[ https://issues.apache.org/jira/browse/CLOUDSTACK-9348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15382757#comment-15382757 ]
ASF subversion and git services commented on CLOUDSTACK-9348: ------------------------------------------------------------- Commit ea48e95bdd1641c752eb573fe448aac6478cecd1 in cloudstack's branch refs/heads/master from [~williamstev...@gmail.com] [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=ea48e95 ] Merge pull request #1601 from shapeblue/nio-aggressive-selector CLOUDSTACK-9348: Reduce Nio selector wait timeThis reduced the Nio loop selector wait time, this way the selector will check frequently (as much as 100ms per iteration) and handle any pending connection/tasks. This would make reconnections very quick at the expense of some CPU usage. /cc @swill @kiwiflyer guys can you please apply this fix in your env and test if you're still able to produce any Nio related error b/w mgmt server(s) and kvm agent(s) not being able to connect quickly. Please also watch out for any increased CPU usage (there should not be any significant change), in which case we may increase the timeout from 100ms to 200-400ms. * pr/1601: CLOUDSTACK-9348: Reduce Nio selector wait time Signed-off-by: Will Stevens <williamstev...@gmail.com> > CloudStack Server degrades when a lot of connections on port 8250 > ----------------------------------------------------------------- > > Key: CLOUDSTACK-9348 > URL: https://issues.apache.org/jira/browse/CLOUDSTACK-9348 > Project: CloudStack > Issue Type: Bug > Security Level: Public(Anyone can view this level - this is the > default.) > Reporter: Rohit Yadav > Assignee: Rohit Yadav > Fix For: 4.9.0 > > > An intermittent issue was found with a large CloudStack deployment, where > servers could not keep agents connected on port 8250. > All connections are handled by accept() in NioConnection: > https://github.com/apache/cloudstack/blob/master/utils/src/main/java/com/cloud/utils/nio/NioConnection.java#L125 > A new connection is handled by accept() which does blocking SSL handshake. A > good fix would be to make this non-blocking and handle expensive tasks in > separate threads/pool. This way the main IO loop won't be blocked and can > continue to serve other agents/clients. -- This message was sent by Atlassian JIRA (v6.3.4#6332)