[jira] [Created] (CLOUDSTACK-9506) HA problem

["David Gil (JIRA)" <ji...@apache.org>]

David Gil created CLOUDSTACK-9506:
-------------------------------------

             Summary: HA problem
                 Key: CLOUDSTACK-9506
                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-9506
             Project: CloudStack
          Issue Type: Bug
      Security Level: Public (Anyone can view this level - this is the default.)
    Affects Versions: 4.2.0
         Environment: Centos
            Reporter: David Gil


Today our production cloudstack had a problem that a compute node went into alert and VM's and VR's quit working. When trying to move VM's to another node it came by no communication and it was unable to move the VM. We had to do a lot of manual items to get everything back up but need HA to work so that VM and VR move when a problem is found.

We had this problem a few months ago with node going into disconnected state and had to do the same thing.

What we had to do was login to the database and put item in a stopped state. Deleted the VR and restart network to get the VM's back up and running. Has anyone seen this before with HA not moving VM's quick enough and they get stuck in this state? What can be done so these VM's do not get into this state?

Thank you



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)