Hide Forgot
Created attachment 548983 [details] vdsm.log from host in the affected data center Description of problem: Back end storage hosting the virtual machine experienced an unexpected outage causing a running VM to enter an unknown state. Version-Release number of selected component (if applicable): RHEV 3.0 Beta How reproducible: Unknown Steps to Reproduce: 1. Running VM 2. Associated storage domain unexpectedly went off-line 3. Hosted VM entered an "unknown" state and is unmanageable Actual results: Unable to start, stop, shutdown or delete VM. Unable to execute a destroy on the associated storage domain. Utilizing the REST API to force VM delete results in the following error: <?xml version="1.0" encoding="UTF-8" standalone="yes"?><fault><reason>Operation Failed</reason><detail>[Cannot remove VM. VM is running.]</detail></fault> Command executed: curl -X DELETE -H "Accept: application/xml" -H "Content-Type: application/xml" -u admin@internal:xxxxxx -k -d "<action><force>true</force></action>" https://rhevm3-beta:8443/api/vms/3461f1fe-d2a4-4f9f-beda-5e360967e8ba Error when trying to execute a destroy action ont eh storage domain: Cannot destroy Storage. Storage status illegal. Expected results: Expect through the management portal the ability to remove a defunct VM and/or it's associated storage domain. REST API should also allow forceful VM deletion regardless of VM state. Additional info: vdsm.log attached.
we can't destory a VM if we can't reachout to the domain to remove its disks but we do have auto-recovary that will make the domain and host responsive again (once outage ends) and this will let you manage the VM again.