(1) How do I run WebLog from my browser?

You don’t. WebLog is intended to be run directly from the UNIX command prompt or via the “cron” command. Its output is an HTML file, which can be viewed through your browser.

(2) Where do I find the access log that WebLog uses as input?

I’ve no idea. You’ll need to talk to your system administrators. Most (but not all) systems provide their users with access logs, but there’s no way for me to tell what they’re named or where they’re located on your system.

(3) Does WebLog handle compressed log files?

Yes, it can automatically uncompress log files using the “gunzip” command. In order to ensure that the program correctly uncompresses and processes the log file, though, make sure that you specify the log file name without the extension in your configuration file! (In other words, for example, if your log file is “dailylog.gz” you should define the $LogFile variable as “dailylog” rather than as “dailylog.gz.”)

(4) The name of my log file is different each day. It’s always in the format, “www.YYMMDD.” How am I supposed to define $LogFile?

As of WebLog version 2.00, you can use an asterisk in the variable definition as a “wildcard.” For example, if your log files come to you named “www.YYMMDD” you can simply define $LogFile as “www.*”. This will tell WebLog to analyze any file whose name matches the pattern. This can also be useful if you have several log reports backed up, and want to run WebLog on all of them at once.

(5) My log files contain only IP numbers rather than fully-qualified domain names. How to I find out what domains are visiting my site?

As of version 2.00, WebLog can resolve IP numbers to domain names for you.

(6) What if I have multiple days’ files accumulated when I try to run WebLog, or the log file contains data from several days?

The latter situation presents no problem whatsoever. As to the former, well, see the answer to question #4.

(7) I keep getting “Out of Memory” errors when I try to run WebLog. What can I do?

If your site is very busy, you probably won’t be able to keep a full list of domains that have visited. The sorting of that list is a major memory hog; turning it off will probably solve your problem. (As an intermediate step, you can try falling back from a full list to a second-level list.) Please note that even with the domains list turned off, you will still get the “top level” tracking, so you’ll still know how many visits you got from each “type” (.com, .net, .de, .au, etc.) of domain.

If that doesn’t help, your log files might just be too large. Try breaking them down into smaller chunks.