Description of problem: When attaching host USB device to VM, the VM failed to run. Issue occurred on AMD and Intel hosts and, RHEL7 and windows 2012 VMs and different USB devices (DOK and Keyboard). Same VMs run properly when detaching host USB device. Observing engine.log shows the next ERROR: 2015-09-08 16:41:35,933 ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (ForkJoinPool-1-worker-5) [] Correlation ID: null, Call Stack: null, Custom Event ID: -1, Message: VM rhel7_amd is down with error. Exit message: local variable 'device' referenced before assignment. 2015-09-08 16:41:35,934 INFO [org.ovirt.engine.core.vdsbroker.VmAnalyzer] (ForkJoinPool-1-worker-5) [] Running on vds during rerun failed vm: '4e47202c-f55a -47c2-ae8f-2d0567c266b9' Version-Release number of selected component (if applicable): rhevm-3.6.0-0.13.master.el6 sanlock-3.2.4-1.el7.x86_64 qemu-kvm-rhev-2.3.0-22.el7.x86_64 vdsm-4.17.5-1.el7ev.noarch libvirt-client-1.2.17-5.el7.x86_64 How reproducible: Consistently Steps to Reproduce: 1. Attach host USB device to VM. 2. Run VM. 3. Actual results: VM failed to run and return to "down" state. Next audit log displayed at webadmin UI: VM win2012_intel is down with error. Exit message: local variable 'device' referenced before assignment. Expected results: VM should run wuth USB device attached. Additional info: engine and vdsm logs attached. issues time: AMD host with RHEL7 VM with USB DOK attached: 2015-09-08 16:41:32 Intel host with windows 2012 VM and USB keyboard attached: 2015-09-08 16:48:52
Created attachment 1071376 [details] engine.log
Created attachment 1071377 [details] amd host vdsm.log
Created attachment 1071378 [details] intel vdsm.log
severity changed to urgent: since using GPU attached to VM without using USB mouse and keyboard is almost worthless. unless the user is using keyboard & mouse of an VM console.
The fix will be included in fix for 1248105 as that is similar issue.
(In reply to Martin Polednik from comment #5) > The fix will be included in fix for 1248105 as that is similar issue. this is not a similar issue, this is a _same_ issue *** This bug has been marked as a duplicate of bug 1248105 ***