Article View: comp.sys.ibm.pc.misc
Article #1300Re: What is the best anti-virus program???
From: Bleys Ahrens
Date: Sat, 29 Aug 1992 03:17
Date: Sat, 29 Aug 1992 03:17
98 lines
5997 bytes
5997 bytes
All Anti-Virus programs are not created equal. Most anti-virus packages on the market today (Norton, Central Point, Untouchable, MacAffee) rely on pattern scanning as their primary method of detecting a virus. In other words, they have a database of some sort containing hundreds of strings or attributes that known viruses produce when they infect a computer. This has worked relatively well in the past, but unfortunately it is rapidly becoming unpractical. When new viruses and strains are identified, the appropriate patterns must be added to the databases of pattern scanning programs. This may be done with updates to the program on disk, via download or manually typing strings into the database. The problem is that an average of three new viruses are being created everyday. Add to this, the number of new mutating viruses which change their appearance and method of operation as they spread and it quickly becomes impossible to keep up with the spread of viruses. As more and more users become interconnected though networks like the Internet, Compuserve, Fidonet, etc. it becomes increasing easy to move a virus from one side of the world to the other. Another relatively new and dangerous sort of virus is the stealth viruses. These viruses intercept calls from anti-virus programs as they scan memory and disks. Boot sector stealth viruses generally make a copy of the uninfected original areas and direct scanning programs to the location of the copy, which appears to be normal. Stealth file viruses on the other hand are often able to remove themselves from a file as it is being scanned, so that the file appears clean. After the scan is complete, the virus in memory then moves its code back into the infected files. Other recent threats to computer security include the publishing of a book that tells how to write viruses and includes source code. (I would prefer not to publicize it and intentionally left out the title.) Another new threat is the recent posting on various BBSes of a some programs that help people write mutating viruses. In fairness to the makers of various anti-virus software manufacturers, it must be pointed out that most programs on the market include methods other than pattern scanning. Most also include TSR's which monitor for suspicious disk and memory reads and writes. Some also include various CRC checking methods which run some sort of algorithm on each file to create a checksum which is then compared each time the file is executed. A simple checksum works relatively well if you are absolutely sure that your files aren't already infected. The problem is that some viruses are smart enough to change the checksum values stored on the disk. So what is the optimal solution... Well, I have been evaluating several programs lately (Norton Anti-Virus, Central Point Anti-Virus, Fifth Generation Untouchable, Certus Novi, Intel LanProtect, Brightworks Development Sitelock and MacAffee Scan) and the one that seems have the best potential to deal with the possibility of stealth and mutating viruses is the relative newcomer, Novi from Certus. This product does do some pattern scanning, but that is not the basic paradigm used by this product. As I understand from my research, all executable files have essentially the same type of header records. This contains the basic information about the program and points to the beginning of the code to run. What a virus will do is attach to the end of the file and change the initial pointer to the start of the virus code. At the end of the virus code is a pointer to the start of the program code. Thus, the user runs the program, the virus executes and most likely places itself into memory and then executes the program. Novi works by checking the critical areas of the file (or disk boot records for a boot sector scan) to see that everthing is the way it should be. If pointers appear invalid or out of place, the product alerts the user to a problem and then depending on the user response, can attempt to remove the improper pointers and code. By not relying on pattern scanning, the program will working with existing viruses as well as new mutations or strains. I am still testing with it and others, but my opinion is that this program offers the best protection into the future, without the need of updates. (This program has also rated very, very well in a variety of recent magazine article and evaluations. It's file integrity checking is very fast and is compatible with the major LAN OSes and with Windows.) I suspect this article will provoke quite a few responses, and I hope that some of the above anti-virus manufacturers with be available for comments. I'm sure the folks from Certus can tell you a great deal more about their product and would be more than happy to send you some literature. (I'm not here to sell products for them, so I would prefer not to answer a zillion questions about exactly how the product works.) Bleys Ahrens Disclaimer: I am not affiliated in any way with any of the above mentioned products and/or companies. I am an IS professional with a large international corporation. While the above evaluations occured on company time, the opinions and views are strictly my own. Comments generally welcomed. Direct flames to /dev/nul... -- bleys@infopls.chi.il.us (Bleys Ahrens) Infoplus BBS, +1 708 537 0247, v32bis. Home of Infoplus.
Technical Details
Message-ID:
<ec19PB3w165w@infopls.chi.il.us>Path:
arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!nucsrl!ddsw1!infopls!bleysReferences:
<1992Aug26.142321.1@gallua.gallaudet.edu>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!nucsrl!ddsw1!infopls!bleys From: bleys@infopls.chi.il.us (Bleys Ahrens) Newsgroups: comp.sys.ibm.pc.misc,comp.os.msdos.apps,comp.binaries.ibm.pc.d Subject: Re: What is the best anti-virus program??? Message-ID: <ec19PB3w165w@infopls.chi.il.us> Date: Fri, 28 Aug 92 22:17:49 CDT References: <1992Aug26.142321.1@gallua.gallaudet.edu> Organization: INFOPLUS support, Wheeling, IL Lines: 98