single or multiple range statement

[Douglas Armstrong]

I have seen problems with this in our network, but we have a lot of old
gear,printers Win'95 boxes and old *nix distros that need to be supported
forsoftware on them, that balk talking to a .0 or .255 that are not
network/broadcasts. for this I have them removed from my config.

Doug Armstrong
www.ovationdata.com[1]


George C. Kaplan wrote: Simon Hobson wrote: However, you would be better
advised to use : range 10.10.16.1 10.10.16.254 range 10.10.17.1 10.10.17.254
and so on The reason is that there are some faulty IP stacks out in the real
world which are written on the assumption that things are using class C
networks - hence .0 would be the network, and .255 the broadcast address.
SInce there are a small number of devices that can't cope with having a .0
or.255, it will save hassles to avoid them in case you ever get one of these
devices on your network. Is this really a problem nowadays? We have a few
address pools that span a /24 boundary, and we haven't bothered to skip over
the .0 and .255 addresses. I don't recall ever getting trouble report for
anyof these addresses. Has anyone else on this list run across clients
(recently) that couldn't handle a .0 or .255 address? 


--- Links ---
   1 http://www.ovationdata.com