Access Stats 1.12 Problem

Hello all together!

This script doesn't work, but I don't know why.

My files are in

/logstat/gif
/logstat/logfiles
/logstat/log.gif
...

log.pl, stat.pl and ip.log are in

/cgi-bin/logstat/

My log.pl:

Ma stat.pl:

# url to script
$cgiurl = "http://www.web82.server-drome.net/cgi-bin/logstat/stat.pl";

# url to the image files without trailing "/"
$gif_url = "/logstat/gif";

# path to the log files from the server root without trailing "/" (chmoded to 777 (drwxrwxrwx))
$base_dir = "/srv/www/htdocs/web82/html/logstat/logfiles";

# url to the log files without trailing "/"
$log_url = "/logstat/logfiles";

# minimum visits before to show in list
$host_min = 1;
$browser_min = 1;
$os_min = 1;
$referer_min = 1;

# show visits from all countries
$show_ccodes = "yes";
$show_max = 15; # only valid if $show_ccodes="no"

# log file extension
$log_file_ext = "txt";

# misc configurations
$font_size = "8pt"; # font size
$table_width = 700; # table width
$max_bar_length = 300; # bar lenght of host,os,browser,etc. in relation to $table_width
$max_bar_day_length = 170; # bar lenght of day should be set in relation to $table_width
$hits_by_weekday = "yes";
$top_countries =   "yes";
$top_browsers =    "yes";
$top_os =          "yes";
$top_host =        "yes";
# Done
###########

There are no error messages, but it doesn't log anything.

The code in my html-file:

May I delete these lines:

So, it's not HTML 4.01 transistonal.

Where is my mistake? I want use relavive paths, if it's possible.

Okay, first thing I see wrong is that the base_dir in both Log.pl and Stat.pl are not the same. Both of them need to be the full path.
/srv/www/htdocs/web82/html/logstat/logfiles

The other thing may be that you are trying to use relative paths - some times that works, some times it doesn't. At lot depends on how the cgi-bin has been set up on the server or by the webhost.

On the last javascript portion, you may take out the noscript portion, however it will not count people who do not have their javascript turned on when they arrive at your pages. The noscript portion forces, at the very least, a visit to be counted in the log as a "noscript" on your print out.

These are the things that stood out - not guaranteeing it will solve the problems.

Hoping that helps

Make sure you use the complete url in the url variables. and notand so forth for the rest of the url stuff.

Great! Thank you very much, it works @ET!

@Carbonize: Thank you for your answer, too, but it works now (with relative paths).

One more thing. This script sais that I'm using the Internet Explorer, but I'm using firefox.

Sorry, as I watched the statistics, another user was on my page an this user has git the IE.

Thank you.

I wouldn't read to much into the browsers it says visitors use. Opera identifies it's self as Internet Explorer by default and it is easy to alter the user agent sent by Mozilla based web browsers/

Does this mean, it's impossible to get the real browser names?

Not impossible - just takes work to write the script. See my other response to your other thread you started.

http://proxy2.de/forum/viewtopic.php?t=4131

Impossible, no. But unless you are going to start checking for DOM models using javascript then no you cannot.