 WebProWorld Dev Forum |
Beginner PHP & SQL Tutorial Needed
I want to add a database to my website and am trying to find starter info on learning SQL & PHP. I have found several sites, however they are still over my head.
Learning Servers
I was wondering where online I can learn how to built and install my own server? Are there different type of servers? If yes, are there online forums and such where I could learn each one?
MS Access "Numeric Field Overflow" Error
My computer has plenty of power and it is a simple database so I do not know how rebooting is helping but it works every time. I moved this database to another computer and the same error occurs.
|
 |
| Recent Articles |
Securely Erasing Data
If you're planning on recycling, donating, or even scrapping your old PC, it's important to ensure that you've securely erased all data from your hard drives. Too
BMC's $33 Million Identity Management
In an attempt to be more competitive in the small and medium-sized business market, BMC Software has agreed to buy Calendra for $33 million.
Storage Costs Decrease
The ability to store data is one of the rudimentary functions of a computer. Numerous devices are used in order to keep copies of work and/or backups.
ProFTPd, wu-ftpd, and general ftp security
FTP in general has a long and sad history of security problems. If you need to run an ftp server, you need to keep careful track of vulnerabilites and exploits that may make for a very unhappy day.
Configuring Basic Cisco Router Security
Network security is a hot topic today, and will only increase in importance in the months and years ahead.
COLT's Switched Ethernet VPN Built upon Cisco's Platform
Cisco announced that COLT Telecom Group plc is delivering its new pan-European managed Switched Ethernet VPN service built upon Cisco's multiservice optical transport platform.
Layered Security
Layered Security is becoming an over-used buzzword for a very simple concept. Let's say you went to your local bank to cash a check.
|
| |
|
01.27.05
Troubleshooting: Preserve The Scene
 By A.P. Lawrence
You've seen it on television or in the movies: important evidence is lost at a crime scene because someone wasn't careful enough to preserve it.
A crowd of curiousity seekers destroys important footprints, a helpful housekeeper thoughtfully washes the bathroom sink, wiping away fingerprints. The police investigators show up and are angry at the carelessness of the patrol officer who should have protected the scene. A crime scene investigator always wants to preserve and protect evidence.
Obviously, computers provide evidence too. If information on a hard drive were part of a crime investigation, that hard drive would immediately be removed and kept safe for analysis. The forensic investigators would clone it, and work with the copy so that their work didn't accidentally destroy original evidence. The procedures for doing this are quite involved, and will include taking MD5 hashes of the original and the copy so that integrity can be maintained and insured.
In a way, any computer problem is a crime. Perhaps not in the legal sense, but it is misbehaviour - maybe by a person using the computer, or maybe by the computer itself. It may be petty, or quite serious. It can cost every bit as much or more than a real crime might. It only makes sense that you should approach any problem with an eye toward not destroying information you might need. Most computer problems aren't important enough to justify all that a professional forensic investigator would do (or if they are that important you probably don't have the luxury of time to do all that), but before you jump in to fix things, you should sit back for just a moment and think about whether there is any evidence that needs to be preserved, and what you might need to do - or more usually NOT do - to preserve it.
As an example of this, we'll take the case of a backup that did not contain a file that should have been on it. The situation was that Machine A was backed up to tape nightly, and that tape in turn was restored to Machine B every morning. The purpose of this was to have a reasonably current system that could be pressed into service in the event of sudden failure of the main machine. Nowadays, most companies with a similar need would employ some form of clustering and redundant storage instead, but this backup-restore scheme is inexpensive and does meet minimal needs for some situations.
 Any redundancy plan is pointless if it isn't tested. It's not enough to just restore data to the standby machine: you have to verify that the system is capable of doing the work it may have to do. Toward that end, this company had a procedure where the person responsible for restoring the tape would also login to the standby and run a particular report. The report chosen was one that would access many files, so if the correct output were produced, it could be reasonably assumed that data was properly restored and the system was capable of use in an emergency. This was just a final check: the original backup would have also been verified while still on Machine A, and of course the restoration to B would also report any errors encountered. This report was just another assurance that all was well.
But one morning things were not well. The backup on the live machine had completed sucessfully, and had even read back everything written to tape and verified that each file matched what was stored on the hard drive. The restoration on the standby reported no errors reading the tape there. The standby hard drive didn't complain of any problems writing the data. But the report didn't produce the right results. More specifically, it seemed to be getting part of its data from an older file. Indeed, the specific file was easily identified, and it could be seen the version on Machine B was older than expected. This was not a file that gets modified or even used at all every day, but it was still important data. Fortunately, this problem was discovered by the integrity test rather than by real need, but the next question obviously was "How do we fix this?"
I bet you have answers, or at least guesses, already. This shouldn't be a very difficult problem: either the file is on the tape and it did not restore, or it was not on the tape at all. In the first case we look for the problem on the standby box, and in the second we go look at the backup procedures on the live box. Simple stuff, this should be done in minutes, right?
Sure. It should be simple, and the overwhelming likelihood is that it is a quick and easy diagnosis and fix. We can probably knock this one off before our coffee gets cold. Piece of cake, no sweat, why do they even bother me with such trivial stuff?
But what if it isn't? What if this somehow turns into one of those nasty little problems where you keep muttering things like "but it can't.." and your brain is bouncing off two apparently contradictory observations? What then? Your coffee is getting cold, your boss wants to know why this isn't fixed yet, and you have absolutely no idea. Why do they always give you the hard ones?
So, before "piece of cake" turns into "Huh?", let's think about what we can do to preserve evidence we might need if this thing does turn nasty on us.
Read the Rest of the Article.
*Originally published at APLawrence.com
About the Author:
A.P. Lawrence provides SCO Unix and Linux consulting services http://www.pcunix.com |
