[ https://issues.apache.org/jira/browse/CASSANDRA-11492?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15225939#comment-15225939 ]
Sylvain Lebresne commented on CASSANDRA-11492: ---------------------------------------------- I suppose it's a multi-node cluster? Schema changes in C* are currently unfortunately inherently racy. It is thus up to the client to wait for schema agreement between schema changes (if they conflict at least, which they do in this case). In theory, this is handled under the hood by your client driver but that obviously a bit dependent of said driver. If you give us more informations on which driver you use for this (or, if it's cqlsh, which version of cqlsh this is), we might be able to say for sure if that's the problem or if it's something else. > Crash during CREATE KEYSPACE immediately after drop > --------------------------------------------------- > > Key: CASSANDRA-11492 > URL: https://issues.apache.org/jira/browse/CASSANDRA-11492 > Project: Cassandra > Issue Type: Bug > Reporter: Seth Rosenblum > > In a test environment, I run in immediate sequence: > {code:sql} > DROP KEYSPACE IF EXISTS "keyspacename"; > CREATE KEYSPACE "keyspacename" > WITH REPLICATION = {'class': 'SimpleStrategy', 'replication_factor': > 1}; > {code} > And I end up with a {{java.lang.AssertionError: null}} error triggered during > the CREATE statement. > https://gist.github.com/sethrosenblum/d6b450b4455eeb28550f4038a0928bcf -- This message was sent by Atlassian JIRA (v6.3.4#6332)