Bang away
--#--
Now, tail the data file: <i>tail -f /usr/temp/data</i> and 
watch what happens when you click the submit button. You should 
see the data added on each submit.<br>
<br>
<code>
Fri Nov 12 19:22:40 1999	John	crypto	4
Fri Nov 12 19:22:52 1999	John	crypto	4
Fri Nov 12 19:23:03 1999	John	crypto	4
Fri Nov 12 19:25:27 1999	John	crypto	4
</code>
<br>
If you're not seeing this, be sure your Web server has the 
authority to write to /usr/temp/data.  If need be, give 
/usr/temp and /usr/temp/data a mode of 777.<br>
<br>
Change the field values in the HTML form to something different
so you'll recognize them, and add a pause to the Perl 
script, say, 7 seconds: <i>sleep(7);</i>. Add the sleep 
statement somewhere between where the script grabs the 
data and where it writes it to the file.<br>
<br>
When you run the form now, you'll see a little hourglass or 
wristwatch or whatever your OS uses to tell you to park it 
for a minute. But forget that. You're in a hurry. Hammer on the 
submit button a couple or few times.<br>
<br>
Finally, you'll get the Thanks! screen. Meanwhile, your data 
file starts to grow.<br>
<br>
<code>
Fri Nov 12 19:22:40 1999	John	crypto	4
Fri Nov 12 19:22:52 1999	John	crypto	4
Fri Nov 12 19:23:03 1999	John	crypto	4
Fri Nov 12 19:25:27 1999	John	crypto	4
Fri Nov 12 19:48:10 1999	Bubba	foozball	999
Fri Nov 12 19:48:12 1999	Bubba	foozball	999
Fri Nov 12 19:48:13 1999	Bubba	foozball	999
Fri Nov 12 19:48:14 1999	Bubba	foozball	999
Fri Nov 12 19:48:16 1999	Bubba	foozball	999
</code>
<br>
Note, however, that the time on your Thanks! screen matches only 
the <i>last</i> set of data. The user is left with the impression 
that all that hammering on the submit button finally pushed the 
data through, but only one set.<br>
<br>
OK. We've clarified the problem. What do we do about it?
<br>