Tasks and Responsibilities
The basic tasks and responsibilities of a Network Administrator are:
• Staff training and support
• Software installation, maintenance, and upgrading
• Hardware installation, maintenance, and upgrading
• Research and troubleshooting
• Routine administration and maintenance
• Network documentation
• Database supervision
• Knowledge and experience with the type of network the company has
• Knowledge and experience with your office applications
• Knowledge and experience with network software installation
• Knowledge and experience with hardware installation and upgrades
• Ability to teach and train about network use and software skills
• Knowledge and experience with network email
• Dedication to appropriate uses of computer technology (the ability to use and adapt older equipment)
Some of the day-to-day support tasks a network administrator should perform include:
• Doing backups of computers, and of the server(s)
• Adding and deleting new network users
• Making sure that virus protection software is up to date, and clean any viruses that infect staff computers
• Training and assisting staff in using their hardware and software efficiently
• Troubleshooting any routine problems that staff cannot fix on their own, such as a problem with the printer or a computer that freezes repeatedly
A good network administrator keeps things working smoothly day to day, but also tries to grow with the job and improve the way the organization uses technology. For instance, a network administrator might take an interest in learning all about the organizational database and developing better ways to automate and integrate it into the work of the organization.
Bear in mind that a network administrator cannot solve all of your technical problems. It is normal to bring in consultants to do more specialized and complicated service and support. Think of it as the difference between a general contractor and a specialty contractor.
A Day In The Life Of A Network Administrator
There hardly ever is an average day in the life of a network admin. When I arrive at work, the first task for me is to check the servers to insure they are operating properly. I have found that one of the quickest ways to do this is to simply log on to the network. Once logged on, I check for all my network resources such as access to the internet, access to email and access to my drive mappings. The login script is a text file created by the admin is used to give the user mappings specific to their rights level. Drive mappings are “pointer” to specific folders or files on the network. By simply logging on to the network, I can check the basic operation of my main production devices: the file server, email server, the firewall, the router and the Domain Controller. This is not a complete check, but it my reveal problems that other users will face once they log on.
After logging on, I go to the server room to check each server. My computer mentor taught me that whenever you enter the server room, have your eyes, hears and even nose open to anything that is different. On one occasion I entered the server room and heard my email server constantly processing emails. After further inspection I found that my server had been infected by the “Nimda” (Admin spelled backwards) virus. This was a fast-spreading Internet worm and file infector in pure and original form. It arrived as an embedded attachment, README.EXE file, in an email that had an empty message body and, usually, an empty subject field. Unlike most virus, it did not require the email receiver to open the attachment for it to execute. It used a known vulnerability in Internet Explorer-based email clients to execute the file attachment automatically. It had four modes of spreading: via email, via network shared drives, via unpatched IIS (Internet Information Server) servers and via file infection. My IIS server had not been properly patched against this virus. Fortunately, the vendor of my virus protection software quickly provided an application to clear the server of the virus. That was a “sad face” day.
Each server has a log file that tracks application events, system events and security events. It is very important to check these logs daily because they can be very helpful as a preventive maintenance tool. Errors seen in these logs should be investigated to see if they are symptoms of a larger problem looming in the background.
As I check the logs I also check the status of the previous nights backup. Most backup software has a separate log to track the performance of the backup job. The backup should have completed without any errors. Any error should be invested further. I have learned that a backup job that states that it completes successfully may not be useful in time of need. Kind of like having a new spare tire that has a leak in it. On one occasion, the email server needed to be rebuilt from backup tape due to database corruption. None of my backup tapes were any good because they had been backing up corrupt data. That was a definite “sad face” day. Another thing my mentor taught me, but I hadn’t incorporated was a means to test the validity of the backups. Since that incident, I have built a “Test Network” where I can take a backup tape from a production server and perform a restore on the test network just as I would in the event of an emergency. I also eject the previous night’s tape and insert a tape for the current day. Most administrators have a tape rotation schedule that should adequately cover most foreseeable problems.
Normally by the time I’ve finished checking the servers, the users have started arriving and logging onto the network. Useless there is a project that I’m working on that requires me to stay in the server room, I usually return to my office and start return any email or phone calls that my have come in since the previous workday. I also return to my office because being visible to the users is an important part of being a network administrator. The users should be able to feel that is around in case they are in need of assistance. Visibility can also serve as a deterrent to the users doing anything that would detrimental to their workstations. A average day may include having to go retrieve from a backup tape a file that a user accidentally deleted the day before. I would then go back to the server room, find the tape with the most recent version of the deleted file, insert it, and restore the file. I have found that when a user does something of this nature, it’s good to provide some “negative reinforcement” so that they will be less likely to do it again. This can be done making the user wait and sweat it out for a while. Don’t make the restore process seem too easy or the user will take it for granted.
Another average task of an administrator is creating new user accounts. This is done by using which ever utility that comes with the network operating system. For an NT network, it is User Manager for Domains. For a Novell network the utility is called Network Administrator. Either utility will allow the admin to create templates that can be used for creating most users. The administrator can create groups with the utilities that can used to further control access to network resources. To create accounts using User Manager for Domains, you must first log into then admin workstation as the administrator. The admin workstation is a computer in the domain that has the various network utilities installed. After logging into the admin workstation, click the “Start” button, “Programs”, “Administrative Tools” and then click “User Manager for Domains.” On the “User” menus, click “New User”. I’ve learned that is it best to create a user called “template” when you initially set up the network that has generally setting you would want all users to have i.e a home directory on the network. After selecting “New User”, enter the appropriate information in the spaces provided, i.e. Username, Full Name, Description, Password and Confirm Password. You then select the appropriate password options, and then click “Add”. The “New User” dialog box reappears and is cleared so that you can add another user. Finally, click “Close” to return to the User Manager window.
If it is necessary to create a home folder for a user account, double click on a previously created user. The “User Properties” dialog box appears. Click “Profile” and the User Environment Profile dialog box appears. Under “Home Directory”, click “Connect”. In the To box, type \computer_nameusers\%username% (where computer_name is the name of the server where the home folder will be stored. Click OK to return to the User Properties dialog box, then OK to return to the User Manager Window.
There are many other configuration settings you can chose for each user but the best policy is to be consistent when creating users and only make special selection when the need arises. This will help in the troubling of login problems later.
A network administrator will also need to setup new workstations from time to time. I have learned that the best approach to this is to purchase the same model workstation (Vectra Vli8 for example) from the same manufacture, (Hewlett Packard for example) to ensure that the hardware is consistent. This consistency helps in predicting possible hardware failure and software compatibility. The administrator should have a check list of things that must be performed on a new workstation before it reaches the end user. This list would include such things as specific applications, desktop settings, etc. The goal is to make the transition from old workstation to new workstation nearly transparent to the end user. If the old workstation can be put back in service elsewhere, the administrator would possibly format the drives (effectively erasing all the previous data) and reassign the workstation in a position that is less demanding than the previous one. I have found that installation of new workstations not only makes the end user more productive, but generally the task of administering the workstation is easier because of the improvement in technology i.e. newer operating system. Some administrators have found it useful to sue a procedure called “ghosting” to speed up the process of setting up multiple new workstations. When this process, a workstation is configured with general setting sufficient for most users. Special software is used to “copy” an image of that configuration. When a new workstation or workstations is needed to be configured the imaged is then copied to the hard drive. For this procedure to work, the workstation must have the same or very similar hardware as the original workstation the image was copies from. Otherwise there will be multiple errors and the administrator will spend time more correcting errors than he/she would have doing the configuration manually.
This paper is the property of getfreeessays.com Copyright © 2002-2003
