[Ovirt-devel] [Fwd: host not shown in default hardware pool]

Fri Apr 17 14:46:23 UTC 2009

[root at physical log]# dig +short srv _ovirt._tcp.priv.ovirt.org
0 0 80 management.priv.ovirt.org.

dns seems to be ok. I verified on host/node + managment that the
resolvers point to .50.1 and .50.2. and the srv/a resolves properly.

physical keeps saying:
Apr 17 16:35:48 physical libvirt-qpid: GSSAPI Error: Unspecified GSS
failure.  Minor code may provide more information (No credentials cache
found)

management keeps saying:
Apr 17 16:35:48 management qpidd[2158]: 2009-apr-17 16:35:48 warning
Client closed connection with 501: internal-error: Sasl error: SASL(-1):
generic failure: GSSAPI Error: Unspecified GSS failure.  Minor code may
provide more information (No credentials cache found)
(qpid/client/SaslFactory.cpp:226)

On physical I do not have a /var/log/ovirt.log (neither on management).
I have:

[root at management log]# ll ovirt-server/
total 968
-rw-r--r-- 1 ovirt ovirt   6048 2009-04-17 16:32 db-omatic.log
-rw-rw-rw- 1 root  root    9400 2009-04-17 15:40 host-browser.log
-rw-r--r-- 1 ovirt ovirt   5263 2009-04-17 16:32 mongrel.log
-rw-r--r-- 1 ovirt ovirt 950770 2009-04-17 16:38 rails.log
-rw-r--r-- 1 ovirt ovirt   3783 2009-04-17 16:32 taskomatic.log

and

[root at physical log]# ll ovirt-host.log
-rw-r--r-- 1 root root 0 2009-04-16 16:24 ovirt-host.log

Probably most information goes into /var/log/messages?

I've been running also the following test:
service libvirt-qpid stop
export QPID_TRACE=1
libvirt-qpid --broker <hostname-of-server>

Please see the output attached as file.

Any further ideas? Why is the kerberos failing? The client side gives
up. That's what I read from the logs.

regards,

Ingmar

Joey Boggs wrote:
> 
> On the node try running:     dig +short srv _ovirt._tcp.priv.ovirt.org
> it should return the ip address of your appliance, that could cause the
> gssapi errors since it's not able to find to the server. Also on the
> node can you attach /var/log/ovirt.log which will let us know what's
> going on when the ovirt services start.
> 
> 
> 
> Ingmar Schraub wrote:
>> Interesting... the host shows now up after running "ovirt-identify-node
>> -v -s 192.168.50.2 -p 12120". It is unavailable though.
>>
>> I shut the appliance down, re-started libvirtd on the host, powered the
>> appliance back on. Host is still unavailable.
>>
>> I ran the test "ruby /usr/share/ovirt-server/qmf-libvirt-example.rb",
>> but this gave me only
>> Connecting to amqp://management.priv.ovirt.org:5672..
>> ----------------------------
>> ----------------------------
>>
>> I believe that the kerberos authentication fails (as the logs I provided
>> indicate). But why? date/time is in sync on host/appliance. /etc/hosts
>> knows about management.priv.ovirt.org and physical.priv.ovirt.org,
>> hostname is set properly, ...
>>
>> I have currently only 1 physical machine (2xIntel Xeon Quadcore, 8GB
>> RAM, 4xnic,...). On this host (ovirtbr0 - 192.168.50.1, eth0 -
>> 192.168.100.3), I have the 'appliance' running which is the 'management'
>> station (eth0 - 192.168.50.2).
>>
>>
>> Here is the log excerpt from db-omatic.log as requested.
>>
>> INFO Fri Apr 17 13:23:26 +0200 2009 (2276) dbomatic started.
>> ERROR Fri Apr 17 13:23:36 +0200 2009 (2276) Unable to determine qpid
>> server fromm
>>  DNS SRV record, retrying..
>> INFO Fri Apr 17 13:23:41 +0200 2009 (2276) Connecting to
>> amqp://management.priv..
>> ovirt.org:5672
>> ERROR Fri Apr 17 15:43:26 +0200 2009 (2276) Error in db-omatic: SIGTERM
>> ERROR Fri Apr 17 15:43:26 +0200 2009 (2276)
>> /usr/share/ovirt-server/db-omatic/dbb
>> _omatic.rb:433:in
>> `sleep'/usr/share/ovirt-server/db-omatic/db_omatic.rb:433:in ``
>> check_heartbeats'/usr/share/ovirt-server/db-omatic/db_omatic.rb:470:in
>> `main'/uss
>> r/share/ovirt-server/db-omatic/db_omatic.rb:474
>> INFO Fri Apr 17 15:45:18 +0200 2009 (2297) dbomatic started.
>> ERROR Fri Apr 17 15:45:28 +0200 2009 (2297) Unable to determine qpid
>> server fromm
>>  DNS SRV record, retrying..
>> INFO Fri Apr 17 15:45:33 +0200 2009 (2297) Connecting to
>> amqp://management.priv..
>> ovirt.org:5672
>> INFO Fri Apr 17 15:45:33 +0200 2009 (2297) Marking host
>> physical.priv.ovirt.org
>> unavailable
>>
>> regards,
>>
>> Ingmar
>>
>>
>> Joey Boggs wrote:
>>  
>>> Can you also grab the contents of /var/log/ovirt-server/db-omatic.log
>>> from the appliance?
>>>
>>> Just to make sure, your node is the physical machine running the
>>> appliance?
>>>
>>>
>>>
>>> Ingmar Schraub wrote:
>>>    
>>>> Hello,
>>>>
>>>> in addition to my trouble report from this morning, I further
>>>> noticed in
>>>> the logs the following:
>>>>
>>>> HOST:
>>>> -----
>>>> Apr 17 14:15:25 physical libvirt-qpid: GSSAPI Error: Unspecified GSS
>>>> failure.  Minor code may provide more information (No credentials cache
>>>> found)
>>>>
>>>> ovirt-appliance:
>>>> ----------------
>>>> Apr 17 14:15:25 management qpidd[2162]: 2009-apr-17 14:15:25 warning
>>>> Client closed connection with 501: internal-error: Sasl error:
>>>> SASL(-1):
>>>> generic failure: GSSAPI Error: Unspecified GSS failure.  Minor code may
>>>> provide more information (No credentials cache found)
>>>> (qpid/client/SaslFactory.cpp:226)
>>>>
>>>> It seems like the authentication fails and therefore the appliances
>>>> doesn't 'see' the host.
>>>>
>>>> Does anybody know whether this is a libvirt-qpid problem or a qpidd
>>>> problem or some wrong configuration in ovirt? Is this a known issue or
>>>> did I hit something new?
>>>>
>>>> I am really stuck here...
>>>>
>>>> Thanks a lot and best regards,
>>>>
>>>> Ingmar
>>>>  
>>>> ------------------------------------------------------------------------
>>>>
>>>>
>>>> Subject:
>>>> host not shown in default hardware pool
>>>> From:
>>>> Ingmar Schraub <is at eseco.de>
>>>> Date:
>>>> Fri, 17 Apr 2009 10:49:33 +0200
>>>> To:
>>>> ovirt-devel at redhat.com
>>>>
>>>> To:
>>>> ovirt-devel at redhat.com
>>>>
>>>>
>>>> Hello,
>>>>
>>>> I am new to ovirt and please bare with me if my question has been asked
>>>> before. I googled, studied the wiki and searched the bug list. But I
>>>> couldn't find anything similar.
>>>>
>>>> I've just set up a new server with F10, fully patched and installed
>>>> ovirt using the Rawhide build instructions.
>>>>
>>>> All seems to work nicely, but I am stuck at one point.
>>>>
>>>> After 'sudo ovirt-install-node-stateful' followed by a host reboot I
>>>> started the appliance again. My problem is that the host is not
>>>> shown at
>>>> all in the default hardware pool. Any hints what's going wrong here?
>>>> What could be missing that the appliance doesn't see my host?
>>>> Networking, bridge, etc. is all ok.
>>>>
>>>> I had also to tweak /etc/httpd/conf.d/ovirt-server.conf because the
>>>> ovirt-appliance httpd doesn't start with the config file provided.
>>>> There
>>>> is already a ticket open on bugzilla. (#489819).
>>>>
>>>> Thanks a lot in advance.
>>>>
>>>> regards,
>>>>
>>>> Ingmar
>>>>
>>>>  
>>>> ------------------------------------------------------------------------
>>>>
>>>>
>>>> _______________________________________________
>>>> Ovirt-devel mailing list
>>>> Ovirt-devel at redhat.com
>>>> https://www.redhat.com/mailman/listinfo/ovirt-devel
>>>>         
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: libvirtqpid.log
Type: text/x-log
Size: 8982 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/ovirt-devel/attachments/20090417/7e2c42aa/attachment.bin>