[ovirt-users] New problem with hosted-engine during "Configuring the management bridge"

[Bob Doolittle]

Thanks John.

When hosted-engine aborts, it uninstalls everything.
So there is no webadmin available.

I've tried modifying the VDSM python code (e.g. 
/usr/share/vdsm/storage/multipath.py and 
/usr/lib64/python2.7/site-packages/vdsm/tool/service.py) to see/work 
around what's going wrong, but oddly VDSM does not seem to be affected. 
I suspect the original code (or bytecodes) are cached somewhere. 
Restarting vdsmd service has no effect. I'd really appreciate some 
insight there so I can work around it.

Does oVirt 3.4.1 work more smoothly with F20? I chose F19 because it's 
more stable at this point, thinking that things would be more likely to 
work smoothly. It's not turning out that way... I'm willing to start 
over with F20 if there's no path forward.

-Bob

On 05/14/2014 07:25 PM, John Taylor wrote:
> Bob,
> I remember something like this with an all-in-one install a while back
> where that error showed up,  but it was kind of red herring with
> multipath because the real problem was  ovirtmgmt bridge didn't get
> created. And that was a known problem I think with f19 see
> http://www.ovirt.org/OVirt_3.4_TestDay
> "Important Note: Known Fedora 19 bug: If the ovirtmgmt bridge is not
> successfully installed during initial host-setup, manually click on
> the host, setup networks, and add the ovirtmgmt bridge. "
>
>   If you can get to webadmin could you try to setup ovirtmgmt manually
> on the host.
>
> -John
>
> On Wed, May 14, 2014 at 6:58 AM, Bob Doolittle <bob at doolittle.us.com> wrote:
>> I could really use some help on this one. My efforts to debug VDSM via
>> instrumenting the python code are not working - the compiled code must be
>> cached somehow.
>>
>> Something is wrong with the way the multipathd service is being restarted.
>> It doesn't look too me that systemctl is even being called for it.
>>
>> Thanks,
>>    Bob
>>
>> On May 13, 2014 1:12 PM, "Bob Doolittle" <bob at doolittle.us.com> wrote:
>>> Maybe this isn't the actual problem after all.
>>>
>>> I replaced /sbin/multipath with a script runs the old version, but the
>>> suppresses those errors and returns exit status 0.
>>> But "vdsm-tool service-reload multipathd" is still failing and I don't
>>> know why.
>>>
>>> I have attached my vdsm.log file.
>>>
>>> Any guidance appreciated. I'll try digging through the python code for
>>> service.py and see if I can catch it when the multipath configuration is in
>>> place to see the exact issue.
>>>
>>> -Bob
>>>
>>> On 05/13/2014 12:27 PM, Bob Doolittle wrote:
>>>
>>> Hi,
>>>
>>> I have started a new installation as specified in the 3.4.1 release notes
>>> (fresh Fedora 19 install, yum localinstall
>>> http://resources.ovirt.org/pub/yum-repo/ovirt-release34.rpm).
>>>
>>> This is failing in the step "Configuring the management bridge".
>>>
>>> Based on the vdsm.log, it appears I am hitting:
>>> Bug 988995 - vdsm multipath.py restarts mutipathd, cutting the branch vdsm
>>> sits on
>>>
>>> "multipath -F" is returning "invalid keyword: getuid_callout" and it
>>> appears that this is causing vdsm-tool to abort (although the command exit
>>> status is 0 and the bug report says that those are only harmless warnings).
>>>
>>> There is no workaround stated in that bug report.
>>>
>>> Help?
>>>
>>> -Bob
>>>
>>>
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>