Hello Simon!
By default, Webalizer's country identification comes from reverse DNS records that Web server wrotes to logs. Every time client connects Apache tries to obtain it's hostname by IP address and writes hostname to logs. Webalizer simply extracts hostname suffix and resolves country name using special table.
The problem is: not all clients has hostname associated. Many ISP doesn't care about this. So Webalizer "gets lost".
To overcome this problem you can use software that has database to associate IP ranges directly to country. I've written a patch that integrates GeoIP library support into Webalizer: http://sysd.org/proj/log.php#glzr
On this address there's a patch to Webalizer source and static binary for Linux. Good luck!


----- Original Message -----
From: simon benedict <sbdcunha@...>
Date: Wed, 21 Apr 2004 09:26:40 -0700 (PDT)
To: webalizer@yahoogroups.com
Subject: [webalizer] usage by country

> Hi All
> 
> I am running webalizer on Linux Box and using the log files from IIS to generate the statistics.
> 
> but the usage by country shows me 87%  unresolved/unknown
> 
> 
> http://weblog.baladia.gov.kw/usage_200404.html
> 
> 
> i really dont know if it should be this way and how do i get rid of this so that my pie chart graph shows all countries
> 
> do i have to change any settings in my IIS or the webalizer.conf file
> 
> 
> thnks and appreacite
> 
> 
> regards
> 
> simon
>
>
>
>
> ---------------------------------
>
>
>
>
>
> > ATTACHMENT part 2 application/x-pkcs7-signature name=smime.p7s
>
>
>            
> ---------------------------------
> Do you Yahoo!?
Yahoo! Photos: High-quality 4x6 digital prints for 25˘
--
______________________________________________
Check out the latest SMS services @ http://www.linuxmail.org
This allows you to send and receive SMS through your mailbox.


Powered by Outblaze


Webalizer homepage: http://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/


ttp://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/

---

Yahoo! Groups Links

• To visit your group on the web, go to:
  http://groups.yahoo.com/group/webalizer/
   
• To unsubscribe from this group, send an email to:
  webalizer-unsubscribe@yahoogroups.com
   
• Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Hello:

If you Apache, you'll surely have an utility called "logresolve". It
parses the Apache logfiles and resolves all IPs it encounters there.

logresolve < access_log_not_resolved > access_log_resolved

It will parse "access_log_not_resolved" and will create
"access_log_resolved". Change names accordingly.

It may take more than one hour for 100 megas of logs, but you only
need to run it once.

You can convert all logs with only IPs on it to resolved logs, so you
can deactive the dns resolving in webalizer so it goes faster. It also
understand logs where some IPs have already been resolved and others
haven't.

You can desactive the HostnameLookup feature in Apache so it runs
faster, and then run logresolve on the rotated logs, so you have the
best of both worlds.

It seems to be able to understand Common and Combined logs from
servers others from Apache (I tried a caucho.resin log and it worked).

enventa2000 wrote:
> Hello:
>
> If you Apache, you'll surely have an utility called "logresolve". It
> parses the Apache logfiles and resolves all IPs it encounters there.

we need to be careful with statements like this... some of these utils are not
available on other operating systems that apache runs
on... in my case, IBM's OS/2 Warp 3 ;)

[trim]

> You can convert all logs with only IPs on it to resolved logs, so you
> can deactive the dns resolving in webalizer so it goes faster. It also
> understand logs where some IPs have already been resolved and others
> haven't.

even better would be to let apache resolve the addresses, wouldn't it? if apache
can't resolve the address, no other tool will be
able to resolve them either... in many cases, anyway...

> You can desactive the HostnameLookup feature in Apache so it runs
> faster, and then run logresolve on the rotated logs, so you have the
> best of both worlds.

but your system still spends the same time (or longer) when you resolve after
the fact, doesn't it?


--
         _\/
        (@@)                      Waldo Kitty, Waldo's Place USA
__ooO_( )_Ooo_____________________ telnet://bbs.wpusa.dynip.com
_|_____|_____|_____|_____|_____|_____ http://www.wpusa.dynip.com
____|_____|_____|_____|_____|_____|_____ ftp://ftp.wpusa.dynip.com
_|_Eat_SPAM_to_email_me!_YUM!__|_____|_____ wkitty42 -at- alltel.net

> > If you Apache, you'll surely have an utility called "logresolve".
[snip]
> we need to be careful with statements like this... some of these
utils are not available on other operating systems that apache runs
> on... in my case, IBM's OS/2 Warp 3 ;)

ops, you are right, I use RedHat 9.0, which is filled (or bloated)
with lot of utilities that I will never use.

Anyways, it is very useful when you have not resolved logfiles. I'm
sure there are similar utilities and even scripts, but I don't konw
them.

[snip]

> even better would be to let apache resolve the addresses, wouldn't
it? if apache can't resolve the address, no other tool will be
> able to resolve them either... in many cases, anyway...
>

Since logresolve is provided with Apache, they will both have the same
abilities.


> > You can desactive the HostnameLookup feature in Apache so it runs
> > faster, and then run logresolve on the rotated logs, so you have
the
> > best of both worlds.
>
> but your system still spends the same time (or longer) when you
resolve after the fact, doesn't it?


Yes, but Apache resolves the addresses while it is at the same time
atempting to serve the pages. Sites where Apache takes too much time
to serve the pages may attempt to desactive the feature. If anything
goes wrong, they can still resolve the non-resolved logs later.
tel.net

You are still executing the webalizer binary file. You need to execute
the geolizer binary file instead.

You need to execute the "webalizer" binary file inside the geolizer
directory (I assume you unpacked the files in this directory):

/usr/src/geolizer_2.01-10-static-linux/webalizer

instead of simply executing:

webalizer

The "Invalid Keyword" is because webalizer does not understand that
keyword. Geolizer would never give that error message, because those
keywords are completely valid for him.


Also, use the "-c" switch to tell geolizer where your webalizer.conf
file is (perhaps it is called "sample.conf"). For example:

/usr/src/geolizer_2.01-10-static-linux/webalizer -c /etc/webalizer.
conf


You will notice that geolizer will complain that DNSCache and
DNSChildren are invalid keywords. Comment them if you wish. Remember
that if you drop geolizer and return to use webalizer again then you
will need to uncomment those keywords again, or webalizer will not do
DNS resolving.


--- In webalizer@yahoogroups.com, simon benedict <sbdcunha@y...>
wrote:
>
> hi
>
> I have download unzipped the  geolizer_2.01-10-static-linux
>
> i have made the required change for the Logfile and the outdir
directive respectively and also the  directives
>
> TopCountries    300
> GeoIP           yes
> GeoIPDatabase   /usr/src/geolizer_2.01-10-static-linux/GeoIP.dat
>
> now when i run webalizer it gives me the output as below
>
> Warning: Invalid keyword 'GeoIP' (webalizer.conf)
> Warning: Invalid keyword 'GeoIPDatabase' (webalizer.conf)
> Webalizer V2.01-10 (Linux 2.4.18-14) English
> Using logfile /home/webuser/logs/ncsa1.log (clf)
> DNS Lookup (20): 29 addresses in 0.35 seconds
> Using DNS cache file dns_cache.db
> Creating output in /var/www/html/weblog
> Hostname for reports is 'kmdns'
> Reading history file... webalizer.hist
> Skipping bad record (648)
> Generating report for April 2004
> Generating summary report
> Saving history information...
> 648 records (1 bad) in 0.06 seconds
>
> its says GeoIP is invalid keyword
>
>
>
> and the output is http://weblog.baladia.gov.kw/usage_200404.html
>
> shows me 74 % unresolved/unknown
>
> since I have download the static version of Geolizer i just need the
Geo keywords specifiying the path to the GeoIP.dat file if I am right
and which I have done
>
>
>
> Appreciate if you can let me know
>
>
>
> thanks and Regards
>
>
>
> simon
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> ---------------------------------
> Do you Yahoo!?
> Win a $20,000 Career Makeover at Yahoo! HotJobs

You are still executing the webalizer binary file. You need to execute
the geolizer binary file instead.

You need to execute the "webalizer" binary file inside the geolizer
directory (I assume you unpacked the files in this directory):

/usr/src/geolizer_2.01-10-static-linux/webalizer

instead of simply executing:

webalizer

The "Invalid Keyword" is because webalizer does not understand that
keyword. Geolizer would never give that error message, because those
keywords are completely valid for him.


Also, use the "-c" switch to tell geolizer where your webalizer.conf
file is (perhaps it is called "sample.conf"). For example:

/usr/src/geolizer_2.01-10-static-linux/webalizer -c /etc/webalizer.
conf


You will notice that geolizer will complain that DNSCache and
DNSChildren are invalid keywords. Comment them if you wish. Remember
that if you drop geolizer and return to use webalizer again then you
will need to uncomment those keywords again, or webalizer will not do
DNS resolving.


--- In webalizer@yahoogroups.com, simon benedict <sbdcunha@y...>
wrote:
>
> hi
>
> I have download unzipped the  geolizer_2.01-10-static-linux
>
> i have made the required change for the Logfile and the outdir
directive respectively and also the  directives
>
> TopCountries    300
> GeoIP           yes
> GeoIPDatabase   /usr/src/geolizer_2.01-10-static-linux/GeoIP.dat
>
> now when i run webalizer it gives me the output as below
>
> Warning: Invalid keyword 'GeoIP' (webalizer.conf)
> Warning: Invalid keyword 'GeoIPDatabase' (webalizer.conf)
> Webalizer V2.01-10 (Linux 2.4.18-14) English
> Using logfile /home/webuser/logs/ncsa1.log (clf)
> DNS Lookup (20): 29 addresses in 0.35 seconds
> Using DNS cache file dns_cache.db
> Creating output in /var/www/html/weblog
> Hostname for reports is 'kmdns'
> Reading history file... webalizer.hist
> Skipping bad record (648)
> Generating report for April 2004
> Generating summary report
> Saving history information...
> 648 records (1 bad) in 0.06 seconds
>
> its says GeoIP is invalid keyword
>
> 
>
> and the output is http://weblog.baladia.gov.kw/usage_200404.html
>
> shows me 74 % unresolved/unknown
>
> since I have download the static version of Geolizer i just need the
Geo keywords specifiying the path to the GeoIP.dat file if I am right
and which I have done
>
> 
>
> Appreciate if you can let me know
>
> 
>
> thanks and Regards
>
> 
>
> simon
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
>
>            
> ---------------------------------
> Do you Yahoo!?
> Win a $20,000 Career Makeover at Yahoo! HotJobs



Webalizer homepage: http://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/


ttp://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/

it worked perfect.. suprising

You are still executing the webalizer binary file. You need to execute
the geolizer binary file instead.

You need to execute the "webalizer" binary file inside the geolizer
directory (I assume you unpacked the files in this directory):

/usr/src/geolizer_2.01-10-static-linux/webalizer

instead of simply executing:

webalizer

The "Invalid Keyword" is because webalizer does not understand that
keyword. Geolizer would never give that error message, because those
keywords are completely valid for him.


Also, use the "-c" switch to tell geolizer where your webalizer.conf
file is (perhaps it is called "sample.conf"). For example:

/usr/src/geolizer_2.01-10-static-linux/webalizer -c /etc/webalizer.
conf


You will notice that geolizer will complain that DNSCache and
DNSChildren are invalid keywords. Comment them if you wish. Remember
that if you drop geolizer and return to use webalizer again then you
will need to uncomment those keywords again, or webalizer will not do
DNS resolving.


--- In webalizer@yahoogroups.com, simon benedict <sbdcunha@y...>
wrote:
>
> hi
>
> I have download unzipped the  geolizer_2.01-10-static-linux
>
> i have made the required change for the Logfile and the outdir
directive respectively and also the  directives
>
> TopCountries    300
> GeoIP           yes
> GeoIPDatabase   /usr/src/geolizer_2.01-10-static-linux/GeoIP.dat
>
> now when i run webalizer it gives me the output as below
>
> Warning: Invalid keyword 'GeoIP' (webalizer.conf)
> Warning: Invalid keyword 'GeoIPDatabase' (webalizer.conf)
> Webalizer V2.01-10 (Linux 2.4.18-14) English
> Using logfile /home/webuser/logs/ncsa1.log (clf)
> DNS Lookup (20): 29 addresses in 0.35 seconds
> Using DNS cache file dns_cache.db
> Creating output in /var/www/html/weblog
> Hostname for reports is 'kmdns'
> Reading history file... webalizer.hist
> Skipping bad record (648)
> Generating report for April 2004
> Generating summary report
> Saving history information...
> 648 records (1 bad) in 0.06 seconds
>
> its says GeoIP is invalid keyword
>
> 
>
> and the output is http://weblog.baladia.gov.kw/usage_200404.html
>
> shows me 74 % unresolved/unknown
>
> since I have download the static version of Geolizer i just need the
Geo keywords specifiying the path to the GeoIP.dat file if I am right
and which I have done
>
> 
>
> Appreciate if you can let me know
>
> 
>
> thanks and Regards
>
> 
>
> simon
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
> 
>
>
>            
> ---------------------------------
> Do you Yahoo!?
> Win a $20,000 Career Makeover at Yahoo! HotJobs



Webalizer homepage: http://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/


ttp://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/

--- In webalizer@yahoogroups.com, simon benedict <sbdcunha@y...>
wrote:
> thnaks a lot for the quick reply
>
> btw I was executing the the right binary .  i was running geolizer
binary cause i was moving to /usr/src/geolizer_2.
01-10-static-linux/webalizer with cd command and then running
webalizer
> and also I had the sample.conf in the same directory

>
> but as per u when I started to run the below command from the root
>
> /usr/src/geolizer_2.01-10-static-linux/webalizer -c
/usr/src/geolizer_2.01-10-static-linux/sample.conf
>
> it worked perfect.. suprising


:) The joys of understanding what the PATH variable does....

execute this your command line:

echo $PATH

You will see the list of directories where linux (or unix) searches
for executable. It was finding the webalizer executable in one of
those directories. To execute files outside of that list of
directories you need to specify the whole path to the executable:

Specify the whole route:
/usr/src/geolizer_2.01-10-static-linux/webalizer

Move to the directory, and execute:
cd /usr/src/geolizer_2.01-10-static-linux/webalizer
./webalizer

Move to a directory, and specify a relative route:
cd /usr/src
./geolizer_2.01-10-static-linux/webalizer

This can be written also like this:

cd /usr/src
geolizer_2.01-10-static-linux/webalizer


Another relative route:
cd usr/src/apache
../geolizer_2.01-10-static-linux/webalizer



The "./" part means that the route to the file is the actual
directory. "../" is to move one directory up.

If you had not had webalizer installed, then linux would have
complained that it couldn't find the file, even if the file was in
your current directory. Linux (and unix) will never look at the
current directory to find an executable unless you tell them do so.
You could of course add "./" to PATH, but it is dangerous, because you
could be in the wrong directory, and run a different script than you
intended and break something.

--- In webalizer@yahoogroups.com, simon benedict <sbdcunha@y...>
wrote:
> thnaks a lot for the quick reply
> 
> btw I was executing the the right binary .  i was running geolizer
binary cause i was moving to /usr/src/geolizer_2.
01-10-static-linux/webalizer with cd command and then running
webalizer
> and also I had the sample.conf in the same directory

> 
> but as per u when I started to run the below command from the root
> 
> /usr/src/geolizer_2.01-10-static-linux/webalizer -c
/usr/src/geolizer_2.01-10-static-linux/sample.conf
> 
> it worked perfect.. suprising


:) The joys of understanding what the PATH variable does....

execute this your command line:

echo $PATH

You will see the list of directories where linux (or unix) searches
for executable. It was finding the webalizer executable in one of
those directories. To execute files outside of that list of
directories you need to specify the whole path to the executable:

Specify the whole route:
/usr/src/geolizer_2.01-10-static-linux/webalizer

Move to the directory, and execute:
cd /usr/src/geolizer_2.01-10-static-linux/webalizer
./webalizer

Move to a directory, and specify a relative route:
cd /usr/src
./geolizer_2.01-10-static-linux/webalizer

This can be written also like this:

cd /usr/src
geolizer_2.01-10-static-linux/webalizer


Another relative route:
cd usr/src/apache
../geolizer_2.01-10-static-linux/webalizer



The "./" part means that the route to the file is the actual
directory. "../" is to move one directory up.

If you had not had webalizer installed, then linux would have
complained that it couldn't find the file, even if the file was in
your current directory. Linux (and unix) will never look at the
current directory to find an executable unless you tell them do so.
You could of course add "./" to PATH, but it is dangerous, because you
could be in the wrong directory, and run a different script than you
intended and break something.




Webalizer homepage: http://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/


ttp://www.webalizer.org
Webalizer for NT: http://www.medasys-lille.com/webalizer/

What program do you use to read logs? I use wordpad, but it's not so good
with large logs.

Also it seems like all of you run your own server - my website is hosted by
a hosting company, so I have no control over Webalizer functions at all.

Anne
http://www.afan.dk

Anne wrote:
> What program do you use to read logs? I use wordpad, but it's not so good
> with large logs.

any decent file /viewer/ should work... you don't need an editor just to look...
in some cases, i use the old dos based LIST.COM by
vern buerg... it is available in newer versions and is being updated, IIRC... i
believe that there is an obvious url in the name, too ;)

> Also it seems like all of you run your own server - my website is hosted by
> a hosting company, so I have no control over Webalizer functions at all.

that shouldn't really matter as long as you can get to your logs and maintain
whatever config options you need to maintain...

--
         _\/
        (@@)                      Waldo Kitty, Waldo's Place USA
__ooO_( )_Ooo_____________________ telnet://bbs.wpusa.dynip.com
_|_____|_____|_____|_____|_____|_____ http://www.wpusa.dynip.com
____|_____|_____|_____|_____|_____|_____ ftp://ftp.wpusa.dynip.com
_|_Eat_SPAM_to_email_me!_YUM!__|_____|_____ wkitty42 -at- alltel.net

ok, i know this has been discussed, and I've searched the archives, but
can't come up with anything good. so here's my question, and maybe a good
answer can be added to the FAQ or something.


I am running Apache 1.3.x, and i have 1 log file (for 1 domain), and it
just grows and grows until i manually rotate it (usually the 1st week of
the month).  Then i user egrep and break it out by month, so i'm left with
a file like april.2004.log.  then i manually edit the webalizer.conf file
to point to that log, and then run it for the months stats.

This is a long, boring process, and now we want to get them updated daily,
i have full control over the server and cron, so that's not an issue, it's
just the layout i need help with.


Should i run a cron script 1 second after midnight to get the previous day
in 1 file, name it daily.log, then fireoff the webalizer, then when it's
done, have it rename the log to like 050904.log (whatever day the log is
for) or something.  Or is there a MUCH easier way to do this?

Thanks

brian

Any particular reason why you want to rotate the logs daily instead of
monthly?  Doesn't seem like space is an issue..  You can still generate
daily stat reports without having to rotate your logs you know :)

--

On Mon, 10 May 2004, Brian Quinn wrote:

> ok, i know this has been discussed, and I've searched the archives, but
> can't come up with anything good. so here's my question, and maybe a good
> answer can be added to the FAQ or something.
>
>
> I am running Apache 1.3.x, and i have 1 log file (for 1 domain), and it
> just grows and grows until i manually rotate it (usually the 1st week of
> the month).  Then i user egrep and break it out by month, so i'm left with
> a file like april.2004.log.  then i manually edit the webalizer.conf file
> to point to that log, and then run it for the months stats.
>
> This is a long, boring process, and now we want to get them updated daily,
> i have full control over the server and cron, so that's not an issue, it's
> just the layout i need help with.
>
>
> Should i run a cron script 1 second after midnight to get the previous day
> in 1 file, name it daily.log, then fireoff the webalizer, then when it's
> done, have it rename the log to like 050904.log (whatever day the log is
> for) or something.  Or is there a MUCH easier way to do this?
>
> Thanks
>
> brian
>
>
>
> Webalizer homepage: http://www.webalizer.org
> Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
> ttp://www.webalizer.org
> Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
>
> Yahoo! Groups Links
>
>
>
>
>

i wasn't sure i could run them daily and rotate them monthly, i guess i
should have figured that out if i would have thought about
it.  Thanks.   Anything special i need to do?  does it have a problem
running while the logs are still being written too?

Thanks

Brian




At 09:22 PM 5/10/2004, you wrote:

>Any particular reason why you want to rotate the logs daily instead of
>monthly?  Doesn't seem like space is an issue..  You can still generate
>daily stat reports without having to rotate your logs you know :)
>
>--
>
>On Mon, 10 May 2004, Brian Quinn wrote:
>
> > ok, i know this has been discussed, and I've searched the archives, but
> > can't come up with anything good. so here's my question, and maybe a good
> > answer can be added to the FAQ or something.
> >
> >
> > I am running Apache 1.3.x, and i have 1 log file (for 1 domain), and it
> > just grows and grows until i manually rotate it (usually the 1st week of
> > the month).  Then i user egrep and break it out by month, so i'm left with
> > a file like april.2004.log.  then i manually edit the webalizer.conf file
> > to point to that log, and then run it for the months stats.
> >
> > This is a long, boring process, and now we want to get them updated daily,
> > i have full control over the server and cron, so that's not an issue, it's
> > just the layout i need help with.
> >
> >
> > Should i run a cron script 1 second after midnight to get the previous day
> > in 1 file, name it daily.log, then fireoff the webalizer, then when it's
> > done, have it rename the log to like 050904.log (whatever day the log is
> > for) or something.  Or is there a MUCH easier way to do this?
> >
> > Thanks
> >
> > brian
> >
> >
> >
> > Webalizer homepage: http://www.webalizer.org
> > Webalizer for NT: http://www.medasys-lille.com/webalizer/
> >
> >
> > ttp://www.webalizer.org
> > Webalizer for NT: http://www.medasys-lille.com/webalizer/
> >
> >
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
>
>
>Webalizer homepage: http://www.webalizer.org
>Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
>ttp://www.webalizer.org
>Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
>
>Yahoo! Groups Links
>
>
>
>

I need to track dynamic pages for both a php site and a CF site.  What I mean is
I have a
"list" page with several images.  When I click on an image, I get a detail page
with
information about the image. All this is database driven.  So my list page is
list.php and
my detail page is detail.php?id=123 (for example).  So I need to know how many
times
database record 123 (id=123) was viewed.  How do I set Webalizer to track this? 
I know
Web Tredns can do this sort of thing.

Thanks!

forphil

2) the countries option show unresolved/unknown 100 %

 

do i need to make any changes in my squid.conf file

 

appreciate your help

 

thanks and regards

 

simon

 

Hello:

To track dinamic pages you'll have to download the source code, make a
small change to it, and compile the source. Then use the program you
just compiled instead of the binary version (the one you don't need to
compile).

The change is:

Open the file "webalizer.c". You need to change the "isurlchar" code.
Search for "isurlchar" until you find something like this:

/*********************************************/
/* ISURLCHAR - checks for valid URL chars    */
/*********************************************/

int isurlchar(unsigned char ch)
{
    if (isalnum((int)ch)) return 1;           /* allow letters,
numbers...    */
    if (ch > 127) return 1;                   /* allow extended chars..
.      */
    return (strchr(":/\\.,' *-+_@~()[]",ch)!=NULL); /* and a few
special ones */
}



You need to change the last line for this one instead. Notice the
addition of "?":

    return (strchr("?:/\\.,' *-+_@~()[]",ch)!=NULL);

After this, all "?" are treated as a normal URL char.



NOTE: The search string feature will stop working, because it relies
on the meaning of "?" as a special character separating the URL from
the query string...



NOTE: Notice that these three URLS are three different URLs:

detail.php?id=123
detail.php?id=123&p=2
detail.php?p=2&id=123

If you have more than one parameter, you may consider grouping (this
will only work after changing "isurlchar" and compiling):

GroupURL ?id=123   id 123
HideURL  ?id=123



You can find some discussion on security issues from doing this and
some other random oddities by searching the function name in the group
messages:

http://groups.yahoo.com/group/webalizer/messagesearch?query=isurlchar



--- In webalizer@yahoogroups.com, "forphil75" <forphil@e...> wrote:
> I need to track dynamic pages for both a php site and a CF site.
What I mean is I have a
> "list" page with several images.  When I click on an image, I get a
detail page with
> information about the image. All this is database driven.  So my
list page is list.php and
> my detail page is detail.php?id=123 (for example).  So I need to
know how many times
> database record 123 (id=123) was viewed.  How do I set Webalizer to
track this?  I know
> Web Tredns can do this sort of thing.
>
> Thanks!
>
> forphil

> hi,

Hello.

>
> I have recently used webalizer to process my squid log files and
there are 2 queries i want to clearify
>
> i am using the patched webalizer and when I run it i get the below
errors . the squid access.log file size was about 9 mb
>
> 1) Warning: Truncating oversized request field
> Skipping bad record (50155)
> Warning: Truncating oversized request field [50236]
> Warning: Truncating oversized request field [50915]
> Warning: Truncating oversized request field [51177]
>

These are only warnings. You may want to attempt to solve the "bad
record" warning, because of its potencial for problems in the future.
The number in parenthesis is the line number in the log file.

Launch your favourite text editor and open the squid logfile. Go to
line number 50155 and inspect the line and the lines around it. You
will probably find a log entry cut in half, or some other strange
thing. Delete the faulty line or correct it if you prefer so.

The "oversized request field" are very long URLs.  You can edit the
code and increase the size of the internal fields storing the URLs,
but notice that this will cause webalizer to increase menory usage.
You may run out of RAM if your machine has very little memory.

In webalizer.h you'll find a line like this:

#define MAXURLH  128                   /* Max URL field size in htab
     */

You need to change 128 to a higher value. If you have a 32 bit
machine, it is more efficient to increase this value in increases of
32. Try 128+32=160 or 128+32*2=192

#define MAXURLH  160

You'll probably still have warnings unless you put a very high value.
Normally those are attempts to overflow some buffer in your server, or
some sites which have exagerately long request strings for some
reason.

> 2) the countries option show unresolved/unknown 100 %
>
>
>
> do i need to make any changes in my squid.conf file

Hum, have you activated the DNS resolution in webalizer.conf? This
line should be uncommented. No need to uncomment the DNSCache line.

DNSChildren    10

Also notice that you need to compile using this option in the
"configure" script:

make clean
./configure --enable-dns
make
make install


Is your server capable of resolving addresses? (try executing
"nslookup some_ip" in the command line)

--- In webalizer@yahoogroups.com, "t_stangel@y..." <t_stangel@y...>
wrote:
> hey there,
> i have a very unusual problem with webalizer. well im not sure its
webalizer at all but that is where the problem seems to be comming
from. every week or so, one of my hostees stats directory gets renamed
from "stats/" to "etc/". then the cron daemon goes nuts and emails me
about it. it only gets more confusing when the stats directory is
still there but its empty.
>
> each one of my hostees have separate logs so i have a simple perl
script which processes them all every hour in the cron.hourly
directory. the system is redhat 8 and it has only happened to one
hostee. thanks for the help in advance!


You perhaps have some script in cron.weekly which is moving that
directory? Could you copy/paste the cron email text and a "ls -ld" and
a "ls -l" of both directories right after the problem appears, just to
check ownerships and timestamps? Also the crontab line for cron.
weekly, and a "ls" of /etc/cron.weekly to check if the directories
were created around the time a particular script is running.

In any case, you may use a very ugly "patch" to solve the problem by
adding these lines in the webalizer script before running webalizer.
(ejem, I just remembered that you use perl, not bash, just translate
adequately to perl)

PROBLEMATIC_DIR="/hostee/stats/directory/"
mkdir $PROBLEMATIC_DIR/stats/
cp -rf $PROBLEMATIC_DIR/etc/* $PROBLEMATIC_DIR/stats/

This will copy all files in "etc/" to "stats/". Notice that files in
"stats/" holding the same name as some other file in "etc/" will get
overwritten.

NOTE: Notice that "dir/etc" is the same as "dir//etc". I just add "/"
to end of all paths just to be sure that I don't forget later to add
it, so I get many times paths like "/home//user//stats/dir". It will
(in my system) work with no problems.

--- In webalizer@yahoogroups.com, sanket brahmbhatt <sanket_brahm@y...
> wrote:
> Hello to everybody in the group.
>
> This is my first question. Being a newcomer and not totally
acquainted with Webalizer, I want to know whether it is designed for
only UNIX families or can be used for Windows too. Can anybody
elaborate on that?

There is a NT version. The author of it is in this group.

Webalizer for NT: http://www.medasys-lille.com/webalizer/

ah okay. i thought about doing that in perl with a if statement but i am just so baffled at why it might be doing this. i will post all of that once it happens again. thanks!

-todd

> This is my first question. Being a newcomer and not totally acquainted
> with Webalizer, I want to know whether it is designed for only UNIX
> families or can be used for Windows too. Can anybody elaborate on that?

The webalizer runs across pretty much every platform out there except
MacOS (MacOSX is however supported).  There is a win32 binary available
on our web/ftp site.

--
Bradford L. Barrett                      brad@...
A free electron in a sea of neutrons     DoD#1750 KD4NAW

The only thing Micro$oft has done for society, is make people
believe that computers are inherently unreliable.

Hey everyone.  First question here...  I inherited an existing site /
webalizer install.  The only thing I need to do right now is have it report
on 1 or 2 pages in addition to the Top 20 that show up now.  What do I need
to do, to have it report stats on individual pages?  Any help would be
greatly appreciated, thank you!

Tim Chetelat
tim@...
http://www.twisted-reality.com/

AllURLs  yes

--

On Wed, 19 May 2004, Tim Chetelat wrote:

> Hey everyone.  First question here...  I inherited an existing site /
> webalizer install.  The only thing I need to do right now is have it report
> on 1 or 2 pages in addition to the Top 20 that show up now.  What do I need
> to do, to have it report stats on individual pages?  Any help would be
> greatly appreciated, thank you!
>
> Tim Chetelat
> tim@...
> http://www.twisted-reality.com/
>
>
>
>
>
> Webalizer homepage: http://www.webalizer.org
> Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
> ttp://www.webalizer.org
> Webalizer for NT: http://www.medasys-lille.com/webalizer/
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
--
Bradford L. Barrett                      brad@...
A free electron in a sea of neutrons     DoD#1750 KD4NAW

The only thing Micro$oft has done for society, is make people
believe that computers are inherently unreliable.

You could increase the TopURLs value so that those
pages appear in the listing and then group those pages
so they appear grayed to increase visibility.

GroupURL  /images/index-0.htm
HideURL   /images/index-0.htm

GroupURL  /forums/topics/user-index-0.php
HideURL   /forums/topics/user-index-0.php

Be aware that these keywords are matching substrings,
so things like this would group both pages together:

GroupURL  index-0.
HideURL   index-0.



--- Tim Chetelat <tim@...> wrote:
> Hey everyone.  First question here...  I inherited
> an existing site /
> webalizer install.  The only thing I need to do
> right now is have it report
> on 1 or 2 pages in addition to the Top 20 that show
> up now.  What do I need
> to do, to have it report stats on individual pages?
> Any help would be
> greatly appreciated, thank you!
>
> Tim Chetelat
> tim@...
> http://www.twisted-reality.com/
>
>
>


=====
Enric Naval
Estudiante de Informática de Gestión en la Udl (Lleida)
GRIHO enhanced webalizer.conf
http://griho.udl.es/webalizer/webalizer.conf.txt




__________________________________
Do you Yahoo!?
SBC Yahoo! - Internet access at a great low price.
http://promo.yahoo.com/sbc/

--- "Bradford L. Barrett" <brad@...> wrote:
>
> AllURLs  yes

This creates a complete list that you can use, but you
may need to create a "nice looking" list, so that
other people (clientes or supervisors), may easily
find the info they are looking for. You sometimes need
to chew the information for them, so you could grep
the file the list is in and create a nice report.

If you give them that list and tell them to search for
the URL by hand, they will probably never do it,
because there is so much info in it that they will
feel overhelmed.

>
> --
>
> On Wed, 19 May 2004, Tim Chetelat wrote:
>
> > Hey everyone.  First question here...  I inherited
> an existing site /
> > webalizer install.  The only thing I need to do
> right now is have it report
> > on 1 or 2 pages in addition to the Top 20 that
> show up now.  What do I need
> > to do, to have it report stats on individual
> pages?  Any help would be
> > greatly appreciated, thank you!
> >
> > Tim Chetelat
> > tim@...
> > http://www.twisted-reality.com/
> >
> >
> >
> >
> >
> > Webalizer homepage: http://www.webalizer.org
> > Webalizer for NT:
> http://www.medasys-lille.com/webalizer/
> >
> >
> > ttp://www.webalizer.org
> > Webalizer for NT:
> http://www.medasys-lille.com/webalizer/
> >
> >
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
> --
> Bradford L. Barrett
> brad@...
> A free electron in a sea of neutrons     DoD#1750
> KD4NAW
>
> The only thing Micro$oft has done for society, is
> make people
> believe that computers are inherently unreliable.
>


=====
Enric Naval
Estudiante de Informática de Gestión en la Udl (Lleida)
GRIHO enhanced webalizer.conf
http://griho.udl.es/webalizer/webalizer.conf.txt




__________________________________
Do you Yahoo!?
SBC Yahoo! - Internet access at a great low price.
http://promo.yahoo.com/sbc/