Hide Forgot
Description of problem: HA VMs are not restarted on different host(s) if NonResponsive host is off (detected by successful power management status action) and start power management action failed Version-Release number of selected component (if applicable): 3.0 and later How reproducible: 100% Steps to Reproduce: 1. Create 2 or more hosts cluster with power management properly configured 2. Run HA VMs on host1 and make this host non responsive by turning the power off 3. Make sure that power management status of host1 can be properly detected, but power management start action failed 4. Check that fence of host1 failed and HA VMs are not restarted on different hosts although we detect that host1 is turned off using power management Actual results: HA VMs are not restarted on different host even though host1 is powered off Expected results: HA VMs are restarted on different hosts Additional info:
Verify with: rhevm-4.0.4.2-0.1.el7ev.noarch Hosts with PM: OS Version:RHEL - 7.2 - 9.el7 OS Description:Red Hat Enterprise Linux Server 7.2 (Maipo) Kernel Version:3.10.0 - 327.28.3.el7.x86_64 KVM Version:2.3.0 - 31.el7_2.21 LIBVIRT Version:libvirt-1.2.17-13.el7_2.5 VDSM Version:vdsm-4.18.13-1.el7ev Steps: 1. Create 2 hosts in cluster with power management properly configured 2. Run HA 2 VMs and 1 none HA VM on host1 3. Turn power off host_1 from power management and setup host not to power up(from remote PM iLO4 in this case) 4. Check HA VMs are started on host_2 Results: HA VMs run on host_2 - PASS
My customer is experiencing the same issue. He is on the latest RHV 4.0 release. His environment is secure, so logs will not be possible. His environment was just upgraded from RHEV 3.6 to RHV 4.0 in hopes that the issue would be resolved, however it still exists. His hardware is running iDrac 7 as the only difference from the reported bug.
(In reply to Donny Davis from comment #2) > My customer is experiencing the same issue. He is on the latest RHV 4.0 > release. His environment is secure, so logs will not be possible. > > His environment was just upgraded from RHEV 3.6 to RHV 4.0 in hopes that the > issue would be resolved, however it still exists. > > His hardware is running iDrac 7 as the only difference from the reported bug. As you can see this bug is targeted to 4.0.4, which is not yet officially released. Also fix for this bug was backported into 3.6.9 in BZ1372812 which was released a few days ago ...