Re: how is it possible to assign ip ending in 255 ?
> Hello,...
>
> I work in a IT-Network help desktop (help desk that is designed to
> fix/assist in network problems) and few weeks ago i got a call from a
> service Technion that asked me to check ips that end in 255.
>
> At first i didn't believe him (thought he was making a prank on me or
> just being an #$$).
> Any way i tested and i was able to find the problem (the pc had a
> buggy card) but i started to wonder -
> since ip 255 is for broadcasting how is it that even go tru routers ?
The simple answer is that, despite everything you may have been
taught, .255 is not automatically a broadcast address. Neither is .0
automatically a network address.
The broadcast address is the last IP in the subnet.
So if you have a /23 (netmask of 255.255.254.0), it'd be like this:
192.168.0.0 - not usable (network address)
192.168.0.255 - useable
192.168.1.0 - useable
192.168.1.255 - not useable (broadcast address)
The old paradigm of .255 is from the days when the ipv4 space was
classful.
Unfortunately, entirely too much network gear wants to treat an ip
space like it's a bunch of /24's, and some tcp/ip stacks won't respond
to .255's (Windows XP is, I believe, broken in this regard, but it's
supposedly fixed for Vista) so network admins still have to act
like .255's aren't usable if they want to ensure *everyone* can connect
to an IP.
> while it is possible to use ip such as 10.15.0.0 broadcast
> 10.15.255.255 (and 10.15.15.255 can be used) but Afair it should not
> be transfered via routers (as 169.254.x.x).
It all depends on how it's subnetted. If that IP space is broken up
into a /19 (255.255.224.0), then 10.15.255.255 would indeed be a valid
IP (10.31.255.255 would be the broadcast subnet)
If you already have a good grasp on subnetting, try to wrap your skull
around VLSM and it should become a bit more clear. But as a general
rule, don't assign .0 and .255 as live IP's unless you know it's
something you can get away with
Reply to: