
<!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>LDAP Forever</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/211140558534.html</link>
<description>
&lt;p&gt;More investigation of the ldap manager problem, but nothing
found. On a happier note, Greg asked for help on an sql query and I
suggested a way to do it. I'll be away at the &lt;a
href="http://us.pycon.org/TX2006/HomePage"&gt;Python conference&lt;/a&gt; until
next Tuesday.&lt;/p&gt;

</description>
</item>

<item>
<title>LDAP Debugging</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/201140472447.html</link>
<description>
&lt;p&gt;I added some more debugging prints, this time directly in the
Python ldap module. I got the parameters and then wrote a Python
script and a Perl script that as far as I can tell, do the same
thing. However, they return valid results and Zope does not. I'm still
trying to figure out the difference.&lt;/p&gt;
</description>
</item>

<item>
<title>Stuck on LDAP</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/171140213252.html</link>
<description>
&lt;p&gt;Here's what I know about the problem. When logging in, the code
tries to bind to ldap with user name and password. It doesn't have
that info, so it binds anonymously. Then it queries ldap to get the
distinguished name from ldap. This query fails, for unknown reasons,
so it assumes the user is anonymous. This is an unacceptable
result. so it asks for the password again. The problem is clearly with 
the query, but I haven't figured out why yet.&lt;/p&gt;
</description>
</item>

<item>
<title>LDAP Continued</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/161140124169.html</link>
<description>
&lt;p&gt;Narrowing down the problem, but still no joy.&lt;/p&gt;
</description>
</item>

<item>
<title>Continued Work on LDAP</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/151140041098.html</link>
<description>
&lt;p&gt;Contrary to what I previously thought, the ldap manager is going
through the LDAP user folder. However, it is failing to return any
results from the ldap query. I'm still trying to figure out why.&lt;/p&gt;
</description>
</item>

<item>
<title>More LDAP</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/141139953926.html</link>
<description>
&lt;p&gt;I ran the problem past Harry, and he couldn't come up with
anything. So I spent the rest of the day studying the code, which is
perversely opaque in this area.&lt;/p&gt;
</description>
</item>

<item>
<title>LDAP Authorization</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/131139868726.html</link>
<description>
&lt;p&gt;From Zope Authorization to LDAP authorization. LDAP authorization
is failing. It looks like the code is searching for the getUserDN
method and logging in as anonymous when it fails to find it. The only
place I see this method defined is in LDAPUserFolder. It's unclear to
me how LDAPUserFolder is invoked to override the default user
folder.&lt;/p&gt;
</description>
</item>

<item>
<title>Flotsam</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/101139607844.html</link>
<description>
&lt;p&gt;Mike informed me that the calendar on this website was creating
problems for the search appliance, so I disabled it. (Actually I tried 
to fix it and failed. I'll look at it again when I find time.) The
script to re-analyze the network statistics finished yesterday. I took 
the output and used it to update the network statistics files.&lt;/p&gt;
</description>
</item>

<item>
<title>Authorization Change Done</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/091139520795.html</link>
<description>
&lt;p&gt;We were having trouble testing my authorization fix over on battra, 
so I moved it over to gorgo. We were able to complete testing
there. In light of the results, I revised my fix of the code,
retested, then deleted the debugging prints and handed the code over
to Mike for more serious testing.&lt;/p&gt;
</description>
</item>

<item>
<title>More Zope</title>
<link>http://www-int.stsci.edu/~bsimon/weblog.cgi/post/081139435865.html</link>
<description>
&lt;p&gt;I'm still debugging the authorization checking code. It seems to
work well on the single url case, but not so well on the two url
case. We haven't got squid set up yet, which makes testing more
difficult.&lt;/p&gt;
</description>
</item>

</channel>
</rss>
