<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments for /dev/root</title>
	<atom:link href="http://devroot.org/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://devroot.org</link>
	<description>Rants of a grumpy sysadmin living in Switzerland.</description>
	<pubDate>Mon, 01 Dec 2008 19:57:45 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.3</generator>
		<item>
		<title>Comment on Optimize network throughput on your Apple TV by lynxman</title>
		<link>http://devroot.org/2007/05/01/optimize-network-throughput-on-your-appletv/#comment-71</link>
		<dc:creator>lynxman</dc:creator>
		<pubDate>Thu, 13 Nov 2008 12:36:56 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/05/01/optimize-network-throughput-with-appletv/#comment-71</guid>
		<description>@Graham Booker

Oh thanks for the correction, I'll modify the article accordingly :)

@DJLM It should be okay to increase them a bit, but I found that to have more dangers than gains, some of this kernel parameters don't play well when increased beyond this limits so please make sure that you have a way to go back to your previous settings!</description>
		<content:encoded><![CDATA[<p>@Graham Booker</p>
<p>Oh thanks for the correction, I&#8217;ll modify the article accordingly :)</p>
<p>@DJLM It should be okay to increase them a bit, but I found that to have more dangers than gains, some of this kernel parameters don&#8217;t play well when increased beyond this limits so please make sure that you have a way to go back to your previous settings!</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Swisscom roaming GPRS data fraud by David</title>
		<link>http://devroot.org/2007/12/18/swisscom-roaming-gprs-data-fraud/#comment-69</link>
		<dc:creator>David</dc:creator>
		<pubDate>Sat, 02 Aug 2008 17:55:09 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/12/18/swisscom-roaming-gprs-data-fraud/#comment-69</guid>
		<description>Thanks guys for those explanations...

I was wondering why my Swisscom online bill showed that i used "30kb" many times a day, even if i didn't use any webservice at that time.

Unfortunately, most of the time my Carrier changes from Swisscom to SFR (France) as i live on the edge of Leman lake, which is close to the frontier.

Then i believe that my iPhone is just "checking" if GPRS works when i get connected to the new carrier. And then i'm billed 30KB :-/

I'm gonna check if it's possible for the iPhone to "stick" just with 1 unique carrier. Otherwise i'll proboably just switch off GPRS when i don't use it... while fighting with Swisscom helpdesk.</description>
		<content:encoded><![CDATA[<p>Thanks guys for those explanations&#8230;</p>
<p>I was wondering why my Swisscom online bill showed that i used &#8220;30kb&#8221; many times a day, even if i didn&#8217;t use any webservice at that time.</p>
<p>Unfortunately, most of the time my Carrier changes from Swisscom to SFR (France) as i live on the edge of Leman lake, which is close to the frontier.</p>
<p>Then i believe that my iPhone is just &#8220;checking&#8221; if GPRS works when i get connected to the new carrier. And then i&#8217;m billed 30KB :-/</p>
<p>I&#8217;m gonna check if it&#8217;s possible for the iPhone to &#8220;stick&#8221; just with 1 unique carrier. Otherwise i&#8217;ll proboably just switch off GPRS when i don&#8217;t use it&#8230; while fighting with Swisscom helpdesk.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on IBM xServer 305 on CentOS 4 - Kernel incompatibilities by camelia</title>
		<link>http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-68</link>
		<dc:creator>camelia</dc:creator>
		<pubDate>Sat, 17 May 2008 01:57:14 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-68</guid>
		<description>Hello

I have sentos 5 kernal version 2.6.18-53.1.14.el5 
i added noapic and will see if it will works!

Thanks</description>
		<content:encoded><![CDATA[<p>Hello</p>
<p>I have sentos 5 kernal version 2.6.18-53.1.14.el5<br />
i added noapic and will see if it will works!</p>
<p>Thanks</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Optimize network throughput on your Apple TV by Graham Booker</title>
		<link>http://devroot.org/2007/05/01/optimize-network-throughput-on-your-appletv/#comment-67</link>
		<dc:creator>Graham Booker</dc:creator>
		<pubDate>Wed, 30 Apr 2008 03:05:40 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/05/01/optimize-network-throughput-with-appletv/#comment-67</guid>
		<description>You forgot to put the "sudo" in front of the echo commands. An alternative to the following:

sudo echo "kern.ipc.somaxconn=512\ 
net.inet.tcp.mssdflt=1460\ 
net.inet.tcp.sendspace=98304\
 net.inet.tcp.recvspace=98304\ 
kern.ipc.maxsockbuf=1048576\
 net.inet.udp.maxdgram=57344\ 
net.inet.udp.recvspace=42080\ net.local.stream.sendspace=98304\ net.local.stream.recvspace=98304\ 
net.inet.tcp.delayed_ack=0\ 
net.inet.tcp.rfc1323=1\ 
net.inet.tcp.rfc1644=1\ 
net.inet.tcp.newreno=1" &#62; /etc/sysctl.conf</description>
		<content:encoded><![CDATA[<p>You forgot to put the &#8220;sudo&#8221; in front of the echo commands. An alternative to the following:</p>
<p>sudo echo &#8220;kern.ipc.somaxconn=512\ <br />
net.inet.tcp.mssdflt=1460\ <br />
net.inet.tcp.sendspace=98304\<br />
 net.inet.tcp.recvspace=98304\ <br />
kern.ipc.maxsockbuf=1048576\<br />
 net.inet.udp.maxdgram=57344\ <br />
net.inet.udp.recvspace=42080\ net.local.stream.sendspace=98304\ net.local.stream.recvspace=98304\ <br />
net.inet.tcp.delayed_ack=0\ <br />
net.inet.tcp.rfc1323=1\ <br />
net.inet.tcp.rfc1644=1\ <br />
net.inet.tcp.newreno=1&#8243; &gt; /etc/sysctl.conf</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Configuring samba server in Mac OS X Leopard by Lou Jarvis</title>
		<link>http://devroot.org/2007/10/26/configuring-samba-server-in-mac-os-x-leopard/#comment-66</link>
		<dc:creator>Lou Jarvis</dc:creator>
		<pubDate>Fri, 25 Apr 2008 18:43:32 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/10/26/configuring-samba-server-in-mac-os-x-leopard/#comment-66</guid>
		<description>Are the changes you are making to increase the buffer in RAM at the AppleTV device?  I think the AppleTV device buffers the content which is sent over the LAN through its RAM and video RAM which allows it to play back more smoothly and isolates it from network interference....is this correct?</description>
		<content:encoded><![CDATA[<p>Are the changes you are making to increase the buffer in RAM at the AppleTV device?  I think the AppleTV device buffers the content which is sent over the LAN through its RAM and video RAM which allows it to play back more smoothly and isolates it from network interference&#8230;.is this correct?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on IBM xServer 305 on CentOS 4 - Kernel incompatibilities by William</title>
		<link>http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-65</link>
		<dc:creator>William</dc:creator>
		<pubDate>Wed, 26 Mar 2008 00:25:24 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-65</guid>
		<description>And I would like to add this is an x305 :-)</description>
		<content:encoded><![CDATA[<p>And I would like to add this is an x305 :-)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on IBM xServer 305 on CentOS 4 - Kernel incompatibilities by William</title>
		<link>http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-64</link>
		<dc:creator>William</dc:creator>
		<pubDate>Wed, 26 Mar 2008 00:24:57 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/02/05/ibm-xserver-305-on-centos-4-kernel-incompatibilities/#comment-64</guid>
		<description>Is this still in issue with kernal version 2.6.18-53.1.14.el5 ?

I am having this magically issue of after 24hours, system becomes un-responsive, with no messages in /var/log/messages or /var/log/dmesg</description>
		<content:encoded><![CDATA[<p>Is this still in issue with kernal version 2.6.18-53.1.14.el5 ?</p>
<p>I am having this magically issue of after 24hours, system becomes un-responsive, with no messages in /var/log/messages or /var/log/dmesg</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Optimize network throughput on your Apple TV by DJLM</title>
		<link>http://devroot.org/2007/05/01/optimize-network-throughput-on-your-appletv/#comment-63</link>
		<dc:creator>DJLM</dc:creator>
		<pubDate>Sat, 08 Mar 2008 20:04:34 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/05/01/optimize-network-throughput-with-appletv/#comment-63</guid>
		<description>Just wondering, are any of these values able to be safely increased any further?  I'm down to maybe one skip per video I watch, so it wouldn't need to be much of an increase?

Thanks!</description>
		<content:encoded><![CDATA[<p>Just wondering, are any of these values able to be safely increased any further?  I&#8217;m down to maybe one skip per video I watch, so it wouldn&#8217;t need to be much of an increase?</p>
<p>Thanks!</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Optimize network throughput on your Apple TV by DJLM</title>
		<link>http://devroot.org/2007/05/01/optimize-network-throughput-on-your-appletv/#comment-62</link>
		<dc:creator>DJLM</dc:creator>
		<pubDate>Thu, 06 Mar 2008 12:10:31 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/05/01/optimize-network-throughput-with-appletv/#comment-62</guid>
		<description>Thank you so much, I have been using the TV in exactly the same way, AFP files played over it's Airport interface, and suffering the same annoying issues...  You read my mind exactly as to what I suspected the problem was but you saved me looking up and figuring out what to set the variables to!

Many thanks!!!</description>
		<content:encoded><![CDATA[<p>Thank you so much, I have been using the TV in exactly the same way, AFP files played over it&#8217;s Airport interface, and suffering the same annoying issues&#8230;  You read my mind exactly as to what I suspected the problem was but you saved me looking up and figuring out what to set the variables to!</p>
<p>Many thanks!!!</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Swisscom roaming GPRS data fraud by Joduba</title>
		<link>http://devroot.org/2007/12/18/swisscom-roaming-gprs-data-fraud/#comment-51</link>
		<dc:creator>Joduba</dc:creator>
		<pubDate>Tue, 18 Dec 2007 18:21:34 +0000</pubDate>
		<guid isPermaLink="false">http://devroot.org/2007/12/18/swisscom-roaming-gprs-data-fraud/#comment-51</guid>
		<description>Hi Marc,

I had a similar problem, but in my case with ATT US.

After a long and dummy discussion with both help desk and technical department, the main conclusions are:

1) My phone is like shit
2) My phone is not compatible with ATT US
3) It's a ATT problem

It's clear that last two points doesn't make sense for an iphone bought in 5th Ave ;-)

Anyway, the main problem is in the concept it self... I can accept to pay 30Kb per "session" but then I did the following question...

"If in 1 minute or less my phone opens 3 connections, I will be charged 30 or 90 Kb ?"

And winner is 90Kb, so...

GPRS is a "data switching" protocol, there is no "circuit" switching". So the protocol is designed to work "connecting" for every single packet.

Is like to say a packet minimum size is 30Kb...

So this rule is incorrect, as they are based on "common usage", not in the protocol. They have to invoice packets, or bytes, but not connections as connection is not intrinsic of the protocol and can be done in different ways...

In my point of view, is an abuse of dominant position or a cartel situation and governments has to act and clearly say this is ILLEGAL! (or IMMORAL ).

As end users we do not have any "claim" method...

I can't understand why Europeans or world GSM citizens we do not unite against abusive roaming practices.</description>
		<content:encoded><![CDATA[<p>Hi Marc,</p>
<p>I had a similar problem, but in my case with ATT US.</p>
<p>After a long and dummy discussion with both help desk and technical department, the main conclusions are:</p>
<p>1) My phone is like shit<br />
2) My phone is not compatible with ATT US<br />
3) It&#8217;s a ATT problem</p>
<p>It&#8217;s clear that last two points doesn&#8217;t make sense for an iphone bought in 5th Ave ;-)</p>
<p>Anyway, the main problem is in the concept it self&#8230; I can accept to pay 30Kb per &#8220;session&#8221; but then I did the following question&#8230;</p>
<p>&#8220;If in 1 minute or less my phone opens 3 connections, I will be charged 30 or 90 Kb ?&#8221;</p>
<p>And winner is 90Kb, so&#8230;</p>
<p>GPRS is a &#8220;data switching&#8221; protocol, there is no &#8220;circuit&#8221; switching&#8221;. So the protocol is designed to work &#8220;connecting&#8221; for every single packet.</p>
<p>Is like to say a packet minimum size is 30Kb&#8230;</p>
<p>So this rule is incorrect, as they are based on &#8220;common usage&#8221;, not in the protocol. They have to invoice packets, or bytes, but not connections as connection is not intrinsic of the protocol and can be done in different ways&#8230;</p>
<p>In my point of view, is an abuse of dominant position or a cartel situation and governments has to act and clearly say this is ILLEGAL! (or IMMORAL ).</p>
<p>As end users we do not have any &#8220;claim&#8221; method&#8230;</p>
<p>I can&#8217;t understand why Europeans or world GSM citizens we do not unite against abusive roaming practices.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
