quick subnetting tutorial/refresher

[xyyz]

I have my sem3 skills exam tomorrow... and I found out the core of the exam is going to be subnetting...

This is bloody brilliant because subnetting is my WEAKEST link (so to speak). I've had a hard enough time with extended access lists and their placement (I know they are closest to the source... but is that on the closest interface or the exit interface of the router?)

I mean i've spent alot of time with subnetting and just when I think I have it I ask myself something and totally confuse myself.

I have Lemele's book (which I think was a total waste of money), I have access to the online curriculum (useless in this department) and I have the Coriellous (sp?) routing and switching book.

The skills exam is at 1:30 pm and I need to make my skills and understanding razor sharp by then.

If anyone can run a quick tutorial for me I'll name my first born after you, your username included. Otherwise, i'd be very appreciative if you could tell me of some sources I have access to at ohh... 2:11am.

 

[ScottMac]

What always worked for me when taking tests with subnetting questions was to make a visual aid. For computer-based tests, I make the chart before I hit the "start" key.

Draw a grid, 16 spaces wide, from right-to-left label the grid at the top 1,2,4,8,16,32,64....etc and on the bottom, label it with the corresponding subnet mask (again, right-to-left)255,254,252,248,....

When you start getting questions about how many hosts per network, most efficient masking for X number of networks with y number of hosts each...etc, you have a chart right in front of you to work it out.

For "Which network is this host in?" kind of questions, it's fairly fast & easy when you look at the grid, or worse case, use bit tallys under the grid.

Remember that the bit values in the "Network" part of the address can be a zero or a one. Having a "255" in the network part of the subnet mask defines the RANGE of available subnets...the mask is all ones, the actual address probably isn't.

If you're getting questions on CIDR blocking and supernetting, refer to the grid from above, tally the bits, find the highest number of common bits...that's (most likely)your supernet.

Remember that the address class is based on the most significant bits of the address, not the subnet mask.

"Network Number" is all bits in the host portion of the address is zero, "Broadcast Address" is all one in the host portion. Everything in-between is a valid address (by CONVENTION). All zeros or all ones CAN be a valid host address, depending on the foundation of your network operating systems. The old BSD system used all ZEROS as a broadcast address....so on the old BSD system, 255 is a valid host address. The System V UNIX system used all ones as a broadcast address(and that seems to be the "winner&quot😉..so on a System V based network, a host address of Zero is OK. SO...by the RULES you CAN have a zero or 255 host address, by CONVENTION, those addresses are usually not used for hosts.

That's about it for me (it's still early, and I'm still pre-caffinated.....)

Good Luck

Scott