Skip to main content

Proposed network design with redundancy for a charity

More
17 years 10 months ago #17740 by panos

This is a normal lan with internet connection. The only but I see is the decision regarding the placment of isa servers, why not a cisco routers with firewall capabilities instead of the isa servers to save cost.


sose



Hi Sose, thanks for the comment, describing this as a normal lan with internet connection makes me fell much more confident for the design. Have you ever built such a network with multiple internet connections and just one router?

The ISA servers already exist so I'm going to just reuse them in order to save the cost.
More
17 years 10 months ago #17742 by panos

Still not too sure how you want to connect the remote office, its not clear what traffic is required to go between all the different locations. You mention http traffic, if thats just it then you can get away without having any logical link to the different offices. If they do need to link up, then possible a VPN solution would be the way to link the different sites to your main office.

As for the question about the ISA Loadbalancing, it is reletivly easy to do this within ISA 2004. Now, you can configure it through the ISA Snap-In and it will configure the WNLB for you (where with ISA 2000 you had to configure both independantly and then it didn't work very well so you had to rely on third part products to acheive it reliably)

Hope it helps, sorry i cannot really comment on the router side of things, i am currently studying the CCNP so its not my area of expertise.



Hi Smurf,

the traffic is all going to http and that's why I'm not going to use a VPN (might do in the future if security is a concern). Thanks for the advice on ISA 2004, it sounds easy to setup and I'll probably give it a try.

Good luck with your CCNP, I suppose it's really worth the effort!

Regards,
Panos
More
17 years 10 months ago #17743 by panos

Where is the egress point? It already fails because you only have one router.


I understand that creating a single point of failure is not a wise business and network decision but I can't think of a better way to approach this without breaking the bank :S

Do you have any recommendations on how this could be improved?

Thanks for replying in the first place :D
Panos
Time to create page: 0.119 seconds