Bug 1140633

Summary: Can't run VMs on IBM_PowerKVM host
Product: Red Hat Enterprise Virtualization Manager Reporter: Lukas Svaty <lsvaty>
Component: vdsmAssignee: Yaniv Bronhaim <ybronhei>
Status: CLOSED DUPLICATE QA Contact: Lukas Svaty <lsvaty>
Severity: urgent Docs Contact:
Priority: medium    
Version: 3.4.1-1CC: bazulay, danken, ecohen, gklein, iheim, lpeer, lsvaty, ofrenkel, oourfali, rbalakri, Rhev-m-bugs, ybronhei, yeylon
Target Milestone: ---   
Target Release: 3.4.3   
Hardware: ppc64   
OS: Linux   
Whiteboard: infra
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-09-14 15:51:33 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Infra RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
engine.log
none
vdsm from host none

Description Lukas Svaty 2014-09-11 11:40:38 UTC 
Description of problem:
Host is successfully added to engine.
However OS Version: is displayed as unknown and all the versions (libvirt, vdsm..) as [N/A]. CPU Name recognized correctly as PowerPC 8.

When trying to start VM host is filtered out by scheduling policy (see attached logs).

Version-Release number of selected component (if applicable):
av12_ppc
engine:
ovirt-host-deploy-1.2.3-1.el6.noarch
otopi-1.2.3-1.el6.noarch
host:
vdsm-xmlrpc-4.14.15-5.gitb02671b.fc20.noarch
vdsm-4.14.15-5.gitb02671b.fc20.ppc64
vdsm-cli-4.14.15-5.gitb02671b.fc20.noarch
vdsm-python-zombiereaper-4.14.15-5.gitb02671b.fc20.noarch
vdsm-python-4.14.15-5.gitb02671b.fc20.ppc64
kernel-3.10.23-1700.pkvm2_1.1.ppc64


How reproducible:
100%

Steps to Reproduce:
1. Add ppc host to cluster
2. Start VM on local storage of cluster

Actual results:
Host filtered out by scheduling policy

Expected results:


Additional info:
consiguration of engine
psql:
added 2:PowerPC 8:powernv,model_power8:POWER8_v2_1.1:ppc64

config:
rhevm-config -s ClusterEmulatedMachines="rhel6.5.0,pseries"
Comment 1 Lukas Svaty 2014-09-11 11:40:59 UTC 
Created attachment 936515 [details]
engine.log
Comment 2 Lukas Svaty 2014-09-11 11:41:16 UTC 
Created attachment 936516 [details]
vdsm from host
Comment 3 Omer Frenkel 2014-09-14 07:21:59 UTC 
Lukas, please note that the scheduling filtering that you see in the log is for the rerun, after the vm fail to run the first time, there is no another host to run on.

the real problem is the first run failure:
Thread-91::DEBUG::2014-09-11 07:32:18,750::BindingXMLRPC::1074::vds::(wrapper) return vmCreate with {'status': {'message': 'Unknown host operating system', 'code': 9}}

we need to understand why vdsm doesnt report the host information, can you please attach full vdsm log?

not sure where exactly this should go, setting to infra first to take a look on this.
Comment 5 Dan Kenigsberg 2014-09-14 15:51:33 UTC 
vdsm-4.14.15-5.gitb02671b.fc20.ppc64 does not include the fix for this bug (http://gerrit.ovirt.org/32653)

*** This bug has been marked as a duplicate of bug 1131965 ***