Vanessa Avalos
Cisco Networking Academy Semester 3
Chapter 8
Network Management

Network documentation is the most critical and least performed task.  Network documentation starts with your work journal but also includes cutsheet diagrams, logical diagrams, MDF & IDF layouts, Software listings, Maintenance records and security measures.

The Cutsheet
The Logical Diagram Cutsheet indicates where, what type of, length and termination type of cabling.  It also includes location of jacks.

MDF & IDF Layout
The physical and logical layout of the MDF & any IDFs includes:  
physical rack mounts 
auxiliary equipment 
servers, patch panels
labels
configurations

server & workstation details about each host attached to the network – 
    such as information on make & model, drives (CD-ROM, floppy), serial number, ram, cards (NIC, sound card), asset tags, 
    configuration details – IP & MAC addresses, IRQ, DMA and base memory addresses
    physical location, purchase date and warranty information

Software Listings – software configurations, operating system, 

Maintenance Records – a list of all repairs done on all equipment.  This will help predict and solve any potential or future problems.

Security Measures – this includes only soft security such as:
User rights, Firewall support and password definitions.

Physical Security – Identify how the MDF and IDFs are locked; who has access to these rooms and why; how are the hosts protected; who has physical access to the system;

User policies – such as what is and is not permited, consequences, minimum user ID an password lengths, rules for password content, user access to resources.  Make sure the network policies do not conflict with company policies

All these documents mentioned above create a set of network documents for the system.  This document set will allow upgrades and maintenance to procede in an organized manner.  It can be a starting place for upgrades.  Or a way to recover from a network failure.  THE DOCUMENTATION MUST BE CONTINUALLY UPDATED

Network access or network security – two parts:
1. Unauthorized access – both internal and external
2. Recover from catastrophic events

    Network access starts with security:
        Security measures such as:  minimum password length, maximum password age, unique passwords, logging on a specific times.

Data recovery: Protecting Data From Loss
There are many methods to do this.  Three of the more popular methods are: Tape backup, fault tolerant disks and uninterruptible power supplies UPS.

Tape backups are among the most common because they have a very low expense and a large data capacity.  But the data is stored sequentially and must be recovered the same way.

There are different types of backups because they cause a drain on resources.
Most types work with a flag or switched archive bit.  This bit is stored in a file and turned on when the file is created or modified.  The bit turned on means that the file needs to be backed up.  After the file is stored on the tape the bit is switched off.
Most companies recommend that backups be stored off the premises in case of fire or water damage.

Five types of backup operations:

Full Backup - all files on disk are stored to tape and archive bit for all files turned off

Incremental Backup - backs up all files that have been created or modified since the last full backup, only works in conjunction with a full backup and that any file has its archive bit turned back on

Differential backup - Backs up all the files that have been created or modified since the last full backup. The archive bit is not reset. Each time a differential backup is done, all of the files modified or created since the last full backup will be stored again

Copy Backup - Backs up user selected files to tape. Does not reset the archive bit to off

Daily Backup - Backs up only the files that are modified on the day of the backup. Does not reset the archive bit to off

The first three backup procedures are the most widely used.

To do incremental backup
On Monday do full backup. This will turn off all archive bits
On Tuesday do incremental backup on a separate tape.  This stores all files modified on Tuesday and resets their archive bit
This process is to be repeated for each day of the week and each day with a separate tape
This gives a complete backup of all files modified during that week.

An advantage to this is it require least amount of time per day, so it has least impact on network resources.
A disadvantage is, if you need to restore the backup you must first restore full backup and then all the incremental tapes in order, and this takes a great deal of time.

Redundancy Techniques: Protecting Data through fault tolerant storage devices
This type of redundant set of devices is categorized by RAID (Redundant Array of Inexpensive Disks) levels 0-5.

RAID 0 - Stripes data across multiple disks, no parity, so there is no redundancy. 

RAID 1 - Disk mirroring (disk duplexing) writes data to two identical partitions on separate hard disks thus creating an automatic backup. Disk duplexing uses two hard disk controller cards as well as two hard disks to prevent the controller card as being the single point of failure for the system as in disk mirroring. 

RAID 2 - Writes data across multiple hard disks, with error checking. This system is not used any more because it requires expensive disk modifications to work. 

RAID 3 - Stripes data one byte at a time and has a dedicated parity drive. A good, but expensive redundancy choice. Because of the expense, this solution is not used very often either. 

RAID 4 - Stripes data one sector at a time and has a dedicated parity drive. An expensive redundancy choice that is very slow on data writes to the disk. Because of the expense and the slowness of writing, this solution is not used very often either. 

RAID 5 - Stripes data and parity across multiple disks (at least three for RAID 5). By mixing the parity across all of the disks, a separate parity disk is not required and yet full data redundancy is achieved. Data writes to the disk are still slow, but the cost isn't so high. One other important fact about RAID 5 is that on a Windows NT system the boot and system partitions cannot be located on a RAID 5 disk array.

A volume is refers to a physical unit of storage. Volume sets are used quite often in networks. They provide NO data redundancy; they are just a way to refer to large storage areas as a single unit.

Many environmental factors can effect your network, such as static, dust, dirt and heat. 
Take precautions to protect yourself and your equipment from static electricity, dust, dirt and excessive heat.  These precautions include: 
Making sure that all of the equipment's power switches are OFF before hooking it up; 
grounding yourself before touching the inside of the computer; 
keeping the environment in which equipment is used clean and free of contaminants; 
no smoking around equipment; 
keeping food and drinks away from equipment;
making sure that your work area leaves the computers vents open;
making sure that you place the computer on a solid support area;

Power conditioning:
Protect your equipment from irregularities in your building's electrical wiring. The easiest way to do this is to put it on separate circuits in your building.  This will solve some of your problems.  You can also use other devices such as:
Isolating transformers: Controls voltage spikes and high frequency noise.
Regulators: Maintains a constant output voltage despite changes in voltage, handles brownouts and voltage surges.
Line Conditioners: This is a regulator with an isolating transformer built in.
Uninterruptible power supplies: This device will allow computer to run even if there is a power failure.

EMI and RFI
Problems with the network can actually be caused by the equipment itself.  Power supplies, monitors, fluorescent lights, large electric motors, and electrical wiring, can cause Electro-Magnetic Interference (EMI) and Radio Frequency Interference (RFI).  These types of problems can be very difficult to diagnose and are usually discovered by use of diagnostic software and hardware.

Software Viruses: A type of software whose sole purpose is to disrupt the operation of a network.
You can take some steps to avoid damage by viruses by:
Being careful about getting software, without knowing specifically where it comes from. 
Being wary of other people using your computer with their disks. 
Using a current virus checker on all computers.

Network baselines:

Network performance is a measurement of a network's quickness and reliability. Every combination of computer and network hardware, software, and cabling will have a different network performance. In order to know when your network is performing poorly, you must have a measurement to compare against.  This measurement is called a baseline. It is established after network has been installed and configured properly.
You can use this base line to determine the heath of network, as network grows periodically update it.
Always verify the operation of equipment after a move.

Peer to Peer – 10 or less computers, sharing off a drive with other people.  Doesn’t need specialized software.
Advantages are its cheaper to create and operate, it allows users to control their own resources, it does not require a dedicated server, and requires no additional software except the operation system. 
Disadvantages are it has no central point of management, the user shares resources on their machine, and a limitation of no more than ten users

Client/server – a real network, advantages, centralized users – 
Advantages: a centralized point of user, security, and resource management; servers can provide specific resources to clients. Also provides access to resources
Disadvantages: there is now a single point of failure, requires specially trained personnel to maintain the network, and special network software and hardware add greatly to the cost but
a client-server network is really the only choice for businesses with more than ten users

UNIX – The Unix operating system comes in many variations, Unix is a multi-user operating system that supports multiprocessing, multitasking, and multithreaded applications. The operating system is Kernel based and uses NFS

NetWare  - comes in many different versions, uses a combination of 2 file services. The is 1st FAT (file allocation table), which is used for DOS, the 2nd is DET (Directory Entry Table) a proprietary Novell file system which provides for both file and directory security on the server.

NT - just like Unix, a multi-user operating system that supports multiprocessing, multitasking, multithreaded applications and kernel based, uses both FAT16 and NT's own proprietary system, NTFS (New Technology File System)

Server management and control – 
NOS - like an OS for networks, its main function is to control the network. It does this by establishing network users rights, login accounts, profiles, policies, etc.

Network trouble shooting – 
The scientific method:
1.  Observe some aspect of universe
2.  invent a theory
3.  test it
4.  modify it
5.  return to step 3
6.  keep testing.  

Example: User calls, can’t get on internet.  Help desk writes report gives to you. Talk to user, ask What did you do?  Tells you, “nothing different.”  Check logs, computer was upgraded last night, computer network drives incorrectly configured, check network neighborhood,  OK.  Ping server, gateway & NIC (127.0.0.1).  Decide might be problem with server. Ping server from next desk, get OK!  Server is fine, backbone is fine & server connection to backbone.  Server IDF ping server, connection fine.  Get new cable on patchpanel and ping.