On Fri, Aug 03, 2007 at 05:16:23PM -0400, Peter Memishian wrote:
> 
>  >  svcadm restart network/physical
>  > And after restarting networking:
>  > -bash-3.00# ifconfig -a
>  > [..]
>  > iprb0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
>  >         inet 10.99.1.41 netmask ffffff00 broadcast 10.99.1.255
>  >         groupname PROD
>  >         ether 0:2:b3:11:de:dd
>  > iprb0:1: 
> flags=9040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,NOFAILOVER> mtu 
> 1500 index 2
>  >         inet 10.99.1.81 netmask ffffff00 broadcast 10.99.1.255
>  > iprb0:2: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
>  >         inet 0.0.0.0 netmask 0
>  > iprb0:3: 
> flags=9040842<BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,NOFAILOVER> mtu 
> 1500 index 2
>  >         inet 10.99.1.81 netmask ffffff00 broadcast 10.99.1.255
>  > iprb0:4: 
> flags=9040842<BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,NOFAILOVER> mtu 
> 1500 index 2
>  >         inet 10.99.1.81 netmask ffffff00 broadcast 10.99.1.255
>  > [..]
>  > 
>  > so after restarting network/physical SMF always adds additonal logical 
> interfaces (addif in /etc/hostname.<if> as mentioned in Sun Network Services 
> Guide ) 
> 
> I don't believe that network/physical expects to be "restarted", but this
> may have changed post-NWAM; cc'ing nwam-discuss.

If you've only enabled network/physical:default, the presence
of nwam should not have changed anything for you; the nwam
project simply added an alternative service instance
(network/physical:nwam).

And I think meem is correct: there is no "stop" method for
network/physical:default; so restarting that service is the
same as simply re-running the start method...which attempts
to configure the interfaces as directed by the /etc/hostname*
files.  Multiple plumbs of the same interface probably result
in error messages (check the service's log file); multiple
'addif' commands will result in multiple logical interfaces.

-renee
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to