<?xml version="1.0" encoding="utf-8"?>
<feed version="0.3" xmlns="http://purl.org/atom/ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xml:lang="en">
  <title>The Site With the Lamp</title>
  <link rel="alternate" type="text/html" href="/blog/" />
  <modified>2008-09-06T14:29:16Z</modified>
  <tagline></tagline>
  <id>tag:,2008:/2</id>
  <generator url="http://www.movabletype.org/" version="3.33">Movable Type</generator>
  <copyright>Copyright (c) 2008, raditha</copyright>
  <entry>
    <title>Qmail Mail Stuck in the Queue</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001463.html" />
    <modified>2008-09-06T14:29:16Z</modified>
    <issued>2008-09-06T00:48:03+00:00</issued>
    <id>tag:,2008:/2.1463</id>
    <created>2008-09-06T00:48:03Z</created>
    <summary type="text/plain">The other day, staff at a company that I occaisionaly act as a consult for, were complaining about not having any mail. You would think that might make anyone happy but unfortunately they weren&apos;t. When I checked their mail server...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>The other day, staff at a company that I occaisionaly act as a consult for, were complaining about not having any mail. You would think that might make anyone happy but unfortunately they weren't. </p>

<p>When I checked their mail server (qmail + vpopmail + bincimap) which in fact I had setup for them sometime back, I found that thousands of messages were stuck in the queue without being placed into the inboxes. The good thing about qmail is that once the mail reaches the queue, it's safe. It will not go missing unless there is a hard drive failure.</p>

<p>Mails being stuck in the queue is quite common with qmail - usually though it's the remote deliveries that jam up - in this case it was the local deliveries. It turns out that qmail local had got stuck while placing messages in the postmaster's mailbox. With this setup whenever a message is recieved for a non existent user it's delivered to the postmaster (and a bouncer is given to the sender). In reality it wasn't such a smart thing to have done - no one bothers to check those messages and there were so many in Maildir/new that <em>ls</em> also seemed to crash. Even <em>ls | wc</em> didn't return a count.</p>

<p>The folder was completely deleted and qmail was configured to simply bounce those messages without placing a copy in the postmaster box. I also realized that it would be a good idea to use the <a href="/blog/archives/001381.html">good receipt to patch</a> which I had used successfully on one my other mail servers.</p>

<p>The Good Receipt To patch (goodrcptto) results in the mail server rejecting emails to non existent users at the transaction stage - that makes life a lot easier for the mail server and saves quite a lot of bandwidth - it also makes the Double Bounce Trim (doublebouncetrim) patch redundant.</p>

<p><br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Corrupt CF card</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001462.html" />
    <modified>2008-09-04T03:54:19Z</modified>
    <issued>2008-09-04T03:44:17+00:00</issued>
    <id>tag:,2008:/2.1462</id>
    <created>2008-09-04T03:44:17Z</created>
    <summary type="text/plain">Last december, I lost some important images thanks to the Compact Flash card on my Nikon D70S becoming corrupt. That time though, only a small number of files had disappeared from the listing or had gremlins in their file names....</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Last december, I lost some important images thanks to the Compact Flash card on my Nikon D70S becoming <a href="/blog/archives/001318.html">corrupt</a>. That time though, only a small number of files had disappeared from the listing or had gremlins in their file names. That time, I managed to recover most of them by using fsck.vfat on the CF card. So in actual fact only a small number of images were lost.</p>

<p>This time though the card was totally messed up , every single file in the directory listing had gremlins on them. I tried to do a manual fsck.vfat but there were simply too many errors. I soon gave up and re-ran with the -a option. The process seemed to run for an eternity thousands of filenames scrolled through the screen. These automatically renamed filenames are sequentially numbered and when it reached 9999 fsck crashed!</p>

<p>Now this calls for desparate measures. I searched around and came up with a life saving bit of C code that a gentleman by the name of <a href="http://www.cs.washington.edu/homes/oskin/saveimg.html">Mark Oskin</a> had written. The best solution to any problem is usually a simple one - and this bit of code was really simple. I did read about the FAT file system in one of those famous Peter Norton books about 20 years ago but that has long been forgotten. Apparently mark hasn't!</p>

<p>The code worked like a charm and i believe almost all the files were recovered.  It cannot be run on the device itself but fortunately I had already taken a backup with 'dd if=/dev/sdd1 of=cfcard'. Wish I had tried it last time.</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Fredora 9 - Any regrets?</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001461.html" />
    <modified>2008-09-03T13:52:55Z</modified>
    <issued>2008-09-03T13:28:54+00:00</issued>
    <id>tag:,2008:/2.1461</id>
    <created>2008-09-03T13:28:54Z</created>
    <summary type="text/plain">When I moved from Fedora to Debian it wasn&apos;t long before I started regretting that decision. Now I am back on Fedora 9 but this time there are no regrets even though Fedora is as annoying as ever. One of...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>When I moved from Fedora to Debian it wasn't long before I started regretting that decision. Now I am back on Fedora 9 but this time there are no regrets even though Fedora is as annoying as ever.</p>

<p>One of the first annoyances is that Firefox 3 always starts in offline mode. I didn't even know that modern day browsers had an offline mode. I do recall seeing something like that in good old Netscape Navigator 3 or something.</p>

<div align="center"><a href="/blog/images/ff3-offline.gif"><img src="/blog/images/ff3-offline-t.gif" border="0"></a></div>

<p>I am angry but  will live with it because installing FF 3 on Debian broke the whole system!  There was another annoyance but it wasn't so serious.</p>

<div align="center"><img src="/blog/images/ff-assert.gif"></div>

<p>It popped up when I clicked on the search box next to the tool bar , it disappeared on it's own accord after a while.<br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Dialog Broadband - The Court Jester.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001460.html" />
    <modified>2008-09-02T16:44:11Z</modified>
    <issued>2008-09-02T16:26:25+00:00</issued>
    <id>tag:,2008:/2.1460</id>
    <created>2008-09-02T16:26:25Z</created>
    <summary type="text/plain">Dialog Telekom, once the #1 teleco in the country has been reduced to a sad state, they are now occupying the role of court jester , making mistakes at every opportunity adapting the wrong technology, hiring idiots to man their...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Dialog Telekom, once the #1 teleco in the country has been reduced to a sad state, they are now occupying the role of court jester , making mistakes at every opportunity adapting the wrong technology, hiring idiots to man their call centers and back rooms. Profits have been steadily declining and it will not be long before Dialog is in the red. Part of that is due to Dialog Broadband eating up a huge chunk of the parent companies revenues and mauling it's reputation in the process.</p>

<p>when I first wanted to obtain a dialog broadband connection they dragged their heels for nearly a month and the connection was setup only after I wrote to Dr Hans. Speeds were poor and disconnections frequent but I put up with it then came a shock:  Astonishingly my bill payment for March (Rs 12,500) was never credited to my account. This meant it has always been in arrears. Four complaints were made no action was taken</p>

<p>Resigned to the fact that this money might have ended up in someone's pocket, I wanted to disconnect at that time but as a loyal Dialog customer I thought to give Dialog BB one last chance. However to my utter disappointment found that my WiMax line has been  disconnected last thursday.</p>

<p>I immidiately email Dr Hans and my line was reactivated, then someone from Dialog Broadband called to appologise, then less than 24 hours later the line was cut off. Well it might have been one of those frequent outages that Dialog Broadband has made it's trademark, either way it doesn't really matter anymore. I already have a lanka bell wimax connection (which I have been using as a backup because of the frequent failures of the<br />
Dialog link). That will be my primary link from now on.</p>

<p>Lanka Bell Wimax isn't without it's problems - in fact some of them are pretty serious (more on that later) but at least you can't fault them for not trying to serve. On the other hand with Dialog Broadband it's always the customer's fault.</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Swapping Ginger for Chilly</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001458.html" />
    <modified>2008-09-02T00:47:34Z</modified>
    <issued>2008-09-01T16:18:07+00:00</issued>
    <id>tag:,2008:/2.1458</id>
    <created>2008-09-01T16:18:07Z</created>
    <summary type="text/plain">I moved away from Fedora a few weeks ago after becoming increasingly annoyed with it&apos;s messy upgrade process. The distro that I chose was Debian. Here in Sri Lanka we have adage that goes like &apos;giving up ginger to get...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>I moved away from Fedora a few weeks ago after becoming increasingly annoyed with it's messy upgrade process. The distro that I chose was Debian. Here in Sri Lanka we have adage that goes like 'giving up ginger to get chilly'. The saying apparently first came into use a few centuries ago when Portuge invaders were chased away and the Dutch took their place.</p>

<p>In English, you say, out of the frying pan into the fire. If Fedora is almost as bad as windows and <a href="/blog/archives/001449.html">Debian is just as bad</a>. Thus someone moving from Fedora to Debian can claim to have fallen from the frying pan into the fire.</p>

<p>I've been annoyed with it for quite a while but the last straw was when my <a href="/blog/archives/001456.html">Lanka Bell wimax connection crawled to a halt</a>. Amazingly it seemed to be pretty quick on windows. Obviously there is no way that Vista can be better than linux at anything so that means there is some configuration issue. Then I tried my media server which still has Fedora 9 and that was as fast as the Vista lappy I tested the connection with.</p>

<p>It took me quite a while to figure out the cause of the problem - there is a major misconfiguration on the part of Lanka Bell. More on that later - i want to give them time to fix it before making it public. Suffice to say, i was being routed through a network that I shouldn't have been sent through. The trouble is I never could figure out how to make it stop on debian (yep, that really is my fault - I never got  the hang of the /etc/ layout in debian). However fixing it on my fedora box and hardening the firewall was a breeze. </p>

<p>So the logical decision was to remove Debian and put in Fedora once again. Guess what, doing a fresh installation is a lot easier than an upgrade.<br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>JoPPP</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001457.html" />
    <modified>2008-09-01T16:41:22Z</modified>
    <issued>2008-08-30T13:19:40+00:00</issued>
    <id>tag:,2008:/2.1457</id>
    <created>2008-08-30T13:19:40Z</created>
    <summary type="text/plain">We know of PPPoE and POE which are crucial to Wimax implementations that use Alvarian outdoor antennae, but have you ever heard of JoPPP? that&apos;s short for Junk over PPP, which is what is passed back and forth over the...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>We know of PPPoE and POE which are crucial to Wimax implementations that use Alvarian outdoor antennae, but have you ever heard of JoPPP? that's short for Junk over PPP, which is what is passed back and forth over the PPP link when using Lanka Bell WiMax. No wonder it's <a href="/blog/archives/001456.html">crawling</a></p>


<div style="width:600px; height:200px; overflow:auto"><pre>
listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes
19:25:38.846663 IP 192.168.215.138.netbios-dgm > 192.168.215.143.netbios-dgm: NBT UDP PACKET(138)
19:25:48.941528 IP 192.168.215.138.netbios-ns > 192.168.215.143.netbios-ns: NBT UDP PACKET(137): QUERY; REQUEST; BROADCAST
19:25:38.847592 PPPoE  [ses 0x5d5] IP 119.235.2.76.32992 > ns1.slic.lankabellnet.com.domain:  64753+ PTR? 143.215.168.192.in-addr.arpa. (46)
19:25:38.886044 PPPoE  [ses 0x5d5] IP ns1.slic.lankabellnet.com.domain > 119.235.2.76.32992:  64753 NXDomain* 0/1/0 (143)
19:25:39.585971 IP 192.168.215.138.netbios-ns > 192.168.215.143.netbios-ns: NBT UDP PACKET(137): QUERY; REQUEST; BROADCAST
19:25:39.765919 arp who-has 192.168.101.1 tell 192.168.101.58</pre>
</div>

<p>But that's not the real reason for the slowness. Here is the real reason: </p>
<div style="width:600px; height:200px; overflow:auto">
<pre>
op,timestamp 1978135924 17416466>
19:02:30.483407 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 42000 win 501 <nop,nop,timestamp 17416650 1978135809>
19:02:30.608415 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 42000:43400(1400) ack 1 win 14 <nop,nop,timestamp 1978136049 17416497>
19:02:30.733796 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 43400:44800(1400) ack 1 win 14 <nop,nop,timestamp 1978136174 17416528>
19:02:30.733876 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 44800 win 501 <nop,nop,timestamp 17416713 1978136049>
19:02:30.765116 PPPoE  [ses 0x5d5] IP 119.235.2.76.32977 > ns1.slic.lankabellnet.com.domain:  31605+[|domain]
19:02:30.853791 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 44800:46200(1400) ack 1 win 14 <nop,nop,timestamp 1978136334 17416568>
19:02:30.978106 PPPoE  [ses 0x5d5] IP ns1.slic.lankabellnet.com.domain > 119.235.2.76.32977:  31605 1/1/0 (131)
19:02:30.978337 PPPoE  [ses 0x5d5] IP 119.235.2.76.32977 > ns1.slic.lankabellnet.com.domain:  22516+[|domain]
19:02:30.993324 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 46200:47600(1400) ack 1 win 14 <nop,nop,timestamp 1978136499 17416609>
19:02:30.993403 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 47600 win 501 <nop,nop,timestamp 17416778 1978136334>
19:02:31.113364 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 47600:49000(1400) ack 1 win 14 <nop,nop,timestamp 1978136664 17416650>
19:02:31.238419 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 49000:50400(1400) ack 1 win 14 <nop,nop,timestamp 1978136664 17416650>
19:02:31.238496 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 50400 win 501 <nop,nop,timestamp 17416839 1978136664>
19:02:31.363751 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 50400:51800(1400) ack 1 win 14 <nop,nop,timestamp 1978136664 17416650>
19:02:31.483611 PPPoE  [ses 0x5d5] IP truncated-ip - 256 bytes missing! ns1.slic.lankabellnet.com.domain > 119.235.2.76.32977:  22516 5/7/7[|domain]
19:02:31.484040 PPPoE  [ses 0x5d5] IP 119.235.2.76.42697 > ik-in-f101.google.com.www: S 354522469:354522469(0) win 5808 <mss 1452,sackOK,timestamp 17416901 0,nop,wscale 7>
19:02:31.518226 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 51800:53200(1400) ack 1 win 14 <nop,nop,timestamp 1978136914 17416713>
19:02:31.518301 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 53200 win 501 <nop,nop,timestamp 17416909 1978136664>
19:02:31.643226 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 53200:54600(1400) ack 1 win 14 <nop,nop,timestamp 1978136914 17416713>
19:02:31.759264 PPPoE  [ses 0x5d5] IP 119.235.2.76.35277 > ug-in-f164.google.com.www: F 310474103:310474103(0) ack 1744444467 win 135 <nop,nop,timestamp 17416969 1243138488>
19:02:31.759330 PPPoE  [ses 0x5d5] IP 119.235.2.76.35279 > ug-in-f164.google.com.www: F 1:1(0) ack 1 win 54 <nop,nop,timestamp 17416969 3704698737>
19:02:31.768711 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 54600:56000(1400) ack 1 win 14 <nop,nop,timestamp 1978137174 17416778>
19:02:31.768789 PPPoE  [ses 0x5d5] IP 119.235.2.76.50493 > nms01.hns.net.in.www: . ack 56000 win 501 <nop,nop,timestamp 17416972 1978136914>
19:02:31.888696 PPPoE  [ses 0x5d5] IP truncated-ip - 1280 bytes missing! nms01.hns.net.in.www > 119.235.2.76.50493: . 56000:57400(1400) ack 1 win 14 <nop,nop,timestamp 1978137174 17416778>
19:02:31.896932 PPPoE  [ses 0x5d5] IP 119.235.2.76.32977 > ns1.s
</div>

<p> Now to try to figure out what's causing it. They at Lanka bell simply don't know.</p>
]]>
      
    </content>
  </entry>
  <entry>
    <title>Lanka Bell Crawls to a halt.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001456.html" />
    <modified>2008-08-28T12:50:14Z</modified>
    <issued>2008-08-28T12:45:02+00:00</issued>
    <id>tag:,2008:/2.1456</id>
    <created>2008-08-28T12:45:02Z</created>
    <summary type="text/plain">Lanka Bell Wimax has been slower than Dialog but last evening it disappeared all together. Fell off the face of the earth. A connection would be established but you cannot use it, nothing can be pinged, sshed, telneted or fingered....</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Lanka Bell Wimax has been slower than Dialog but last evening it disappeared all together. Fell off the face of the earth. A connection would be established but you cannot use it, nothing can be pinged, sshed, telneted or fingered.</p>

<div align="center"><img src="http://www.speedtest.net/result/314811357.png"></div>

<p>Then normal service was restored last this morning. Assuming that 80 kbps download speed can be called normal for a 2mbps line. Even that was only for a brief period. It disappeared all too soon.</p>

<p>But their response was pretty good, more on that later.<br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>PHP5 XSLT on Debian.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001450.html" />
    <modified>2008-08-15T03:17:03Z</modified>
    <issued>2008-08-15T06:59:24+00:00</issued>
    <id>tag:,2008:/2.1450</id>
    <created>2008-08-15T06:59:24Z</created>
    <summary type="text/plain">Another side effect of using the unstable repo to install FF3 was that it messed up my apache. Debian does advice against using the unstable repo. That means it really isn&apos;t suited for anyone who wants to use the latest...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Another side effect of using the unstable repo to install FF3 was that it messed up my apache. Debian does advice against using the unstable repo. That means it really isn't suited for anyone who wants to use the latest version of any software. There is a long gap between Debian releases and at that rate we will have to wait forever to see FF3 make it's way into the stable repository. That means there is no choice but to go with the unstable option.</p>

<p>I sorted out the apache issue by uninstalling and reinstalling it. Things went smoothly for a couple of days, until I needed to run an old PHP script. The script is nothing special it just does a bit of XSL transformation. It makes use of the PHP XSLT extension. The trouble is that is not part of debian!!</p>

<p>There is a PHP4-XSLT module and there is a PHP5-XSL module but no PHP5-XSLT how anoying. Once upon a long ago, I quite used to enjoy compiling things from source but now I am too old forit. Nevertheless it seems that I have no choice but to manually compile PHP5. I didn't get far - the PHP configuration script came up with this ridiculous statement</p>

<div style="margin-left:15px; color:#330000"><pre>
checking whether the C compiler (gcc  ) works... no
configure: error: installation or configuration problem: C compiler cannot create executables.
</pre></div>

<p>Turns out this is because libc6-dev is not installed. That is easily rectified but while installing libc6-dev, apt decided to uninstall xorg? how ridiculous is that. Looks like I will be limited to the terminal next time I reboot the machine.</p>

<p>But when I tried to compile PHP again, I ran into the same error once more. Close inspection reveals that APT had choked out with</p>

<div style="margin-left:15px; color:#330000"><pre>Failed to fetch http://ftp2.jp.debian.org/debian/pool/main/g/glibc/libc6-dev_2.7-12_i386.deb</div>

<p>How anoying is that? when Fedora 10 is released, I am definitely switching back.</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Dialog vs Lanka Bell</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001452.html" />
    <modified>2008-08-15T03:18:36Z</modified>
    <issued>2008-08-09T02:19:11+00:00</issued>
    <id>tag:,2008:/2.1452</id>
    <created>2008-08-09T02:19:11Z</created>
    <summary type="text/plain">I am so fed up with Dialog Broadband&apos;s pathetic service that I decided to switch to Lanka Bell instead. Lanka Bell has recently announced WiMax services and already they have signed up more than 250 customers. Dialog on the other...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>I am so fed up with <a href="http://www.raditha.com/blog/archives/001359.html">Dialog Broadband's pathetic service</a> that I decided to switch to Lanka Bell instead. Lanka Bell has recently announced WiMax services and already they have signed up more than 250 customers. Dialog on the other hand has less than 2000 inspite of being the first service provider to come up with the service nearly an year ago.</p>

<p>I haven't yet disconnected the Dialog link and that gave me an opportunity to compare the speeds of the two providers. I was disapointed to find that Lanka Bell is a bit slower.</p>

<table border="0" cellpadding="5">
<tr><td align="center">Lanka Bell</td><td align="center">Dialog</td></tr>
<tr><td align="center"><a href="http://www.speedtest.net"><img src="http://www.speedtest.net/result/306767855.png"></a></td><td align="center"><a href="http://www.speedtest.net"><img src="http://www.speedtest.net/result/306769174.png"></a></td></tr>
</table>

<p>Two tests were carried out one with speedtest.net and the other with dslreports.com both reports show that Lanka Bell is slightly slower (they are both 2mbps lines)</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Debian worse than Fedora?</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001449.html" />
    <modified>2008-08-08T03:37:20Z</modified>
    <issued>2008-08-08T06:43:20+00:00</issued>
    <id>tag:,2008:/2.1449</id>
    <created>2008-08-08T06:43:20Z</created>
    <summary type="text/plain">Installing Firefox 3 (Iceweasel 3, if you want to call it that) on Debian turned out to be a very painfull task. Memories fade but I don&apos;t recall things being that difficult on Fedora. The trouble didn&apos;t end when the...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Installing Firefox 3 (Iceweasel 3, if you want to call it that) on Debian turned out to be a very painfull task. Memories fade but I don't recall things being that difficult on Fedora. The trouble didn't end when the installation was completed, in fact, it was just the beginning. I am beginning to regret dumping <a href="/blog/archives/001425.html">Fedora in favor of debian</a>.</p>

<p>After FF was put in I found that I couldn't click on many of the shortcuts to folders on my desktop. Well, I could click on it but the folder wouldn't open, a dialog would pop up asking me with what I wanted to open that 'file'. Crazy, it seems that during the <a href="/blog/archives/001441.html">FF3 installation</a>, apt had removed both nautilus and konqueror!!!</p>

<p>I had to go back and reinstall them and a few more besides which had been pulled out by apt. On top of that whenever I try to upgrade an existing package or a new package I keep getting a warning that the package's authenticity has not been verified.</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Macbook Blank Screen solved.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001448.html" />
    <modified>2008-08-07T06:42:37Z</modified>
    <issued>2008-08-07T10:47:05+00:00</issued>
    <id>tag:,2008:/2.1448</id>
    <created>2008-08-07T10:47:05Z</created>
    <summary type="text/plain">The other day I ran into a problem where the screen on the macbook suddenly appeared to power down. It was just black. It stayed the same even after a restart. It appears as if the computer is in sleep...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>The other day I ran into a problem where the <a href="/blog/archives/001443.html">screen on the macbook suddenly appeared to power down</a>. It was just black. It stayed the same even after a restart. It appears as if the computer is in sleep mode, but In actual fact the screen is live - but the back light is switched off. It's as if the energy saver has kicked into action. </p>

<p>I tried a lot of <a href="/blog/archives/001444.html">different remedies</a> posted in different forums, ranging from changing the sleep mode to resetting the SMC and RAM. None of it seemed to work. Firmware updates cannot be applied since the macbook already had the latest version of the firmware. The obvious next step is to roll back the firmware. This is where procrastination and being lazy really helped. I was playing around with the setting on the macbook over a vncviewer when I suddenly thought of switching off power saving. </p>

<p>It turns out that in order to solve the blank screen problem, all you need to do is to switch off the 'Reduce brightness of the built-in display when using this power source' and the 'Automatically reduce the brightness of the display before sleep'. And your LCD springs back to life.</p>

<p>Of course to do this, you need to have VNC. If you don't have VNC you can simply plug in an external monitor (but you need a special adapter for that because there isn't a VGA socket on the macbook). </p>

<p><br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Apple Sucks</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001444.html" />
    <modified>2008-08-06T02:04:23Z</modified>
    <issued>2008-08-06T07:40:34+00:00</issued>
    <id>tag:,2008:/2.1444</id>
    <created>2008-08-06T07:40:34Z</created>
    <summary type="text/plain">Sure windows sucks but apple seems to suck even more. The back-light on the Macbook LCD shutting down (gives the impression that the computer is asleep) will not go away. There are thousands of &apos;me too&apos; posts on the web...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Sure windows sucks but apple seems to suck even more. The back-light on the Macbook LCD shutting down (gives the impression that the computer is asleep) will not go away. There are thousands of 'me too' posts on the web out there. Apple doesn't seem to be able to or want to come up with a solution. There are several solutions proposed by other users. Some of them seem to have worked for some but not for others.</p>

<p>The most obvious thing to do whenever there is a problem with a macbook or a macbook pro is to reset the PRAM (hold down Comman - Option - P - R while rebooting). That didn't solve it. The next is to try to reset the SMC (Unplug power, take out battery, hold down power button for 5seconds , start up again) is not a solution either.</p>

<p>One proposed solution is to switch off the 'safe sleep' mode. This is done by deleting the /private/var/vm/sleepimage file and the /system/library/extensions.mkext file along with the execution of these two commands:</p>

<div style="margin-left:25px; color:#000033">
sudo pmset -a hibernatemode 0 <br/>
sudo nvram "use-nvramrc?"=false
</div>

<p>That unfortunately didn't do anygood either. So it was time to look at suggestions such as updating the firmware and SMC unfortunately both are upto date. The next option is to download the latest OS X update. That happens to be a security update but Apple is notorious more so than Windows when it comes to fessing up about bugs in their software. It's quite possible that their Security updates contain bug fixes to other issues and other updates patch the system to close up security holes.  After downloading a 140Mb file over a <a href="http://www.raditha.com/blog/archives/001296.html">slow dialog broadband connection</a>, I can tell you that the issue is still there.<br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Macbook blank screen.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001443.html" />
    <modified>2008-08-04T05:22:17Z</modified>
    <issued>2008-08-04T05:05:42+00:00</issued>
    <id>tag:,2008:/2.1443</id>
    <created>2008-08-04T05:05:42Z</created>
    <summary type="text/plain">A few months earlier, my father&apos;s Macbook refused to boot up, it wouldn&apos;t produce anything except a blank screen. It wasn&apos;t something that could be fixed by the usual remedies such as resetting the PVRAM. It was a major hardware...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>A few months earlier,  my father's Macbook refused to boot up, it wouldn't produce anything except a blank screen. It wasn't something that could be fixed by the usual remedies such as resetting the PVRAM. It was a <a href="/blog/archives/001302.html">major hardware problem</a> that was covered by the warranty but getting it fixed by the local dealer turned out be <a href="http://www.webquarry.com/~raditha/blog/archives/001303.html">astonishingly difficult</a>.</p>

<p>Now something similar has come up again. The screen had suddenly gone blank just as it would do when the machine goes to sleep. If you close the cover (LCD) , reopen it and hit any key the screen would light up momentarily and go black once again. Increasing the brightness didn't do any good either. The screen isn't completely gone. Apparently it's only the back light that's gone. You can dimly see an outline of what's on the desktop.</p>

<p>This too is a known issue, there are too possible causes. One is a buggy firmware the other buggy hardware. A week earlier Bluetooth on this same machine had been switched off all of a sudden. It had completely disappeared from the menubar at the top as well as system preferences. Fortunately that was cured by a simple restart. This time though restart doesn't get you anywhere.</p>

<p>The first thing to do is to make a backup. This machine get's it's IP using DHCP from the airport. Since it's nearly impossible to keep the screen live to open terminal , run ifconfig and see it's output. The IP could be easily determined by nmap. Then I SSHed into the box and started a backup on it (rsync). It's still running (22GB of data in the home folder). Not going to try anything till the backup is complete.</p>

<p>I could have had the backup completed in minutes by using the gigabit ethernet available here, but that would mean airport has to be turned off and for that you need the screen again. How I wish I had vncserver on it.</p>

<p><br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>Firefox 3 on Debian.</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001441.html" />
    <modified>2008-07-29T13:22:53Z</modified>
    <issued>2008-07-29T11:31:42+00:00</issued>
    <id>tag:,2008:/2.1441</id>
    <created>2008-07-29T11:31:42Z</created>
    <summary type="text/plain">All right, All right, the title should be Iceweasel on Debian. Call a rose by any other name and it would be just as hard to install on Debian. I think I am beginning to regret giving up Fedora. Iceweasel...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>All right, All right, the title should be Iceweasel on Debian. Call a rose by any other name and it would be just as hard to install on Debian. I think I am beginning to regret giving up Fedora.</p>

<p>Iceweasel 3.0 hasn't made it's way into the debian stable repositories that wouldn't really matter if it was possible to install the precompiled binary from <a href="http://www.mozilla.com">mozilla.com</a> unfortunately things don't quite  work out like that because I got an error with the installer (unfortunately forgot to note it down)</p>

<p>So if you are keen to use firefox 3 the only option is to switch to the unstable repository. The documentation advices against mixing the two but sometimes there is no choice.</p>

<p>Adding the unstable repostory to synaptic (which is really adding it to apt) is easier said than done. When you reload after adding the new repo, it errors out.</p>

<div style="margin-left:30px; color:#330000"><pre>E:Dynamic MMap ran out of room
E:Dynamic MMap ran out of room
E:Error occured while processing scim-modules-socket (NewVersion1)
E:Problem with MergeList /var/lib/apt/lists/
http://us.debian.org_debian_debian_dist_sid_main_binary-1386-Packages
E:The package lists or status file could not be parsed or opened.
</pre></div>

<p>The web is full of pundits and some of them have commented that this error is because there are two many repositories. How rediculous is that? fortunately there are some really knowledgable people out there too.  One of them suggested that this is because the default <a href="https://bugs.launchpad.net/ubuntu/+source/apt/+bug/24626">space allocation for the cache is too low</a>.</p>

<p>I reckoned it might be something like this when I first so the error and initially thought of changing the ulimit (this is what needed to be done with k3b on Fedora), but it turns out that on debian ulimit by default is unlimited.</p>

<p>The solution is to add something like the following to the apt configuration files:</p>

<div style="margin-left:30px; color:#000033">APT::Cache-Limit 125829120</div>

<p>The number of course being the new limit. You can choose something appropriate.</p>

<p><br />
</p>]]>
      
    </content>
  </entry>
  <entry>
    <title>sshfs debian</title>
    <link rel="alternate" type="text/html" href="/blog/archives/001438.html" />
    <modified>2008-07-25T05:14:19Z</modified>
    <issued>2008-07-25T07:12:47+00:00</issued>
    <id>tag:,2008:/2.1438</id>
    <created>2008-07-25T07:12:47Z</created>
    <summary type="text/plain">Thought I would play around with SSHFS, which allows you to mount a folder on a remote computer locally using SSH. Obviously that would be much more convenient than using SFTP and would be a hell-ova-lot safer than using No...</summary>
    <author>
      <name>raditha</name>
      <url>http://www.raditha.com</url>
      <email>e4c5@raditha.com</email>
    </author>
    
    <content type="text/html" mode="escaped" xml:lang="en" xml:base="/blog/">
      <![CDATA[<p>Thought I would play around with SSHFS, which allows you to mount a folder on a remote computer locally using SSH. Obviously that would be much more convenient than using SFTP and would be a hell-ova-lot safer than using No File System (NFS). Getting started with it on Debian, though wasn't quite so easy.</p>

<p>When trying to use sshfs to mount a remote directory, i get this error:<br />
fuse: failed to exec fusermount: Permission denied</p>

<p>Apparently it's because I am not a member of the fuser group. I promptly added myself. Still no luck. logged out of the terminal and opened a new one, to see if getting the profile re-read will do the trick. It does but you run into another error.</p>

<p>fusermount: failed to open /dev/fuse: No such file or directory</p>

<p>Then I heard that fuse module needs to be loaded. So I tried the usual modprobe:</p>

<p>sudo modprobe fuse</p>

<p>I still get the same error and I still don't see the /dev/fuse file. I reckoned these profiles on debian are badly messed up. Certainly it seems to completely ignore my /etc/profile and .bash_profile files. So I changed into root once again with su and then gave the modprobe command from there. Lo and Behold. The darned module was loaded and the remote file system was mounted.</p>

<p>Finally we are getting there. <a href="/blog/archives/001429.html">Debian sure has strange ways of doing things</a>.<br />
</p>]]>
      
    </content>
  </entry>

</feed>