
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13484
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Observability
>            Reporter: Gabriel Garcia
>            Priority: Minor
>
> In the CompactionTask class, there is an expression `StringUtils.join(transaction.originals(), ", ")` (present since 1.0 in the form: `StringUtils.join(toCompact, ", ")`) (nowadays in method `reduceScopeForLimitedSpace`).
> If the set of SSTableReader is very large, then the generated log line, which contains full file paths for each reader, will be enormous. This is specially the case when dealing with SSTable files generated from big data sets using CQLSSTableWriter, since it will create many small files. (Over 40k in my case. The single log line will be well in the 10's of MB, potentially causing problems in some log viewers.) To make matters worse, reduceScopeForLimitedSpace() is called multiple times in checkAvailableDiskSpace().



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org