On Linux you can specifiy routes through certain cards. For instance if I have 3
cards with IPs of 10.0.0.1, 10.0.0.2, and 10.0.0.3 and I want traffic to
10.0.0.11 to go originate from 10.0.0.3 I add a route like such:

route add 10.0.0.11 gw 10.0.0.3

Not sure if this will work in a Sun environment and this only controls the
outgoing traffic.


Quoting Eugene Klaus <Eugene.A.Klaus AT FLAGSTAR DOT COM>:

> Terry,
>
> That is the nature of ether net. All outgoing info is through your
> primary card.
> You can receive on any of the three.
>
> Gene Klaus
>
>
>
>
>
> "Barth, Terry (MBS)" <Terry.Barth AT MORTGAGEFAMILY DOT COM> on 11/20/2000
> 08:47:38 AM
>
> Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
>
> To:   ADSM-L AT VM.MARIST DOT EDU
> cc:    (bcc: Eugene A Klaus/Flagstar_notes)
> Subject:  NIC Question
>
>
>
> Hello All -
>
> I have a sun solaris ADSM server that has 3 NIC cards. We would like to
> utilize all 3 NIC cards for backups and restores, independently.  We set
> up
> the cards according to SUN and everything seems fine on that end.
> However,
> when a backup starts off, it sends all data through the first NIC card
> and
> then when it receives data, it spans through all 3 cards. Our switch
> recognizes all 3 cards, but as soon as a backup begins on one of the
> other
> NICS - say for instance, 48.22, the MAC disappears and the switch no
> longer
> recognizes the card.  We feel that there is something on the ADSM
> server
> side that we need to set up so that if we tell a client to do a backup
> and
> designate that the server has an IP address of xx.xx.48.22, that all
> communications will go through this NIC card and not touch the others.
>
> xx.xx.48.20
> xx.xx.48.21
> xx.xx.48.22
>
> My question is, what is it that I need to set up on the ADSM server
> side, so
> that it uses each NIC card independently.
>
> I have placed the IP address in the opt file on the client end. Is
> there
> something that I need to do on the server end?
>
> Any assistance will be appreciated.
>
> Thanks
>
> Terry
>