
<!DOCTYPE rss PUBLIC  "-//Netscape Communications//DTD RSS 0.91//EN"
"http://my.netscape.com/publish/formats/rss-0.91.dtd">
<rss version="0.91">
<channel> 
<title></title> 
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi</link> 
<description>The work log of Bernie Simon</description> 
<language>en</language> 
<item>
<title>Squid</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/291167429746.html</link>
<description>
&lt;p&gt;I spent the day reading the Squid book that I had ordered for info
about configuring squid.&lt;/p&gt;
</description>
</item>

<item>
<title>Scripts Done</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/281167343724.html</link>
<description>
&lt;p&gt;I wrote the rest of the test scripts and ran them. I had a problem
with the copy/paste script, because copy/paste needs cookie
support. So I added cookies to runscript to fix the problem. I grabbed 
a day's worth of urls from the log files so we can have a typical mix
of requests to test.&lt;/p&gt;
</description>
</item>

<item>
<title>It's a Lock</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/271167253377.html</link>
<description>
&lt;p&gt;I got the code in runscript that retrieves existing ids to
work. The tricky part was creating a lock file to keep the same id
from being used, since the script runs multiple concurrent
versions.&lt;/p&gt;
</description>
</item>

<item>
<title>Testing</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/261167170387.html</link>
<description>
&lt;p&gt;The test code I wrote Friday works, but I ran into a problem
writing the scripts. Some of the tests involve changing existing
files, for which purpose I need to obtain them from a folder. I
decided the easiest way to do this is screen scaping, so I started
writing the code.&lt;/p&gt;
</description>
</item>

<item>
<title>Runscript runs</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/221166822713.html</link>
<description>
&lt;p&gt;I finished the testing script that runs back to back commands. It's 
called runscript and it runs to completion in the first test case
without crashing. Which I guess is something.&lt;/p&gt;
</description>
</item>

<item>
<title>Scripting</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/211166735346.html</link>
<description>
&lt;p&gt;I checked out the operations we'll need to do in Zope and verified
that we'll need scripting (back to back commands.) I thought a bit
about how to do that and then started coding.&lt;/p&gt;
</description>
</item>

<item>
<title>More Torture</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/201166652434.html</link>
<description>
&lt;p&gt;I finished the scoping document and mailed it around. Then I
started making the necessary changes in the torture script, which we
will use for testing Zope performance. I modified the script so that
it uses a Zope user name and password, which are necessary because
most of our tests will be with privileged urls. I also modified the
script to take query parameters tacked on the end of the url and turn
them into a POST request. I started testing the revised script and so
far everything works fine. I probably will need to add the capability
to run scripts, that is, collections of urls needed to accomplish a
task, but I'm not sure how to approach that yet. I got my new Palm
(the old one was zorched) and set it up on my computer.&lt;/p&gt;
</description>
</item>

<item>
<title>Zope Improvement Project</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/191166566267.html</link>
<description>
&lt;p&gt;I wrote a Zope script to sum up the size of ST Files and Images in
our system. These files can be removed from our data.fs file if we
move to an external file system and zope and this is probably the
easiest way to decrease our data.fs size. I also installed the ExtFile 
product on battra to see what it looked like. I had a meeting with
Steve, Anee, and Craig to clarify what this project is about and
started to rewrite the scoping document.&lt;/p&gt;

</description>
</item>

<item>
<title>Rescoped</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/181166479671.html</link>
<description>
&lt;p&gt;We discussed the Zope Improvement project again and the
overwhelming concensus is that it should focus on performance. I'll
redo the scoping document to reflect this. I looked at the Zope site
and found a lead to one performance monitoring tool. I also looked for
external file sytem products and the Zope world seems to have settled
on the &lt;a href="http://zope.org/Members/shh/ExtFile"&gt;extfile /
extimage&lt;/a&gt; product.&lt;/p&gt;
</description>
</item>

<item>
<title>TWiki Mail Notification</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/151166220447.html</link>
<description>
&lt;P&gt;Gretchen told me yesterday that TWiki mail notifcation was not
working. I debugged the problem today and determined it was a problem
in the mailnotify.bat script I had created out of the Perl script. It
took a while to find because I am not knowledgable about MS-DOS batch
scrripts (to put it mildly) but it turned out that it didn't like the
taint (-T) option to the perl command and it couldn't find the
setlib.cfg file in the path, because I had moved the batch job to a
new directory. Fixing both problems allowed the script to work and
mail notifications on changes to TWiki should now go out daily.&lt;/p&gt;
</description>
</item>

</channel>
</rss>
