[ https://issues.jboss.org/browse/TEIID-2578?page=com.atlassian.jira.plugin... ] Steven Hawkins commented on TEIID-2578: --------------------------------------- The more I look at the remoting, the more trouble we may be getting ourselves in. Serializing the entire database can be a significant overhead. Also, just like was mentioned with the alter above, consistency can be an issue - performing a drop at the same time from two cluster members that are mutually exclusive. Unless I'm missing something, the DatabaseStore calls EventDistributor.dropDatabase, when in turn is called for each cluster member, which in turn each will call the adminEventDistributor methods to deleteDataSources. Shouldn't only the first succeed? Also what is the notion of server ownership on a databaseDrop? I've pushed changes to https://github.com/teiid/teiid/tree/TEIID-2578 so that we can coordinate on things from there. -- This message was sent by Atlassian JIRA (v6.4.11#64026)