It may seem like there are not enough hours in a day to get everything done. Some things never make it to the top of the “To Do” list. An area that is normally taken for granted is your network, server, and computers. It is easy to assume that everything is OK because you are not having any problems…well, not right now anyway. You could also drive your car for 40,000 miles without changing the oil and probably not have any engine problems. But, would you buy that car based on its reliability and the expectation that it has a lot of miles left in it?
INSTALL WINDOWS UPDATES AND SERVICE PACKS
Microsoft releases fixes, updates, and patches (FUPs) on a regular basis. Currently, Microsoft is releasing these updates on the second Tuesday of each month. This could change by the time you read this article. The service packs are released with less frequency, about once a year for each operating system. (Windows XP and Windows 98 are operating systems.) You can use Windows Update to download and install the latest FUPs. These are changes to the operating system and should be installed with caution. Many larger corporations will test the updates in their labs before they install them on their PCs. I realize that you probably don’t have the staff or test lab, so I recommend that you have your computer support company install the FUPs for you.
CHECK THE BACKUP LOGS AND RUN A TEST RESTORE
I know you are backing up your data every day, right? But are you checking to make sure the backup process is working correctly? Your backup software should create a log file for each backup or add information to a log file each time a backup is run. In some cases, the software may also include a section that shows a summary of each job and whether or not it was completed successfully. You should check the backup log or summary area every day to make sure the backup is working correctly. Get in the habit of checking when you change the media (tape, CD, external hard drive, etc). But (you know there is always a “gotcha” with this tech stuff), you should also run a test restore at least once a month to make sure you can read the data off the media you use for backup. Unfortunately, I do have people tell me that they never checked the logs and/or never tested the backup and lost all of their data. I am not recommending that you restore your entire server or even your application data. I choose a file that is not critical and rename the extension to .old. For example, I change the file manual.pdf to manual.old. I then restore this one file from the backup media to the hard drive.
CHECK THE SELECTIONS FOR THE BACKUP JOB
As part of the setup and configuration of your backup software, you need to select the files and folders that are part of the backup job list. You should check this list at least once a month to make sure you are backing up all of your data. The selections may not be the entire C and D drive (or whatever letters your server has for drives). If you have selected to backup just a few of the folders on the D drive and then install a new software program on D, it may not be added to the backup list. Don’t find out the hard way that your backup was working fine, but it was not backing up all of your data. I see this a lot with certain accounting packages. Such programs will put the data file in the same folder as the application files, under Program Files.
If you check the selections at the same time you are running the test restore, then you have saved yourself an extra trip to the server. Please take the time to look at your folders and files through My Computer or Windows Explorer and not just through the selections list. The list of available files and folders in the selections area may not be refreshing correctly, and you won’t see the new folders.
CHECK THE SERVER EVENT LOGS
If you are running a Windows server, then you have event logs that need to be checked on a monthly basis. You will usually find the logs under the System Administration, Event Viewer. The 2 most common logs to check are the Application and System logs. Usually the events with the blue icon are informational and nothing to worry about. The events with the yellow icon are warnings that should be investigated further. The events with the red icon indicate an issue that needs to be addressed immediately. Here’s the “gotcha”… some red icon events are not critical and can be ignored. If you have driven in Massachusetts, then you understand that red does not always mean stop.
CHECK THE RAID AND DISK HEALTH
Your server may have multiple hard drives in a hardware or software RAID configuration. This means that you are using more than one hard drive to store your data so that your server will survive a single hard drive failure with very little down time. If you are using a hardware RAID configuration, it may be possible to configure it so it sounds an alarm if one of the hard drives fails. There may also be lights on the front of the drives that will indicate a problem. But, the safest way to be sure that the RAID is working correctly is to check it through the RAID configuration software. This will show you the status of the RAID configuration as well the status of each drive.
CHECK THE UPS LOG AND RUN A TEST
This does not mean check to see what you have shipped via UPS. The UPS in this case is the Uninterruptible Power Supply that you have your server and maybe PCs plugged into. This is a battery backup that will keep the server up and running for a short amount of time if you suffer a loss of power. If the UPS is configured to communicate with the server, then it will perform a clean shutdown of the server if the power outage lasts for a predetermined amount of time. The UPS software may include a log file that you can check once a month to see how often the UPS has kicked in to provide power to the server. The log may also show you issues that are not obvious.
CHECK THE FIREWALL LOGS AND SETTINGS
If you have a firewall or cable/DSL router on your network, then it may keep a log of all activity. You should check this log once a month to ensure there is nothing suspicious or obvious going on. One thing that you will notice if you are checking the log around the same time that a worm (like Blaster or Sasser) has been released, is that there will be a lot of activity in the log. The log may not identify the worm, but it should tell you the port number that will also be referenced in the descriptions of the worm. Most, if not all, of the log entries will be harmless, but take the time to check it every month.
CHECK THE ANTIVIRUS SOFTWARE FOR UPDATES, AUTOPROTECT TURNED ON, AND VIRUSES FOUND
You should check your antivirus software on a monthly basis to make sure it is being updated regularly, the auto protection is turned on, and if any viruses have been found recently. If you have a server or more than a few PCs, then you should be using a corporate (or business) edition of the antivirus software. This edition will allow you to manage all of the settings and status from one PC. This means that if you install the management console software on your PC, you can change settings, check the status, and perform other tasks on all of the computers from the comfort of your office (and the server). This does not mean that you install the software that does all of the processing on your PC, just the management console to change the settings.
CHECK WORKSTATIONS FOR UNAUTHORIZED SOFTWARE
This may be too much like Big Brother for some people, but rogue software programs can cause a lot of problems. The rogue software can be an illegal copy of valid, harmless software. I can assure you that the fine for illegal software will be issued in your practice’s name, not in the name of the employee who installed the software. There could be other software installed that causes performance problems with your network or opens your network up to viruses, adware, and spyware. If you rely on looking over your staff’s shoulder when you walk by to see if they are using unauthorized software, then you are not really protecting your network. You would not believe how fast the application can be closed as you approach. The fastest click in the West will occur at the sound of your voice or them catching a glimpse of you coming down the hall.
RUN A DISK DEFRAGMENT, CLEAN OUT THE PCS’ CASE, AND OTHER TASKS
This last set of tasks do not need to be run on a monthly schedule. They should be on a 6-month or 1-year schedule. How often you perform them is entirely based on your environment. If you have the PCs mounted up off the floor in a tiled room, then you may not have to clean out the inside more than once every 2 or 3 years. But, if the computer sits on the floor in a high traffic, carpeted area, then you may want to clean it out every 6 months.
CONCLUSION
I hope that this list is a good start for your network task list. Please understand that these tasks can cause more harm than good if they are performed incorrectly. You should check your ego at the door if you don’t have the technical knowledge or experience to perform the tasks with confidence. The task list is designed to help your network run smoothly, not to create new headaches for you and your staff. Give the list of tasks to your tech and go through it together to see which ones apply to your network.
Mr. Walsh is the owner of NDM Networks in Marlborough, Mass. NDM Networks provides computer and networking services to dental practices and small businesses in Massachusetts and Rhode Island. He can be contacted at (508) 624-9898 or dwalsh@ndmnetworks.com.
