To put it another way -- each number in an IPv4 address being a 1-byte (8-bit) number; let's suppose we're trying to set up a network for 255.255.26.0/22 ... 255 = 1111 1111 255 = 1111 1111 26 = 0001 1010 0 = ....
Since the mask is /22, and the first two bytes gobble up 16 bits, it means we only care about the first six bits, or "0001 10" part, in decimal 26. But now we have a leftover binary "10", the "1" in which can never be tested because it's the 23rd bit and we just said only to test the first 22 bits. While there are certainly IPv4 addresses that would match a network like x.x.26.0/22 by plain old bit math, it's not a sensible combination because it's like saying "I will only accept sandwiches that have both oil *and* mayonnaise on them... although I will actually ignore the presence or absence of mayo and really only care about the oil." --Steve On Fri, May 25, 2012 at 5:23 PM, Kim Longenbaugh <[email protected]> wrote: > Thanks for the info, I was afraid that was the case. > > > > I hope your holiday good. > > > > Kim > > > > From: Reimer, Mark [mailto:[email protected]] > Sent: Friday, May 25, 2012 4:16 PM > > > To: NT System Admin Issues > Subject: RE: subnetting question > > > > Yes. For a /22 mask, you need to be on a 4,8,12… boundary. As in your > subnets below, the /23 all start with an even number and include the next > number (16-17, 18-19, etc) > > > > As you probably know, masks are based on binary math. I can’t quickly find a > tutorial for this, but many exist, as do subnet calculators. > > > > For you Americans, have a great long weekend. Us Canucks had ours last > weekend. > > > > Mark > > > > From: Kim Longenbaugh [mailto:[email protected]] > Sent: Friday, May 25, 2012 2:58 PM > To: NT System Admin Issues > Subject: subnetting question > > > > I must not understand all I know about subnetting…. > > > > Here’s the deal: > > I started out with these two subnets (I’m including the subnet and broadcast > addresses): > > 172.16.8.0/22 = 172.16.8.0-172.16.11.255 > > 172.16.12.0/22 = 172.16.12.0-172.16.15.255 > > > > I planned to segment my campus, so I planned to use the following subnets, > which I wanted to have 512 addresses instead of 1024: > > 172.16.16.0-172.16.17.255/23 not in use yet > > 172.16.18.0-172.16.19.255/23 not in use yet > > 172.16.20.0-172.16.21.255/23 not in use yet > > 172.16.22.1-172.16.23.254/23 in use, working ok > > 172.16.24.1-172.16.25.254/23 not in use yet > > > > However, when I tried to use this subnet, with a /22 for 1024 addresses, my > router doesn’t like it, and when creating a DHCP scope, it tries to make a > superscope. > > 172.16.26.0/22 = 172.16.26.0-172.16.29.255 > > > > It seems to me I should be able to do what I’ve described, but apparently > I’m misunderstanding things. > > So one question is, why can’t I do what I’ve outlined? > > > > Another way to put it is, if I want a /22 subnet, I still have to make them > on the 4,8,12,16,20,24,28….. boundaries? > > ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~ --- To manage subscriptions click here: http://lyris.sunbelt-software.com/read/my_forums/ or send an email to [email protected] with the body: unsubscribe ntsysadmin
