[ https://issues.apache.org/jira/browse/CASSANDRA-8981?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14365769#comment-14365769 ]
T Jake Luciani commented on CASSANDRA-8981: ------------------------------------------- Looks like it's just a timeout. The test is purposely trying to cause a race condition. Perhaps this should be a long-unit test > IndexSummaryManagerTest.testCompactionsRace failing on trunk > ------------------------------------------------------------ > > Key: CASSANDRA-8981 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8981 > Project: Cassandra > Issue Type: Bug > Components: Core > Reporter: Joshua McKenzie > Priority: Minor > Fix For: 3.0 > > > Keep running it repeatedly w/showoutput="yes" in build.xml on junit and > you'll see it time out with: > {noformat} > [junit] WARN 17:02:56 Unable to cancel in-progress compactions for > StandardRace. Perhaps there is an unusually large row in progress somewhere, > or the system is simply overloaded. > [junit] WARN 17:02:56 Unable to cancel in-progress compactions for > StandardRace. Perhaps there is an unusually large row in progress somewhere, > or the system is simply overloaded. > [junit] WARN 17:02:57 Unable to cancel in-progress compactions for > StandardRace. Perhaps there is an unusually large row in progress somewhere, > or the system is simply overloaded. > [junit] WARN 17:02:57 Unable to cancel in-progress compactions for > StandardRace. Perhaps there is an unusually large row in progress somewhere, > or the system is simply overloaded. > {noformat} > I originally thought this was a Windows specific problem (CASSANDRA-8962) but > can reproduce on linux by just repeatedly running the test. -- This message was sent by Atlassian JIRA (v6.3.4#6332)