R/S: Troubleshooting

Don’t do things in random order, but a set of steps.

First step is to define the problem.  Try to isolate the problem. 

Four step troubleshooting:

1 Ping loopback address- trying to isolate the problem Ping 127.0.0.1 When you ping the loopback, you are checking if the TCP/IP software is on the computer, available and working correctly.  There is always a file on the computer that has the loopback address, and if that folder is deleted, the network connections will not work correctly.

2 ping ip address of the local system (192.168.40.10)  When pinging the local system, you are checking if the NIC is working or not.  If these two steps are successful, you know that the user’s machine is ok. 

3 Ping the default gateway ex: ping 192.168.40.1.  A default gateway is the router, which is your translator between the inside network and your outside network. 

4 ping a system outside your local network- google, yahoo, ect.  At this point, we know that the local system is ok, now we are looking to the outside network for the issue.

***These are the specific order that has to be done on the tests and the CCENT test***

More After the Break
 
Commands:

Ping- this is simply a “hey are you there?” The ping is a small amount of packets that are used to check connectivity between two devices.

IPconfig- windows machines  IFconfig – linux/unix/mac : to manage TCP/IP settings such as ip address, default gateway, network mask, DNS setting and DHCP settings.  IPconfig can both display and allow you to change the above information. 

Traceroute (linux/unix/mac); Tracert (windows)- traces the route that a packet takes to get to its destination.  It shows all of the hops(routers) as it goes out.  This can be done when your ping is not coming back correctly.

Arp –a- to check mac table on a system A mac table is the mapping of the IP and physical addresses for your machine.  Show ip arp- works on Cisco devices only.
   
Subnetting

The process of breaking a network into smaller units, or segments, into subnets is called subnetting.  The reason we do subnetting is for better management and troubleshooting inside the network.  Another reason is performance within the network; bandwidth and network traffic.  Adding a router (which allows the subnets to talk to each other) breaks down the broadcast domain.  This is advantageous because broadcast traffic takes up a part of your bandwidth, and filtering unneeded traffic out frees your bandwidth up for other uses.  Another reason to subnet is for higher security.  One of the ways there is higher security is controlling what network gets what traffic.  Another is in the case of attacks- you can shut down the affected network and contain the attack without shutting down the entire network. 

CSMA/CD- carrier sense multiple access collision detection  Carrier is the wire that carries your packets.  Carrier sense monitors your medium.  Multiple access means that more than one machine has access to the same medium.  Collision Detection watches for collisions between packets and informs the machines using the medium when there is a collision to prevent information backups.

How do we subnet?

Class C-32 bits

                N  N  N  H

                First three segments (network) are 24 bits, last one (host) is 8 bits.  The host part is where we will be borrowing bits from to create the subnet.  You must always leave at least 2 bits for the host part.  For class C, you can borrow up to 6 bits.   

192.168.10.0/255.255.255.0  The mask is used to hide the network information of the IP address.  A router doesn’t care about the host part of the IP, just the network.  The way that a router decides which part is which is by looking at the network mask.  It allows the router to decide what to do with the packet- if its internal or meant for external. 

255.255.255.0 is a pain to type out every time, so we use CIDR notation- classless inter domain routing notation.  All we do with this is count the amount of 1’s in a mask.  So, 255.255.255.0 is 11111111.11111111.11111111.00000000 in binary so its /24 in CIDR notation.  SO, if you want to write that mask in CIDR notation format, you would do it with 192.168.10.0/24.  YOU CANNOT USE CIDR FOR A ROUTER! 

130 users in network X.  They wish to break down into two networks, one with 30 users, the other with 100 users. 

Design Requirements

                How many subnets do I need?  2

                How many hosts per subnet? (How many IP address do I need for this subnet?) 100 (this number is the biggest number desired for a subnet.)

These two questions must be answered before moving on with your subnetting. 

So.  I’m going to borrow 1 bit.  This is shown as 2 to the power of 1 (2¹ =2).  The last 8 bits go from 0000 0000 to 1000 0000.  Now the remaining bits are 7, 8-1=7.  To show how many hosts per subnet= 2⁷ – 2.  The 7 is the REMAINING BITS IN THE ADDRESS.  So,  2⁷ -2 = 126.  This means that you have met both requirements for your design.  We subtract 2 from the hosts/subnet to allow the use of the subnet and broadcast addresses, which are not available to be assigned to a host.    

128 64 32 16 8 4 2 1                           Bits Borrowed = decimal value

2⁷ 2⁶ 2⁵ 2⁴ 2³ 2² 2¹ 2⁰                                        1= 128                   4=240

                                                                2=192                   

                                                                3=226

Because we have borrowed something from the host, the mask changes to 255.255.255.128.  128 represents the 1 bit we borrowed.  The CIDR changes to /25 since the number of 1’s has gone up one. 

The next question to answer is What is the block size? This is the same thing as the number of hosts, just without subtracting 2.  2⁷ = 128. 

Subnet Address – the first one is always the same as what you started with- 192.168.10.0.  The second one will be decided by your block size; 192.168.10.128.

Useable Addresses- the range of addresses between the Subnet and Broadcast addresses.  First:  192.168.10.1- 192.168.10.126.  Second:  192.168.10.129-192.168.10.254.

Broadcast Address- the first one is ALWAYS one less than the second subnet you have defined.  192.168.10.127; Second or last is always 255.  192.168.10.255.  Internally the router will flip the remaining host bits to one, and that sets the broadcast address. 

*****

200 users in network A.  They want to set up 3 subnets with 10 users in A1, 50 in A2, and 140 in A3. 

Design Requirements:

How many subnets do I need? 3

How many hosts per subnet? 140

What is the block size?

Subnet Address:

Useable Addresses:

Broadcast Address: 

192.168.50.0/24  This is our starting IP address. 

2²= 4 subnets.  This meets our requirements. 

Remaining bits in host part are 6; 8-2=6.

Hosts per subnet is 2⁶ -2 = 62.  Uh oh!  Does not meet our requirements for host per subnet.  This means that we cannot meet the requirements of this customer using class c addresses.  We must go up a step to class b.

****

Class B subnetting

                N N H H

                16      16

So.  Borrow two bits = 2² = 4 subnets

Remaining bits in host: 16-2 =14.  Our new IP address is 172.20.0.0/16. 

The new mask is 255.255.192.0 or /18.  16 +2 = 18. 

Design Requirements:

How many subnets do I need? 3

How many hosts per subnet?  2^14 – 2 = 16,382

What is the block size? 2^6= 64.  Or, 256-192 = 64.  The 192 is from the new mask.

Subnet Address:   a) 172.20.0.0   b)172.20.64.0      c)172.20.128.0   d)172.20.192.0

Useable Addresses:  a)172.20.0.1 – 172.20.63.254     b)172.20.64.1-172.20.127.245    c)172.20.128.1-172.20.191.254     d)172.20.192.1-172.20.255.254. 

Broadcast Address:   a)172.20.63.255     b)172.20.127.255    c)172.20.191.255      d)172.20.255.255

VLSM (variable length subnet mask)

Allows us to create a routing summery to make it easier for the router to deal with the incoming information. The summery routs must be continuous. 

Wide area networks only need 2 IP addresses. 

Classful routing- done by a piece of software.  All router interfaces having the same mask.  Examples of classful routing protocols are RIPV1 and IGRP.  

Classless routing- router interfaces can have different masks.  We use this to customize/optimize uses of IP addresses.  Examples are RIPV2, EIGRP, OSPF.  You must have a router that supports classless routing. 

How to design a network with classless routing

Starting address is 192.168.60.0/24

Network Name	Hosts Needed	Block Size	CIDR Value	Mask	Subnet Address
A	10	16	/28	255.255.255.240	192.168.60.0
B	30	32	/27	255.255.255.224	192.168.60.32
C	2	4	/30	255.255.255.52

  Block size is the closest number to the hosts allowing for the 2 unusable IP addresses.  Ex- 32 hosts requires 64 block size.   CIDR is 24 plus remaining bits.   The starting of the following networks must be multiples of the block size.  WANs are usually put at either the end of the table or beginning.