[ https://issues.apache.org/jira/browse/CASSANDRA-8112?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
T Jake Luciani updated CASSANDRA-8112: -------------------------------------- Description: nodetool compactionhistory keeps data for 1 week by default and creates a table from the result set in memory. For many systems a week can generate 10's of thousands of compactions in this time (esp with LCS) we should guard against this command allocating too much memory. was: nodetool compactionhistory keeps data for 1 week by default and creates a table from the result set in memory. For many systems a week can generate 10's of thousands of compactions (esp with LCS) we should guard against this command allocating too much memory. > nodetool compactionhistory can allocate memory unbounded > -------------------------------------------------------- > > Key: CASSANDRA-8112 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8112 > Project: Cassandra > Issue Type: Bug > Reporter: T Jake Luciani > Priority: Minor > Fix For: 2.1.1 > > > nodetool compactionhistory keeps data for 1 week by default and creates a > table from the result set in memory. > For many systems a week can generate 10's of thousands of compactions in this > time (esp with LCS) we should guard against this command allocating too much > memory. -- This message was sent by Atlassian JIRA (v6.3.4#6332)