[Users] ovirt-node post-reboot no persistent networks config .. !!! urgent update req for POC coming weekend

[Sven Kieske]

Hi,

thanks for all the clarifications.

We didn't find any documentation on how the node does setup the network
exactly step by step so we were surprised that the configuration we'd
done for eth1 (switch dhcp on) got appended to the ovirtmgmt bridge.
Maybe we didn't find the right docs, or where is this behavior
described?

If the network pages gets disabled after registration with the
management node this would maybe stop some confusion as well :-)

Keep up the great work!

Regards

Sven

On 02/10/13 15:13, Fabian Deutsch wrote:
> Am Mittwoch, den 02.10.2013, 11:26 +0000 schrieb Sven Kieske:
>> Hi,
>>
>> we were now able to test this.
>>
>> The IP does not show up in ovirt-node-setup TUI, however if you drop to
>> console you see this:
>>
>> [root at vroot4 ~]# ip a s
> 
> ...
> 
>> 2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
>>     link/ether 00:25:90:32:e4:88 brd ff:ff:ff:ff:ff:ff
>> 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
>> state UP qlen 1000
>>     link/ether 00:25:90:32:e4:89 brd ff:ff:ff:ff:ff:ff
>>     inet6 fe80::225:90ff:fe32:e489/64 scope link
>>        valid_lft forever preferred_lft forever
> 
> ...
> 
>> 8: ovirtmgmt: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue
>> state UNKNOWN
>>     link/ether 00:25:90:32:e4:89 brd ff:ff:ff:ff:ff:ff
>>     inet 10.0.1.101/24 brd 10.0.1.255 scope global ovirtmgmt
>>     inet6 fe80::225:90ff:fe32:e489/64 scope link
>>        valid_lft forever preferred_lft forever
>> 10: ;vdsmdummy;: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN
>>     link/ether 86:f2:1e:01:7d:4a brd ff:ff:ff:ff:ff:ff
> 
> ...
> 
>> So the address gets switched from eth1 to the virtual nic "ovirtmgmt"
> 
> What do you mean with switched?
> eth0 becomes the slave of the bridge ovirtmgmt, the bridge then get's
> the IP configuration.
> 
>> and does not show up in the TUI.
> 
> When a Node is registered with engine the network page should get
> disabled. That's currently not the case and can be seen as a bug.
> 
>> Is this an intended behavior?
> 
> What behavior exactly?
> 
>> Furthermore, for what reason does ovirt create all those bonding devices?
> 
> That should indeed be fixed.
> 
>> We installed the el6 variant of the iso.
>>
>> An additional info, which I noticed:
>>
>> lsb_release -a does not provide a correct output, it just states:
>>
>> "RedHatEnterpriseVirtualizationHypervisor" which is not very
>> informative, no version information or whatsoever.
> 
> Mike,
> do you know what the output should look like?