Red Hat Bugzilla – Bug 1479796
vmfex hook not working properly
Last modified: 2017-08-16 03:45:21 EDT
Created attachment 1311181 [details] logs Description of problem: vmfex hook not working properly. Seems like vmfex hook is not working as it should. The interface type must be: <interface type='direct'> <mac address='00:00:00:00:00:2a'/> <source network='direct-pool' dev='enp6s0f1' mode='passthrough'/> <virtualport type='802.1Qbh'> <parameters profileid='VMFEX_portprofile_1'/> </virtualport> <target dev='macvtap0'/> But instead, it is started as bridge type: <interface type='bridge'> <mac address='00:00:00:00:00:26'/> <source bridge='vmfex'/> <target dev='vnet0'/> <model type='virtio'/> <filterref filter='vdsm-no-mac-spoofing'/> <boot order='2'/> <alias name='net0'/> I have tested this with the suggested fix(reverted patch) for BZ - 1478054, but it didn't fixed the issue. We believe that it might be another issue with the hook or something else. I don't see any errors in the logs. Version-Release number of selected component (if applicable): 4.2.0-0.0.master.20170808202742.git7c42951.el7.centos How reproducible: 100 on latest master Steps to Reproduce: 1. Configure vmfex via engine-config 2. Configure portfrofile via the UCS manager 3. Create network with vmfex vNIC profile and attach to UCS host 4. Add vmfex vNIC tot he VM and run it Actual results: <interface type='bridge'> <mac address='00:00:00:00:00:26'/> <source bridge='vmfex'/> <target dev='vnet0'/> <model type='virtio'/> <filterref filter='vdsm-no-mac-spoofing'/> <boot order='2'/> <alias name='net0'/> Expected results: <interface type='direct'> <mac address='00:00:00:00:00:2a'/> <source network='direct-pool' dev='enp6s0f1' mode='passthrough'/> <virtualport type='802.1Qbh'> <parameters profileid='VMFEX_portprofile_1'/> </virtualport> <target dev='macvtap0'/> Additional info: We thought this was caused by BZ - 1478054, but the suggested patch(the reverted one for the hooks) didn't fixed the vmfex hook issue.
I believe that it's a dup of 1478054 and like it "solved" by https://gerrit.ovirt.org/#/c/80444/ . I hope a nicer solution follows.
(In reply to Dan Kenigsberg from comment #1) > I believe that it's a dup of 1478054 and like it "solved" by > https://gerrit.ovirt.org/#/c/80444/ . I hope a nicer solution follows. Wrong. As we already said, the fix for BZ 1478054(see comment8#)
(In reply to Michael Burman from comment #2) > (In reply to Dan Kenigsberg from comment #1) > > I believe that it's a dup of 1478054 and like it "solved" by > > https://gerrit.ovirt.org/#/c/80444/ . I hope a nicer solution follows. > > Wrong. As we already said, the fix for BZ 1478054(see comment8#) The fix for BZ 1478054 didn't helped.
This bug report has Keywords: Regression or TestBlocker. Since no regressions or test blockers are allowed between releases, it is also being identified as a blocker for this release. Please resolve ASAP.
This looks like a consequence of the move to Engine XML. This is the hook description: https://github.com/oVirt/vdsm/tree/master/vdsm_hooks/vmfex_dev
Seems similar to other custom properties that are not handled properly with Engine XML.