Attachment 125875 Details for Bug 105177 – Xen Guide (updated)

Xen Guide (updated)

xen-guide.xml (text/plain), 12.32 KB, created by Sven Vermeulen (RETIRED) on 2007-07-24 13:58:51 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Sven Vermeulen (RETIRED)

Created: 2007-07-24 13:58:51 UTC

Size: 12.32 KB

patch

obsolete

><?xml version='1.0' encoding='UTF-8'?>
>
>
><!DOCTYPE guide SYSTEM "/dtd/guide.dtd">
>
><guide link="/doc/en/draft/xen-guide.xml" disclaimer="draft">
>
><title>Configuring Gentoo with Xen</title>
>
><author title="Author">
>  <mail link="swift@gentoo.org">Sven Vermeulen</mail>
></author>
>
><abstract>
>This guide describes how to start using Xen on your Gentoo system
></abstract>
>
>
>
><license/>
>
><version>0.1</version>
><date>2007-07-24</date>
>
><chapter>
><title>Introduction</title>
><section>
><body>
>
><p>
>The <uri link="http://www.xensource.com/">Xen</uri> technology allows you to run
>multiple operating systems on a single physical system, govern resource
>consumption and even migrate domains (which are the virtual environments in
>which a guest operating system runs) from one Xen-powered system to another. Xen
>requires the host operating system to support Xen (which, in this case, will be
>a Linux kernel) but guest operating systems can run unmodified <e>if</e> your
>hardware supports Intel Virtualization Technology (VT-x) or AMD Virtualization
>Technology (SVM). Otherwise your guest operating systems must also support Xen.
></p>
>
><p>
>This guide will talk you through the configuration steps necessary to get Xen up
>and running on Gentoo Linux. We will not discuss Xen itself (the Xen project has
><uri
>link="http://www.cl.cam.ac.uk/research/srg/netos/xen/readmes/user">decent
>documentation</uri> available) nor will we talk about specialized setups that
>might be very interesting for Xen setups but are not Xen-related (like exporting
>Portage through NFS, booting Linux using PXE, etc.)
></p>
>
></body>
></section>
></chapter>
><chapter>
><title>Preparing Domain0</title>
><section>
><title>Introduction</title>
><body>
>
><p>
><e>Domain0</e> is the primary domain under Xen, hosting the host operating
>system which governs all other domains. In this chapter we will prepare an
>existing Gentoo installation to become the host operating system in this domain
>and build the Xen-powered kernel so that Gentoo is ready to host other Xen
>domains.
></p>
>
></body>
></section>
><section>
><title>Rebuilding the Gentoo Installation ?</title>
><body>
>
><p>
>A drastic change that might be necessary is to rebuild the entire Gentoo
>installation with a different <c>CFLAGS</c> setting. Guest operating systems
>running under Xen might otherwise see major performance degradation. If you,
>however, are planning on checking out Xen rather than installing it for
>production use and are not terribly fond of rebuilding all programs, you can
>skip this step. In this case you will notice performance degradation but you
>will still be able to use Xen.
></p>
>
><impo>
>It is advised that, if you change your <c>CFLAGS</c> and build your system with
>a gcc lower than version 4, you do not have <c>-Os</c> set as it has been
>reported to produce broken code.
></impo>
>
><pre caption="Editing the CFLAGS and rebuild the Gentoo installation">
>~# <i>nano -w /etc/make.conf</i>
><comment>(Add -mno-tls-direct-seg-refs)</comment>
>CFLAGS="-O2 -march=pentium4 -pipe <i>-mno-tls-direct-seg-refs</i>"
>
>~# <i>emerge -e world</i>
></pre>
>
><p>
>If you boot your system using an initial ramdisk (initrd) you need to
>rebuild the initrd as well (which is best done by running all steps you would do
>when you rebuild your kernel).
></p>
>
></body>
></section>
><section>
><title>Installing Xen</title>
><body>
>
><p>
>Xen actually contains many components, so you'll need to install a couple of
>packages. Because it is still <uri
>link="/doc/en/handbook/handbook-x86.xml?part=3&amp;chap=3#doc_chap2">~arch
>masked</uri> you first need to unmask it by adding the necessary lines to
><path>/etc/portage/package.keywords</path> and then install them.
></p>
>
><pre caption="Unmasking and Installing Xen">
>~# <i>nano -w /etc/portage/package.keywords</i>
>app-emulation/xen
>app-emulation/xen-tools
>sys-kernel/xen-sources
>
>~# <i>emerge xen xen-tools xen-sources</i>
></pre>
>
>
>
></body>
></section>
><section>
><title>Building the Kernel</title>
><body>
>
><p>
>Next we'll build the Linux kernel with Xen support. This kernel, whose sources
>are available at <path>/usr/src/linux-2.6.x.z-xen</path>, will be our main
>running kernel (i.e. the one running domain 0). In the <c>XEN</c> section you'll
>find drivers for all kinds of input/output, each driver having a <e>backend</e>
>and <e>frontend</e> implementation available. For the domain 0 kernel you need
>to select the <e>backend</e> implementation: these are used by the other
>domains (who use the <e>frontend</e> drivers) to communicate directly with
>the hardware.
></p>
>
><p>
>Of course, don't forget to select <c>Xen-compatible</c> at <c>Processor type and
>features</c>. If you're wondering about networking: each interface in a domain
>has a point-to-point link to an interface on domain 0 (called
><path>vifX.Y</path> where X is the domain number and Y the Yth interface of that
>domain), so you can configure your network the way you want (bridging, NAT,
>etc.)
></p>
>
><p>
>A nice hint is to have the kernel make process store its intermediate object files
>elsewhere so that you can reuse the same kernel tree to build different 
>configurations:
></p>
>
><pre caption="Building the Kernel">
>~# <i>mkdir -p ~/build/dom0 ~/build/domU</i>
>~# <i>make O=~/build/dom0 menuconfig</i>
><comment>(Configure the kernel)</comment>
>~# <i>make O=~/build/dom0 &amp;&amp; make O=~/build/dom0 modules_install</i>
></pre> 
>
><p>
>Once the kernel is built you'll find the kernel image immediately in the 
>build directory (not inside <path>arch/</path> or any other directory) called
><path>vmlinuz</path>. Copy it to <path>/boot</path> and then configure your
>bootloader to use the Xen hypervisor (one of the components installed
>previously) which is stored as <path>/boot/xen.gz</path>. In the bootloader
>configuration, add your newly built kernel as the kernel that Xen should
>boot. For instance, for GRUB:
></p>
>
><pre caption="GRUB Configuration for Xen">
>title Xen 3.0 / Gentoo Linux 2.6.x.y
>root (hd0,0)
>kernel /boot/xen.gz
>module /boot/kernel-2.6.x.y-xen0 root=/dev/hda3
></pre>
>
><p>
>Now reboot your system into Xen. Once you are booted, you need to load the Xen
>daemon:
></p>
>
><pre caption="Loading the Xen daemon">
>~# <i>/etc/init.d/xend start</i>
></pre>
>
><p>
>Now check if you can do whatever you normally do on your system. If this is the
>case, you can edit your bootloader configuration to always boot into Xen and add
>the Xen deamon to the default runlevel so that it is started automatically 
>next time you boot.
></p>
>
></body>
></section>
></chapter>
><chapter>
><title>Creating an Unpriviledged Domain</title>
><section>
><title>Building the Kernel</title>
><body>
>
><p>
>Go to the Xen-powered Linux kernel source and update the configuration. It is
>wise to keep as many topics as possible similar to the main kernel except the
><c>XEN</c> settings where drivers should now have their <e>frontend</e>
>implementation selected instead of the <e>backend</e>. Then build the kernel
>and place the resulting <path>vmlinuz</path> file where you want (we assume this
>is <path>/mnt/data/xen/kernel</path>):
></p>
>
><pre caption="Building the guest kernel">
>~# <i>make O=~/build/domU</i>
>~# <i>cp ~/build/vmlinuz /mnt/data/xen/kernel/kernel-2.6.x.y-xen</i>
></pre>
>
><p>
>It is also possible to create a single kernel image for both the administrative
>domain and the unpriviledged domain. More information about this can be found
>in the Xen user manual.
></p>
>
></body>
></section>
><section>
><title>Creating the Domain Disks</title>
><body>
>
><p>
>For best performance, it is best to dedicate a partition (or logical volume) to
>a domain rather than a file based filesystem. However, if you are going to use
>Xen primarily for tests using a file based filesystem does have its advantages
>(especially regarding maintenance).
></p>
>
><p>
>You can create a file based filesystem using <c>dd</c> and <c>mke2fs</c> (or
>any other file system creation tool). For instance, to create a 2Gbyte ext3
>filesystem:
></p>
>
><pre caption="Creating a file based filesystem">
>~# <i>dd if=/dev/zero of=/mnt/data/xen/disks/ext3root.img bs=1M count=2048</i>
>~# <i>mke2fs -j /mnt/data/xen/disks/ext3root.img</i>
></pre>
>
></body>
></section>
><section>
><title>Configuring a Domain</title>
><body>
>
><p>
>Next we create a Xen configuration file for a domain. You can store these
>configuration files where you want, for instance at
><path>/mnt/data/xen/configs</path>. As an example, we create a configuration
>file for a small Gentoo environment which uses the disk image we created
>previously:
></p>
>
><pre caption="Creating a domain configuration file">
>~# <i>nano -w /mnt/data/xen/configs/gentoo</i>
>
>kernel = "/mnt/data/xen/kernel/kernel-2.6.x.y-xen"
>memory = 512
>name   = "gentoo"
><comment>(Map the disk image to the virtual /dev/sda1)</comment>
>disk   = ['file:/mnt/data/xen/disks/ext3root.img,sda1,w']
>root   = "/dev/sda1 ro"
></pre>
>
><p>
>You can find example configuration files in <path>/etc/xen</path>.
></p>
>
></body>
></section>
><section>
><title>Launching the New Domain</title>
><body>
>
><p>
>Now we're all set and we can launch the new domain. If the disk image contained
>an operating system, we could just create and attach the domain using the
><c>xm</c> command (Xen manager):
></p>
>
><pre caption="Creating and starting a new domain">
>~# <i>xm create /mnt/data/xen/configs/gentoo -c</i>
></pre>
>
><p>
>The domain would be booted inside the terminal in which you executed the
>command. However, in our case, the disk image is empty so the domain wouldn't be
>able to do much. To fix this, you can loop-mount the image and install Gentoo as
>you're used to.
></p>
>
><p>
>If you want to disconnect from the domain, press <path>Ctrl+]</path>. You can
>always reconnect to the domains' console using <c>xm console gentoo</c>.
>However, there is only one console per domain, so only use it when you can't
>access the domain otherwise (for instance, through SSH).
></p>
>
></body>
></section>
></chapter>
><chapter>
><title>Networking on Unpriviledged Domains</title>
><section>
><title>Introduction</title>
><body>
>
><p>
>Xen supports at least two ways of configuring your (virtual) network:
><e>routed</e> and <e>bridged</e>.
></p>
>
><p>
>When selecting the <e>routed</e> approach, the interface inside your
>unpriviledged domain is connected to the virtual interface on your
>administrative domain. On your administrative domain (domain 0), the virtual
>interface is linked together with <path>eth0</path>. The
>interface inside your unpriviledged domain should have an IP address on the same
>network as the interface on the administrative domain. Any communication to 
>that IP address can only occur from the administrative domain, unless you set 
>up specific routing rules.
></p>
>
><p>
>When selecting the <e>bridged</e> approach, <brite>TODO</brite>.
></p>
>
></body>
></section>
><section>
><title>Regular Routed Interfaces</title>
><body>
>
><p>
>Before you set up the interface on your unpriviledged domain, make sure that
>Xen's <path>netloop</path> and <path>netbk</path> drivers are loaded. A quick
>hint: if you have <path>netloop</path> as a module, load it with
><c>nloopbacks=0</c> so that it doesn't create pointless interfaces to the
>loopback device. Then, edit your domain configuration file and add a <c>vif</c>
>instruction to it.
></p>
>
><pre caption="Configuring a virtual interface">
>~# <i>nano -w /mnt/data/xen/configs/gentoo</i>
>
><comment>(Add the vif instruction)</comment>
>vif    = [ 'ip=192.168.1.101, vifname=veth1' ]
></pre>
>
><p>
>In the above example, the interface will be created for the unpriviledged domain
>(in which it will be called <path>eth0</path>) and Xen will ensure that address
>192.168.1.101 will be reachable from the administrative domain through interface
><path>veth1</path>. 
></p>
>
><p>
>This doesn't mean that the virtual <path>eth0</path> interface will automatically 
>have IP 192.168.1.101 assigned to it, but rather that, if you don't give it that 
>IP, it will not be connected with the administrative domain and thus cannot be 
>reached.
></p>
>
><p>
>Now edit <path>/etc/xen/xend-config.xsp</path> as follows to select routed
>network configuration:
></p>
>
><pre caption="Editing xend-config.xsp">
>~# <i>nano -w /etc/xen/xend-config.xsp</i>
>
><comment>(Comment out the following lines)</comment>
><i>#</i>(network-script network-bridge)
><i>#</i>(vif-script vif-bridge)
>
><comment>(Enable the following lines)</comment>
>(network-script network-route)
>(vif-script vif-route)
></pre>
>
></body>
></section>
></chapter>
></guide>

Actions: View

Attachments on bug 105177: 101602 | 102579 | 125262 | 125875 | 125954 | 129096