home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.protocols.tcp-ip:3813 comp.dcom.sys.cisco:985
- Path: sparky!uunet!elroy.jpl.nasa.gov!nntp-server.caltech.edu!mustang.mst6.lanl.gov!ucsbcsl!spectrum.CMC.COM!fennel.acc.com!art
- From: art@acc.com (Art Berggreen)
- Newsgroups: comp.protocols.tcp-ip,comp.dcom.sys.cisco
- Subject: Re: Multiple Subnets and Non-byte boundary netmasks
- Message-ID: <1992Jul24.160634.5645@acc.com>
- Date: 24 Jul 92 16:06:34 GMT
- References: <1992Jul23.224600.11081@polari>
- Organization: ACC, Advanced Computer Communications
- Lines: 23
-
- In article <1992Jul23.224600.11081@polari> jsanchez@polari.online.com writes:
- >A potential customer of ours wants some advice and I thought I would run
- >it past the net. Some/all of this may be a FAQ but bear with me. First,
- >he wants to be able to create lots of subnets and as a consequence is
- >parsimonious with ip addresses. He thinks that being able to run multiple
- >subnets from one router port would allow him to have small subnets (to
- >save IP addresses) and yet put several of them together in one physical
- >ethernet segment. Kinda weird I admit but that is the way customers are.
- >
- >To accomplish the above, he plans on running a netmask that is not on byte
- >boundaries. Nothing in principal wrong with this but are there practical
- >problems that would lead you to avoid this kind of configuration.
-
- Having multiple subnets one a wire is not that uncommon. It's often
- done when a site wants to have all subnets be the same size (hard to
- avoid when running RIP), and some LANs have many more hosts than other
- LANS.
-
- The only problem I can think of right now with non octet aligned masks,
- is that some older host TCP/IP implementations would only take octet
- sized chuncks on netmasks.
-
- Art
-