[ https://issues.apache.org/jira/browse/CASSANDRA-8909?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14352809#comment-14352809 ]
Sam Tunnicliffe commented on CASSANDRA-8909: -------------------------------------------- +1 lgtm Just for future reference, it's preferred to give patch files a .txt extension as it makes it them easier to view in the browser. > Replication Strategy creation errors are lost in try/catch > ---------------------------------------------------------- > > Key: CASSANDRA-8909 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8909 > Project: Cassandra > Issue Type: Improvement > Reporter: Alan Boudreault > Assignee: Alan Boudreault > Priority: Trivial > Fix For: 2.0.13 > > Attachments: replication-strategy-exception-2.0.patch > > > I was initially executing a bad cassandra-stress command and was getting > this error: > {code} > Unable to create stress keyspace: Error constructing replication strategy > class > {code} > with the following command: > {code} > cassandra-stress -o insert --replication-strategy NetworkTopologyStrategy > --strategy-properties dc1:1,dc2:1 --replication-factor 1 > {code} > After digging in the code, I noticed that the error displayed was not the one > thrown by the replication strategy code and that the try/catch block could be > improved. Basically, the Constructor.newInstance can throw an > InvocationTargetException, which provide a better error report. > I think this improvement can also be done in 2.1 (not tested yet). If my > attached patch is acceptable, I will test and provide the right version for > 2.1 and trunk. > With the patch, I can see the proper error when executing my bad command: > {code} > Unable to create stress keyspace: replication_factor is an option for > SimpleStrategy, not NetworkTopologyStrategy > {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)