hi,
First of all I want to convey my thanks, for such a project which helps very much.

While generating a report for a specific month, Iam getting the following error.
perl awebvisit.pl access.*May*
Reading logfile entries...
Can't open access.*May*: No such file or directory
--> 0
Read 0 entries from  to :::

No valid logfile entries were recognised...
Make sure the logfile is in Common Logfile Format or Extended Logfile Format.
If not, you can adapt aWebVisit to your own logfile format (and send me a copy).

And my log file is having the following format.
xx.xx.xx.xxx- - [25/May/2001:17:27:51 -0400] "GET /xxxx/xx.pl HTTP/1.1" 200 128 
Can you give any suggestions which are valuable to me and it will be a great help.
Iam able to generate whole logfile, without any problems. And also Iwant to know how to
configure fly program. Because Iam getting the following after generating.
perl awebvisit.pl access
Reading logfile entries...
Line 5000
--> 0
Read 7246 entries from 25/May/2001:17:27:51 to 04/Sep/2001:10:52:17
Extracting page and link information for 332 sessions...
Found 329 sessions containing at least one page
Calculating page statistics...
Calculating time statistics...
Calculating link statistics...
Generating output files...
f i c h s e 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 
Saving data to file './awebvisit.data'...
Saving statistics to file './awebvisit.stat'...
The aWebVisit reports can now be viewed at './awebvisitf.html'
The 'fly' program is not configured, so there are no web maps available...
I did make in fly-1.6.5.
Iam running all these in HP-UX environment with netscape iplanet is the web server.
Please help me regarding this.
Thank you very much in advance.
bye
krishna

For your first problem, I can only speculate that there are no files matching access.*May*
in the current directory. Can you double-check that ?

In order to use 'fly', you obviously need to install it first, and then you need to tell
aWebVisit where it can find the program.
You do that by adapting the $flyprog variable in the configuration of aWebVisit :

#=======
# 5.a.	Location of the FLY program for graphics - leave blank if not there
#	You can get pre-compiled binaries for Windows 95 & NT, various UNIXes,
#	and other platforms at http://www.unimelb.edu.au/fly/
#=======

$flyprog = '';
#$flyprog = '/usr/local/bin/fly';

hi Mike,
Thanks for your kind reply. And one more thing, I need only the average time spent. Is
there awebvisit.data & awebvist.stat can tell us the total time spent in that URL.
Thatswhat I want. Can you please brief about the total time spent things.
Thanks
krishna

As I mention in "What do the numbers mean ?" at http://mikespub.net/tools/aWebVisit/ the
"average time spent" can only give you a rough *idea* of how long people actually stay on
one page.
It works like this : while the script is walking through the logfiles, it builds up a
'session' for each "visit" that it can identify. Once a session is finished, it calculates
the time difference between each "page" and the next one in that session, and adds that to
the "total time spent" for that page.
That means that only 'Entry' pages and 'Transit' pages will have a time value associated
with them - for 'Exit' pages and 'Hit&Run' pages, there simply *is* no next "page", so we
don't have a time value for those. (it also means that for instance if you have a web page
that also loads a .wav file, and you didn't exclude .wav from what the script identifies
as "pages", you'll probably have a very short "time spent" on the web page itself, and a
more realistic "time spent" on the .wav file...)

OK, so at the end of the logfiles, we have the "total time spent" on each "page" (as far
as we can tell from the logfiles), and we know the number of times that "page" was an
'Entry' or a 'Transit' page. The "average time spent" for each page is then simply :
"average time spent" = "total time spent" / (# of times the page was an entry page or a
transit page)

Both the .stat file and the .data file contain that "average time spent" on pages.
The .stat file only contains the Top N Pages/Entries/Transits/Exits/Hit&Runs, but you can
find the average time spent for each of those pages in the last column.
The .data file is more complete and gives you not only the "average time spent", but also
the # of times each page was an Entry/Transit/Exit/Hit&Run page, so you can recalculate
the "total time spent" for each page if you want to. In fact, in later test versions, I
adapted the .data format to explicitly save the time count and total time spent as well,
so that I could more easily load .data files from previous runs.

One last remark about the values of the "time spent" : since it's the time between two
"pages" in a session, you may get near-zero values for frames or for different elements of
a web page. You may also get near-timeout values for pages with few hits, when a visitor
(or someone else with the same IP) returns to your site nearly at the end of the timeout
period (that's the main reason why I started working on median values in a later test
version). And if you happen to use a webserver that can generate logfiles where the 'hits'
are "out of order" (e.g. IIS), you might even have negative "time spent". So use your
common sense in interpreting the figures...

There, I think I covered everything I could think of regarding "time spent". I wasn't
quite sure exactly what your question meant, so I hope you'll find your answer somewhere
above :)