From 1be510f9529cb082f802408b472a77d074b394c0 Mon Sep 17 00:00:00 2001 From: Nicolas Vigier Date: Sun, 14 Apr 2013 13:46:12 +0000 Subject: Add zarb MLs html archives --- zarb-ml/mageia-dev/2012-March/013674.html | 182 ++++++++++++++++++++++++++++++ 1 file changed, 182 insertions(+) create mode 100644 zarb-ml/mageia-dev/2012-March/013674.html (limited to 'zarb-ml/mageia-dev/2012-March/013674.html') diff --git a/zarb-ml/mageia-dev/2012-March/013674.html b/zarb-ml/mageia-dev/2012-March/013674.html new file mode 100644 index 000000000..13ac1b8e7 --- /dev/null +++ b/zarb-ml/mageia-dev/2012-March/013674.html @@ -0,0 +1,182 @@ + + + + [Mageia-dev] RFT: xen support + + + + + + + + + +

[Mageia-dev] RFT: xen support

+ Maarten Vanraes + alien at rmail.be +
+ Thu Mar 29 22:36:38 CEST 2012 +

+
+ +
Op maandag 11 april 2011 10:33:30 schreef Thomas Backlund:
+> Hi,
+> 
+> xen-4.1.0-1.mga1 is now avaliable in the repos.
+> 
+> And as of kernel-2.6.38.2-4.mga1 there has been some cleanups and addons
+> for xen:
+> 
+> There is now a kernel-xen-pvops that also should work as dom0.
+> 
+> Now this is mostly upstream kenel.org code so its not as fully
+> featured as the old kernel-xen. You can see the features listed here:
+> http://wiki.xensource.com/xenwiki/XenParavirtOps.
+> 
+> There is one exception: our 2.6.38 kernel has xen-netback backend driver
+> added.
+> 
+> So those of you that have been using xen, please test atleast the
+> following:
+> 
+> - check that kernel-server still works as a xen guest
+> - try to use kernel-xen-pvops as dom0 and see how it works
+> (you can also try to use kernel-xen-pvops as a normal xen guest)
+> 
+> Have fun...
+> 
+> --
+> Thomas
+
+i don't know how much people have been testing, but...
+
+dom0 xen didn't work with .xz kernels, i put in a patch for that, and now it 
+works.
+
+in a kvm guest, i've got a working xen dom0 host, but the libvirtd 
+implementation seems unable to find the cpu architecture, and if i force to be 
+the same as the host, i am unable to boot the dom0 anymore. (i think this is 
+likely due to the nature of a KVM guest not really being suitable for a XEN 
+dom0).
+
+i've tried a physical server, but when i finally got to the point of libvirtd, 
+it seemed to not bootup anymore (it was an old disk, which i thought got 
+broken, seeing the IO errors)
+
+I put in a new disk, only to note that the X2 suddenly only had 1 core, which 
+i found very odd, and i noticed the IO errors before  (and also now) were 
+really sata bus errors... (and if i remember correctly this machine had been 
+used with a semi-exploding PowerSupply)... so i think this mobo/cpu is 
+fried... now i need to find new hardware to restart testing again :-( .
+
+note that allthough i can have everything running, i never tested with actual 
+guests, let alone online migrations...
+
+however, i did find out that this xenpvops has 2 different toolchains:
+
+xm and xl. xl seems to do pretty much the same, but apparently doesn't require 
+xend to be running (it reacts directly to the kernel/hypervisor?)
+
+also the xencommons init script seems to start xenconsoled and xenstored on 
+it's own.
+
+i'm thinking of removing both those init scripts and having xend not start as 
+default. (allthough i don't know about xendomains).
+
+i'll keep you guys posted.
+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+

+ +
+More information about the Mageia-dev +mailing list
+ -- cgit v1.2.1