[f-nsp] Server Iron CGC16

Wouter Prins wp at null0.nl
Wed Feb 4 12:09:01 EST 2009


Adam,

I assume this is all in the same layer2 domain and that the default gateway
of the realserver is set to something else as the SI's.

What's important is to have the flow from the server back to the client to
go through the serveriron carrying the active VIP. You can force this by
using source-nat, or by using DSR (which is not an option for you). Another
alternative is to setup VRRP and creating vlans 'above' the SI's, and one
where the realservers sits in. You can avoid asymetric traffic by using
vip-group's. (Only on chassis versions)

If this is not the same layer2 domain, then please explain your exact setup
with configs etc. :)

Regards,
Wouter

2009/2/4 Adam Li <adam_li at gti.com.hk>

> Hi All,
>
> I am rookie in load balancer and Foundry. Some questions are killing me
> recently.
>
> How can I perform link fails fail-over in SSLB? It seems only will
> fail-over when the active SI loss power or both (inbound and outbound)
> link down.
> If just unplug the cable which connecting to the real servers's L2
> switches. It still be active but fail to connect to servers from
> clients. How can I solve it without change another device's
> configuration. Because I have no right to change any another setting
> except the SI and switches.
>
> I have not VRRP setup and no multiple VIPS. Only single VIP with
> sym-priority is used. Anyone please help.
>
> Adam
>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/foundry-nsp/attachments/20090204/44d483e2/attachment.html>


More information about the foundry-nsp mailing list