No, we actually carve out one or more subnet for each VPS host and assign individual IPs to each VPS. Few IPs are wasted.
The only drawback is that a VPS must change IP to be shifted to an alternate node. -- Randy ---------- Original Message ----------- From: "Shaun R." <[email protected]> To: [email protected] Sent: Thu, 27 Aug 2009 09:32:30 -0700 Subject: Re: [c-nsp] Large networks > > With the number of virtual servers most of us are hosting you would run > > out of > > VLAN's very quickly. What I do is static route subnets to host nodes and > > let > > the host nodes do the L3 work. This takes care of MAC address conflicts, > > spoofing, and many other problems. > > So you still carve out a subnet for each vps? For me that didnt make > sense to do because now each customer gets 4 ips (/30) and it seames > like a waist of space. a vps with 40 nodes with consume 2/3's of a > /24. With IPv6 i guess it wouldnt matter :) I was just trying to > do my best to save ips. ebtables acts as a ACL on the edge interface. > > ~Shaun > > _______________________________________________ > cisco-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/cisco-nsp > archive at http://puck.nether.net/pipermail/cisco-nsp/ ------- End of Original Message ------- _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
