Re: novice query
Roy T. Fielding (fielding@liege.ICS.UCI.EDU)
Sun, 22 Dec 1996 01:45:31 -0800
> the problem is - and this is probably
> really obvious for the savvy folks reading
> this tale of woe - my ISP feels it is
> neccessary to gzip the access_log file
> at the end of each day.
>
> pwd
> /home/www/vs/httpd.soleilmoon/logs/OLD
>
> ls -lg access_log*
> -rw-r--r-- 1 www www 46240 Dec 14 00:56 access_log.0
> -rw-r--r-- 1 www www 6321 Dec 14 01:31 access_log.1.gz
> -rw-r--r-- 1 www www 3621 Dec 5 01:37 access_log.10.gz
> -rw-r--r-- 1 www www 6839 Dec 4 01:37 access_log.11.gz
> -rw-r--r-- 1 www www 2904 Dec 3 01:48 access_log.12.gz
> -rw-r--r-- 1 www www 2320 Dec 2 01:35 access_log.13.gz
> -rw-r--r-- 1 www www 5746 Dec 1 09:35 access_log.14.gz
> -rw-r--r-- 1 www www 4650 Nov 30 13:42 access_log.15.gz
> -rw-r--r-- 1 www www 4436 Nov 29 01:38 access_log.16.gz
> -rw-r--r-- 1 www www 6098 Nov 28 01:41 access_log.17.gz
> -rw-r--r-- 1 www www 3198 Nov 27 01:39 access_log.18.gz
> -rw-r--r-- 1 www www 4101 Nov 26 01:40 access_log.19.gz
> ...
>
> clearly, this means i need to configure
> wwwstat to access the files in OLD if i
> want to generate a report for stats covering
> the last month, or last n days. but, i'm
> just not proficient (yet!) with Perl to
> make this work. i know i need to use
> pattern matching in order for wwwstat to
> be able to search all of the neccessary
> gzip'd access_log files, depending on how
> i want to generate the appropriate report.
The command-line shell (not wwwstat) needs to select the files. Use
wwwstat access_log.*
and let the shell determine the file names. wwwstat will automatically
decompress any name ending in .gz
...Roy T. Fielding
Department of Information & Computer Science (fielding@ics.uci.edu)
University of California, Irvine, CA 92697-3425 fax:+1(714)824-4056
http://www.ics.uci.edu/~fielding/