Best Practices – Case Study #1

Let me use on of my long term clients as an example of how we make sure that his office system is protected. This client is a small law firm with 3 attorneys and 5 staff members. The firm has 10 computers and a Novell file server. One of the attorneys lives in another state part of the time and logs in remotely from a laptop.

The law firm has a proprietary case management database that by itself (data only) is just under 100 MB. The firm also has another program that is used for filing bankruptcy claims what has about another 100 MB of data. Also, the groupware program that the office uses for its calendaring and internal email has another 325MB data.

Several years ago the firm started scanning all incoming and outgoing correspondence. At this point, anyone who needs to access a client file can do so via the network. The attorneys and staff no longer have to track down a physical file. This data has grown to almost 40 GB. The scanned data grows every day by about 1 MB.

The system has grown over the years and we have always tried to keep it simple, and also, we have not replaced things that did not need replacing. The Novell server was put into service in early 1993 and other than replacing a hard drive a few months ago, has never upgrade or for that matter, turned off.

When the firm started scanning documents, a new computer was purchased that was used only to store the scanned documents and run the scanning software. This system has a 100 GB hard drive in it. Originally the system had Windows 98 though that was replaced several years ago with Windows 2000. Instead of upgrading the existing OS, a second hard drive was installed with the newer OS and the old drive with all the scanned data was made a secondary drive.

As you might imagine, the data is a bit scattered around. In addition to the main shared data, each computer has it’s own data that needs to be backed up. Internet email is stored by Outlook Express in a hidden folder on each machine. Most of the users also save documents in the local “my documents” directory.

Ok, it’s a mess and a gigantic pain to make sure that everything gets backed up. Trust me, it does and we’ve used our backups many times to save someone’s bacon.

We have setup a machine on the network that works primarily as a backup system. This machine has 2 hard drives installed in the system and a third hard drive that is in a removable tray. We have two drives that we rotate into the removable system.

The primary data that we are concerned with is on the Novell server. This is the case management system, the bankruptcy system and the groupware system. This also is our smallest amount of data.

Due to the nature of the programs and the data, we make a complete backup of this data to a separate directory for everyday. These programs grow each day and over the years we have experienced corruption in our data files. Experience has taught us that sometimes we have to go back a few days to find a backup that has not been corrupted. Data corruption can occur and unless you try to access data in the corrupted area, you may not learn about the corruption right away.

On a daily basis, this same data is backed up to our removable drive. Just to be safe, we also backup the case management data to a USB flash drive, we have two or three of them, which we rotate and take off site.

The data on each local drive is backed up to one of the hard drives in our backup machine. This data is then backed up nightly to the removable hard drive.

The scanned data is also backed up nightly, both to one of the internal hard drives and then again to the removable drive. The hard drive in the server, which is only 4 GB in size is then backed up in it’s entirety to one of the internal hard drives.

To add to the confusion, we also have several images of the server backed up in different locations. At the present time, we do not have images of each local machine. We are in the process of upgrading up to 5 of the machines and as soon as they are configured, an image will be created. And we will take a copy of each image offsite.

Does this sound like it is overkill to you? Sometimes I think it is, and if I were starting from scratch, I would certainly do things a bit differently. I do like the fact that we have data backed up in more than one place. This has saved us several times over the years. I do like the fact that we take the data offsite. The primary database is taken offsite everyday. The rest of the data is take offsite at least once a week, usually a bit more often.

Using the removable drive tray allows us to quickly swap out the offsite storage drives. We recently upgrade the trays so that we don’t even have to turn the systems off.

Our next step is to set up a hard drive in a remote location onto which we copy the data that we take offsite. We are still trying to work out the details on this one. We tried using a USB drive that we could store somewhere a bit farther away from the main office, but someone, OK it was me, dropped it and it crashed.

In the next post I will tell you about another client, this one was a bit more paranoid about losing his data.