[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-users] Re: Xen 4.1.0 compiled from source on Debian Squeeze 6.0
But the Xen manual very clearly mentions the kernel from Debian, and that it's highly recommended over compiling from source? Is this really the best way? Why doesn't it work with the kernel it says it should work with?
I just want to run Xen with Remus support on either CentOs or Debian. The packages of Debian work perfectly, they just lack Remus support. Can't I add remus separately? (e.g. like before it was integrated into xen)
On Thu, Jun 2, 2011 at 2:16 PM, Ian Tobin <itobin@xxxxxxxxxxxxx> wrote:
I would compile it from source and use the menu options I suggested.
Do you have the xen fs mounted?
From: Wouter van Eekelen [mailto:me@xxxxxxx]
Sent: 02 June 2011 12:29
I've reinstalled Debian Squeeze and followed those exact commands, getting the same error:
root@node1:~# /etc/init.d/xencommons start grep: /proc/xen/capabilities: No such file or directory I don't think the kernel would be the issue in the first place, because I don't believe it's included in the source.
As quoted from the README: "Second, you need to acquire a suitable kernel for use in domain 0. If possible you should use a kernel provided by your OS distributor."
On Thu, Jun 2, 2011 at 12:32 PM, Ian Tobin <itobin@xxxxxxxxxxxxx> wrote:
I think make world also compiles a kernel which you don’t need in this case.
Have you tried cd /usr/src
wget http://bits.xensource.com/oss-xen/release/4.1.0/xen-4.1.0.tar.gz
tar xvf xen-4.1.0.tar.gz
cd xen-4.1.0 make xen
make tools make stubdom
cd dist/install
cp -R * /
cd usr/lib64 (if 64bit)
cp -R * /usr/lib64 (if 64bit)
From: Wouter van Eekelen [mailto:me@xxxxxxx]
Sent: 02 June 2011 11:30 I'm simply following the README, "make world" and then "make install" afterwards.
On Thu, Jun 2, 2011 at 12:28 PM, Ian Tobin <itobin@xxxxxxxxxxxxx> wrote: Sorry I didn’t realise. What steps are you doing to compile XEN?
From: Wouter van Eekelen [mailto:me@xxxxxxx]
Sent: 02 June 2011 11:14
Okay, I'm not compiling my own kernel though - I'm using the one in the Debian packages (2.6.32-5-xen-amd64) as suggested at http://wiki.xensource.com/xenwiki/XenDom0Kernels
Should I compile my own kernel instead? On Thu, Jun 2, 2011 at 12:11 PM, Ian Tobin <itobin@xxxxxxxxxxxxx> wrote:
When your in the make menuconfig screen, under Device Drivers select the same as mine, * is for static compile where M is to be used as a module.
Xm is based on python which is being depreciated.
From: Wouter van Eekelen [mailto:me@xxxxxxx]
Sent: 02 June 2011 11:06
To: Ian Tobin
Cc: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] Re: Xen 4.1.0 compiled from source on Debian Squeeze 6.0
Do you have any tutorial/hint on how to statically compile it? And is there any reason it won't just work out of the box?
I've tried this on CentOs 5 too, but the same errors occur there..
And what exactly is the difference between xm and xl? On Thu, Jun 2, 2011 at 12:02 PM, Ian Tobin <itobin@xxxxxxxxxxxxx> wrote:
Statically compile the XEN stuff rather than using it as modules, I had the same grief as you but went away after doing that.
Also xm commands do not work for some reason but xl works fine.
Ian
From: xen-users-bounces@xxxxxxxxxxxxxxxxxxx [mailto:xen-users-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Wouter van Eekelen
Sent: 02 June 2011 09:42
To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Re: Xen 4.1.0 compiled from source on Debian Squeeze 6.0
Seems like part of my e-mail was lost, here is the rest...
At first, when running "xm list", I received this error: "ImportError: No module named xen.xm" I saw that /usr/lib/python2.6/site-packages/xen wasn't in sys.path of Python, so I symlinked it to /usr/lib/python2.6/xen - problem solved Then I received this error: Error: Unable to connect to xend: No such file or directory. Is xend running? xencommons should be started first. So I try starting xencommons:
grep: /proc/xen/capabilities: No such file or directory A thread on this mailing list suggested I added "none /proc/xen xenfs defaults 0 0" to /etc/fstab
I've done so, rebooted, but the issue persisted. Another thread suggested I ran 'insmod xen-evtchn' and 'modprobe xen-evtchn', which gave me this error:
FATAL: Error inserting xen_evtchn (/lib/modules/2.6.32-5-xen-amd64/kernel/drivers/xen/xen-evtchn.ko): No such device
And now I'm rather stuck. What am I doing wrong? On Thu, Jun 2, 2011 at 10:39 AM, Wouter van Eekelen <me@xxxxxxx> wrote:
I installed Debian Squeeze 6.0 on a server and am now trying to install Xen 4.1.0 on it (compiled from source, since the packages don't include Remus support). Here are the steps I did to install it and get around several errors.
1) Installed Debian Squeeze 6.0 with default options and software RAID 1 3) Rebooted into the kernel and made sure it was running: Linux node1 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC 2011 x86_64 GNU/Linux 4) Installed the requirements for compiling: apt-get install zlib1g-dev python-dev libncurses-dev uuid-dev python-dev libncurses-dev libssl-dev xorg-dev uuid-dev build-essential gcc g++ gettext bcc iasl latex2rtf-doc gcc g++ build-essential
5) Downloaded the 4.1.0 source, untarred it and ran "make world" 6) No errors except a warning about latex for documentation, so I ran "make install"
At first, when running "xm list", I received this error:
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users
|
