<?xml version="1.0" encoding="UTF-8"?>
<!-- generator="wordpress/wordpress-mu-1.0" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>

<channel>
	<title>David's Blog</title>
	<link>http://david.surmise.co.uk</link>
	<description>Just another blogs.surmise.co.uk weblog</description>
	<pubDate>Mon, 12 Jul 2010 21:19:36 +0000</pubDate>
	<generator>http://wordpress.org/?v=wordpress-mu-1.0</generator>
	<language>en</language>
			<item>
		<title>SSL Certificates</title>
		<link>http://david.surmise.co.uk/2010/07/12/ssl-certificates/</link>
		<comments>http://david.surmise.co.uk/2010/07/12/ssl-certificates/#comments</comments>
		<pubDate>Mon, 12 Jul 2010 21:19:36 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2010/07/12/ssl-certificates/</guid>
		<description><![CDATA[I&#8217;m currently trying to set up mallard with a proper ssl certificate (free ones are available from www.startssl.com).  The authority requires some sort of client certificate form of authentication to login to their site.  Unfortunately Chrome from Linux doesn&#8217;t yet support this properly.  After much googling I found that I could export [...]]]></description>
			<content:encoded><![CDATA[<p>I&#8217;m currently trying to set up mallard with a proper ssl certificate (free ones are available from www.startssl.com).  The authority requires some sort of client certificate form of authentication to login to their site.  Unfortunately Chrome from Linux doesn&#8217;t yet support this properly.  After much googling I found that I could export my certificate from Chrome and then import it into Firefox:<br />
<code>pk12util -o certfile.p12 -d sql:$HOME/.pki/nssdb -n </code><br />
The certificate&#8217;s name is obtained running<br />
<code>certutil -d sql:$HOME/.pki/nssdb -L</code></p>
<p>(Source: <a href="http://serverfault.com/questions/106619/how-to-export-chrome-user-certificate">serverfault</a>)</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2010/07/12/ssl-certificates/feed/</wfw:commentRss>
		</item>
		<item>
		<title>F-Spot Facebook Extension Bug</title>
		<link>http://david.surmise.co.uk/2009/09/18/f-spot-facebook-extension-bug/</link>
		<comments>http://david.surmise.co.uk/2009/09/18/f-spot-facebook-extension-bug/#comments</comments>
		<pubDate>Fri, 18 Sep 2009 11:34:55 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2009/09/18/f-spot-facebook-extension-bug/</guid>
		<description><![CDATA[There is a bug in the f-spot facebook extension included with Ubuntu Gutsy and Intrepid (F-Spot crashes during authorization exporting to Facebook if friend IDs are large).  Didn&#8217;t want to wait until Jaunty to upload pictures to Facebook, so I&#8217;ve compiled FacebookExport.dll using the upstream patch and copied it over /usr/lib/f-spot/extensions/FacebookExport.dll (making a backup of [...]]]></description>
			<content:encoded><![CDATA[<p>There is a bug in the f-spot facebook extension included with Ubuntu Gutsy and Intrepid (<a href="https://bugs.launchpad.net/ubuntu/+source/f-spot/+bug/405005">F-Spot crashes during authorization exporting to Facebook if friend IDs are large</a>).  Didn&#8217;t want to wait until Jaunty to upload pictures to Facebook, so I&#8217;ve compiled <a href="http://mallard.yellowbill.co.uk/~david/files/2009/09/18/FacebookExport.dll">FacebookExport.dll</a> using the upstream patch and copied it over /usr/lib/f-spot/extensions/FacebookExport.dll (making a backup of course!).  Works for me, but no gaurentees.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2009/09/18/f-spot-facebook-extension-bug/feed/</wfw:commentRss>
		</item>
		<item>
		<title>ntpd</title>
		<link>http://david.surmise.co.uk/2009/01/02/ntpd/</link>
		<comments>http://david.surmise.co.uk/2009/01/02/ntpd/#comments</comments>
		<pubDate>Fri, 02 Jan 2009 00:18:26 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2009/01/02/ntpd/</guid>
		<description><![CDATA[ntpd is now working on mallard (so the clock is nolonger 3min slow).  Apparently
restrict -4 default ignore
was to strict to actually synchronise with other servers, but
restrict -4 default noquery nomodify nopeer notrap
seems to work.

]]></description>
			<content:encoded><![CDATA[<p>ntpd is now working on mallard (so the clock is nolonger 3min slow).  Apparently<br />
<code>restrict -4 default ignore</code><br />
was to strict to actually synchronise with other servers, but<br />
<code>restrict -4 default noquery nomodify nopeer notrap</code><br />
seems to work.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2009/01/02/ntpd/feed/</wfw:commentRss>
		</item>
		<item>
		<title>Results (at last)</title>
		<link>http://david.surmise.co.uk/2008/07/15/results-at-last/</link>
		<comments>http://david.surmise.co.uk/2008/07/15/results-at-last/#comments</comments>
		<pubDate>Tue, 15 Jul 2008 20:00:43 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2008/07/15/results-at-last/</guid>
		<description><![CDATA[It seems to be the done thing to comment on your results here, so I&#8217;ll give a brief overview of mine.
Analogue and Digital Filters, etc, I was quite happy with, though confused as to why about three courses are lumped into one line. Advanced Digital synthesis- well aparently a B- plus a B- plus an [...]]]></description>
			<content:encoded><![CDATA[<p>It seems to be the done thing to comment on your results here, so I&#8217;ll give a brief overview of mine.</p>
<p>Analogue and Digital Filters, etc, I was quite happy with, though confused as to why about three courses are lumped into one line. Advanced Digital synthesis- well aparently a B- plus a B- plus an A makes 76%.. makes sense?</p>
<p>Like Ben I was surpised by Comms, didn&#8217;t think the result went as well as I&#8217;d thought the exam went. Distributed Computing was my best exam, though I did least work for it. I was quite pleased with the ECADs (70% in Media and 72% in Music - pretty good esp considering I&#8217;m rubbish at lab books and I didn&#8217;t really care anymore when I handed them in).  In SWEng I got 76%, which I was quite happy with.  Does anyone know if we get any feedback on SWEng?</p>
<p>So, 79% for the year, bringing my average done to 79%.  So as long as I don&#8217;t screw up too much next year..
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2008/07/15/results-at-last/feed/</wfw:commentRss>
		</item>
		<item>
		<title>Pretending to be an IMAP server using netcat</title>
		<link>http://david.surmise.co.uk/2008/06/11/pretending-to-be-an-imap-server-using-netcat/</link>
		<comments>http://david.surmise.co.uk/2008/06/11/pretending-to-be-an-imap-server-using-netcat/#comments</comments>
		<pubDate>Wed, 11 Jun 2008 23:07:57 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2008/06/11/pretending-to-be-an-imap-server-using-netcat/</guid>
		<description><![CDATA[So, say you want to pretend to be an imap server using nc (can&#8217;t think of any good reason why you&#8217;d want to do this, we were trying to seriously confuse an iphone). So, you do something like:
cat - &#124; nc -l -p 143 -v
And when the line
connect to [127.0.0.1] from localhost [127.0.0.1] 39656
appears you [...]]]></description>
			<content:encoded><![CDATA[<p>So, say you want to pretend to be an imap server using nc (can&#8217;t think of any good reason why you&#8217;d want to do this, we were trying to seriously confuse an iphone). So, you do something like:</p>
<p><code>cat - | nc -l -p 143 -v</code></p>
<p>And when the line</p>
<p><code>connect to [127.0.0.1] from localhost [127.0.0.1] 39656</code></p>
<p>appears you type</p>
<p><code>* OK myIMAPserver</code></p>
<p>If the client is you pretending to be an imap client or even <a href="http://mutt.sourceforge.net/">mutt</a>, you&#8217;ll find that you get a response. If, on the other hand, the client is an iPhone, you&#8217;ll get absolutely nothing. Why? Because the RFC for IMAP states:</p>
<blockquote><p>All interactions transmitted by client and server are in the form of<br />
   lines; that is, strings that end with a CRLF.</p></blockquote>
<p>CR is a carriage return (ascii code 0d), LF is a line feed (ascii code 0a)</p>
<p>But netcat doesn&#8217;t believe in sending CRs (fair enough I suppose, unix uses LF as the new line character, with windows using the CRLF pair).  On OSX you can coax nc into sending a CRLF by typing CTRL+V and then pressing return twice.  Once you do that, the iPhone starts talking to you (in a REALLY insecure manner, but that&#8217;s <a href="http://www.nabble.com/XYMPKI-Authentication-(Yahoo!)-Support-td16776436.html">another story</a></p>
<p>Fun.</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2008/06/11/pretending-to-be-an-imap-server-using-netcat/feed/</wfw:commentRss>
		</item>
		<item>
		<title>gnome-screensaver</title>
		<link>http://david.surmise.co.uk/2007/11/24/gnome-screensaver/</link>
		<comments>http://david.surmise.co.uk/2007/11/24/gnome-screensaver/#comments</comments>
		<pubDate>Sat, 24 Nov 2007 20:45:45 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2007/11/24/gnome-screensaver/</guid>
		<description><![CDATA[Spent a couple of hours trying to fix the screen saver on my laptop.  The unlock dialog box refused to appear, so you had to press ctrl-alt-F1, login to the console and killall gnome-screensaver to get back to the desktop.
This FAQ gave me a few hints.  I found I was getting errors like:
Unable [...]]]></description>
			<content:encoded><![CDATA[<p>Spent a couple of hours trying to fix the screen saver on my laptop.  The unlock dialog box refused to appear, so you had to press ctrl-alt-F1, login to the console and killall gnome-screensaver to get back to the desktop.</p>
<p><a href="http://live.gnome.org/GnomeScreensaver/FrequentlyAskedQuestions#head-d50bc17e7d6f3a51c4715f02c657195e80e26c2c">This FAQ</a> gave me a few hints.  I found I was getting errors like:<br />
<code>Unable to establish service gnome-screensaver: Critical error - immediate abort</code></p>
<p>When I looked in /var/log/authlog I found the line:<br />
<code>Nov 24 19:39:24 localhost gnome-screensaver-dialog: PAM _pam_load_conf_file: unable to open /etc/pam.d/common-password</code></p>
<p>chmod 644 /etc/pam.d/common-password fixed the problem.  I have no idea how this file ended up with the wrong permissions.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2007/11/24/gnome-screensaver/feed/</wfw:commentRss>
		</item>
		<item>
		<title>Mallard moves again</title>
		<link>http://david.surmise.co.uk/2007/10/23/mallard-moves-again/</link>
		<comments>http://david.surmise.co.uk/2007/10/23/mallard-moves-again/#comments</comments>
		<pubDate>Tue, 23 Oct 2007 00:09:52 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Web Development</category>

		<category>Surmise</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2007/10/23/mallard-moves-again/</guid>
		<description><![CDATA[This morning mallard moved the 30 miles or so (back) to Redbus Sovereign House from Rapidswitch&#8217;s Maidenhead data centre with only about 15 minutes of down time.  Not bad considering Google thinks the journey alone should have taken 1hr 10min.

]]></description>
			<content:encoded><![CDATA[<p>This morning mallard moved the 30 miles or so (back) to Redbus Sovereign House from Rapidswitch&#8217;s Maidenhead data centre with only about 15 minutes of down time.  Not bad considering Google thinks the journey alone should have taken 1hr 10min.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2007/10/23/mallard-moves-again/feed/</wfw:commentRss>
		</item>
		<item>
		<title>Timetables</title>
		<link>http://david.surmise.co.uk/2007/10/01/timetables/</link>
		<comments>http://david.surmise.co.uk/2007/10/01/timetables/#comments</comments>
		<pubDate>Mon, 01 Oct 2007 16:55:08 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Course</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2007/10/01/timetables/</guid>
		<description><![CDATA[Our timetables have appeared.

]]></description>
			<content:encoded><![CDATA[<p>Our <a href="http://www.amp.york.ac.uk/internal/ugrad/ttable/index.html">timetables</a> have appeared.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2007/10/01/timetables/feed/</wfw:commentRss>
		</item>
		<item>
		<title>Debian Etch</title>
		<link>http://david.surmise.co.uk/2007/09/11/debian-etch/</link>
		<comments>http://david.surmise.co.uk/2007/09/11/debian-etch/#comments</comments>
		<pubDate>Tue, 11 Sep 2007 22:22:09 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2007/09/11/debian-etch/</guid>
		<description><![CDATA[I upgraded the server that hosts this site from Sarge to Etch yesterday.  I then paniced when I typed &#8216;ps aux&#8217; and saw the line
102       7353  0.0  2.3  11632  3040 ?        Ss   Sep10   [...]]]></description>
			<content:encoded><![CDATA[<p>I upgraded the server that hosts this site from Sarge to Etch yesterday.  I then paniced when I typed &#8216;ps aux&#8217; and saw the line</p>
<p><code>102       7353  0.0  2.3  11632  3040 ?        Ss   Sep10   0:00 /usr/sbin/exim4 -bd -q30m</code></p>
<p>The 102 really should have read &#8220;Debian-exim&#8221;.  This concerned me for a long time (not least of all because exim was also having difficulty connecting to the sql database).  Then I discovered that the new version of ps&#8217;s reaction to long usernames is to display the uid instead.  So where on sarge it displayed &#8220;Debian-&#8221;, on Etch it displays 102.  I had presumed that nss was screwed or something.  But apparently not.  Panic over.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2007/09/11/debian-etch/feed/</wfw:commentRss>
		</item>
		<item>
		<title>pam-pgsql</title>
		<link>http://david.surmise.co.uk/2007/09/04/pam-pgsql/</link>
		<comments>http://david.surmise.co.uk/2007/09/04/pam-pgsql/#comments</comments>
		<pubDate>Tue, 04 Sep 2007 18:16:13 +0000</pubDate>
		<dc:creator>david</dc:creator>
		
		<category>Uncategorized</category>

		<guid isPermaLink="false">http://david.surmise.co.uk/2007/09/04/pam-pgsql/</guid>
		<description><![CDATA[Whilst pam_pgsql logins will work with crypted passwords in your db without the line &#8220;pw_type = crypt&#8221;, passwd (the password changing utility) won&#8217;t.  With the line, passwd does.

]]></description>
			<content:encoded><![CDATA[<p>Whilst pam_pgsql logins will work with crypted passwords in your db without the line &#8220;pw_type = crypt&#8221;, passwd (the password changing utility) won&#8217;t.  With the line, passwd does.
</p>
]]></content:encoded>
			<wfw:commentRss>http://david.surmise.co.uk/2007/09/04/pam-pgsql/feed/</wfw:commentRss>
		</item>
	</channel>
</rss>
