On Sun, 19 Sep 1999, you wrote:
>Ale,
>
> Thanks for your help. I have a couple follow up questions if you don't
>mind.
>
> 1.) re : The monthly file does the following stuff:.....
>
> when you say the monthly file are you referring to
>anamonth.sh reffered to in the cron?
>
Yes. In the following line, it's saying "At 0 minutes past 07 am, 1st day of
the month. You can get more examples in the manpage for crontab: either
man 5 crontab
or just man crontab should do the trick.
>0 07 1 * * (cd /ale/work/stats/; ./anamonth.sh 2>&1 |mail > [EMAIL PROTECTED])
>
> what does the 2>&1 do?
>
It redirects standard error (2) to standard output (1) (I'm not *that* sure
about this, but it works) and then pipes that to my email. This means I get any
error messages as well as the usual analog output in the email.
> 2.) Could you also elaborate on this last line? What exactly is
it doing >to automate the process?
>
> echo "<li><a href=\"$TODAY\" title=\"$THISMONTH\">Month of
>$THISMONTH</a>"
> >> /software/apache/htdocs/site/stats/incl_monthlist.html
>
I'm grouping all reports for any site in a file called /stats/index.html
accessible from some part of that site. This is so any clients can
see it easily. On that page I've got something like the following for every
kind of report (You have to make sure you have server side includes working
for this):
<h3>Weekly Stats</h3>
<ul>
<!--#include virtual="incl_weeklist.html"-->
</ul>
So the line that the .sh file is appending to each time it's run, is actually
adding each new report to that index page, automatically. I just use this as a
time saver: You could probably write some kind of parsing cgi file to be less
error prone (If you run the script twice by accident, you get double entries
in the include file at the moment) ... but this way I get the job done quickly
and I rarely have to go in and do html.
> 3.) Do you have a recomendation of how I could compress the log
>files? Should I do it daily? what would the cron command be for that?
>
We do it nightly (Someone else does that stuff though). It's complicated, but
you could try using the find command, using the -mtime switch, to get just the
most recently made log file, and running gzip just on that one. Maybe it would
be something like:
30 1 * * * find /www/logs/site -name * -mtime -1 -exec gzip -v {} \;
Don't blame me if this command formats your hard disk and sets fire to your
office. But the find command is quite powerful, and you should be able to do
some complex stuff with it once you've fiddled around a bit.
The complicated bit comes in when you start dealing with multiple
sites...moving this gzipped log file to it's appropriate storage directory...
No idea how to do that... Sorry!
Ale
--
Alejandro Fernandez,
[EMAIL PROTECTED]
------------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
List archived at http://www.mail-archive.com/[email protected]/
------------------------------------------------------------------------
