Everything is a Freaking DNS problem

I've ranted before about the difficulties of getting a 2.6 kernel to work with a Silicon Image ATA Controller , Afterall it seemed that it was just a case of bad habits

The problems I was running into was that I could partition a disk, say /dev/hdd1 make a filesystem on it but not mount it, or when I got so lucky to mount it and write on it, at the next reboot the data wasn't there anymore.
I repeatedly spent a couple of minutes looking after a replacement for the MegaIDE that existed in 2.4 for a 2.6 kernel, never found one.
Now it seems that support has been there al the time and that dmraid seemed to be my friend all along,

As soon as you realise that you know you have to stop using /dev/hdwhatever and use /dev/mapper/$devicename , and that seems to work like a charm :)

root@cnode1 ~# dmraid -s
*** Active Set
name   : lsi_cdhajaiciia
size   : 78165359
stride : 128
type   : stripe
status : ok
subsets: 0
devs   : 1
spares : 0
*** Active Set
name   : lsi_cdhajaiciib
size   : 78165359
stride : 128
type   : stripe
status : ok
subsets: 0
devs   : 1
spares : 0
*** Active Set
name   : lsi_cdhajaiciic
size   : 78165359
stride : 128
type   : stripe
status : ok
subsets: 0
devs   : 1
spares : 0
*** Active Set
name   : lsi_cdhajaiciid
size   : 78165359
stride : 128
type   : stripe
status : ok
subsets: 0
devs   : 1
spares : 0

I only needed access to 4 separate disks, but off course the moment you really want to put these disk in raid you'll have to get this into your initrd.

Now it's for sure, today I gave my last talks for this year :) pfiew , December seemed to be the never ending training and workshop month, starting with an Introduction to ITIL course , then travelling to London for a Jboss
Administration course in order to finish the year with the 2nd edition of our own X-Tend-Days , 3 days with 10-15 geeks in a room explaining eachother stuff , topics ranging from docbook over pam , lvm and selinux to linux-ha2, gfs, ocfs2 and Xen. Exhausting but worth the efforts.

Remember my previous rant on Fon not showing the code ? Actually they never replied to my mails, it's not in their faq, and as their webpages are not dated I have no clue since when their source is available so I might have overlooked it before.

But I just found it ,It's very well hidden , but I found it ... I`ll be having a closer look at it and I`ll see if I can document stuff about it.
Don't expect anything to soon however :)

Steven J. Vaughan-Nichols just missed the ball with his
Xen's no killer app article ;(

First of all he is wrong about the lack of platform support. Xen
supports a lot more platforms than Linux 2.4 en 2.6 , namely lots of BSD
Variants , Plan 9 etc. Sun also seems to have Solaris working on Xen and there exists
a Windows version for Xen which is just not available for the general
public. However with the introduction of the VT Chipsets that issue won't
exist anymore.

As for the lack of management part he is completely missing out on how
you manage an infrastructure, the whole point being that there is no
difference between deploying physical or virtual machines and having
virtual machines only makes it easier for you.

Apart from that there are plenty of tools to manage the actual virtual machine stuff,
ranging from the command line over web interfaces to other stuff that
people have been contributing.

Actually the original python scripts that Xen delivered as management
tools have already been ported to C ,rewritten and modified by other
people , hopefully the best of breed of these tools wil eventually become the Xen
default tools

I do however agree on his claims that in order to really beat the * out
of the competition you have to become better than them and tackle new
issues, not "just improve' their stuff.

On top of that he actually forgets about the point that Xen is Free Software

From : Schneier on Security: Microsoft Windows Receives EAL 4 Certification:

"Or are the certifications proof that EAL 4+ isn't worth much?"

Needless to comment..

It's all true what they say .. London is probably one of the hottest spots in the world, when my 770 starts scanning it finds at least 3-4 unencrypted access points.
I`m blogging this from a hotspot I found from my hotel room, no clue who it belongs to, It's called WLAN and it gives you an ip via dhcp from there you're online.

I`m heading back to Antwerp tomorrow evening, flying VLM which takes me from Downtown London back home in less than 2h30 including tubes and the drive back home by car, I still wonder why on earth I ever flew Brussels-Heathrow when Antwerp-London City takes you much faster there.

So it seems I`m also on the spamlist of Prof. Nagib Callaos
whose WMSCI conferences don't really have a good reputation regarding accepting (non)quality papers to their agenda, as documented in SCIgen - An Automatic CS Paper Generator, they often don't even read the papers that are being submitted. Anthony Liekens goes a step further by identifying more conference as Fake :)
http://anthony.liekens.net/index.php/Misc/FakeConferences I`m lucky I googled to see what the conference looked like the past couple of years and didn't find anything .

The System Imager Suite presentation I gave yesterday evening is online , the patches provided by Jo and Bob have already been applied :)

I succeeded in eliminating the 2 phase bootstrap today , more news on that follows in the next couple of days. (On Xen 3.0 , Centos 4.2 and SystemImager)

Today I also received the invitation to speak at LinuxDays.lu

In case you're interrested in Large Scale and/or Automated Linux Deployments I`ll be doing a talk on SystemImager Suite tomorrow at L2U

My good friend Matt Rechenburg of openMosixview fame , is working on a new project kiscsiadm, kiscsiadm is the system-administrator's "swiss knife" when it comes to ISCSI. It is a GUI for common ISCSI-actions which are executed by just a few mouse-clicks.