[ https://issues.apache.org/jira/browse/CLOUDSTACK-7553?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14134778#comment-14134778 ]
ASF subversion and git services commented on CLOUDSTACK-7553: ------------------------------------------------------------- Commit d5a8f1d875667dcab6130b214029f2ec74603db0 in cloudstack's branch refs/heads/master from [~minchen07] [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=d5a8f1d ] CLOUDSTACK-7553: Clean up cached agentMap and pingMap in case of agents connecting back to a different MS. > Channel Closed error after SSVM and CPVM agents reconnect back to clustered > management server after host is disconnected. > ------------------------------------------------------------------------------------------------------------------------- > > Key: CLOUDSTACK-7553 > URL: https://issues.apache.org/jira/browse/CLOUDSTACK-7553 > Project: CloudStack > Issue Type: Bug > Security Level: Public(Anyone can view this level - this is the > default.) > Components: Management Server > Affects Versions: 4.2.0 > Reporter: Min Chen > Assignee: Min Chen > Priority: Critical > Fix For: 4.5.0 > > > 1. Set up a clustered management server setup using Vmware. Note down the MS > id that SSVM agent is connected to from DB host table. > 2. Disable uplink on hypervisor host that runs SSVM. Now in the log we should > see Channel is closed error, which is correct and as expected. > 3. Waited for some time, enable uplink back on hypervisor, SSVM agent will > reconnect back to MS. Make sure that SSVM agent is connected to a different > MS node from the one we noted down in step #1. If not, try #2 and #3 again > until that happens. > 4. Now we will continue seeing "Channel is closed" error from the MS log > noted down in step #1, this is not correct. Command sent to SSVM will fail if > the command is sent from MS node noted down in step #1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)