xen October 2007

xen@lists.fedoraproject.org

32 participants
33 discussions

VFS: Cannot open root device "sda1" or unknown-block(0, 0)
by Haifeng He 15 Oct '07

15 Oct '07

Hi, All I followed the steps in Fedora7VirtQuickStart to install Xen on Fedora 7. I downloaded couples of file system image from http://jailtime.org/. However, every time I tried to create guest OS using those file system image, I got error messages: VFS: Cannot open root device "sda1" or unknown-block(0,0) Please append a correct "root=" boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) An example of Xen configure is: kernel = "/boot/vmlinuz-2.6.20-2936.fc7xen" memory = 256 name = "fedora.fc6" vif = [ '' ] dhcp = "dhcp" disk = ['file:/xen/fedora/fedora.fc6.img,sda1,w', 'file:/xen/fedora/fedora.swap, sda2,w'] root = "/dev/sda1 ro" Having been googling on this topic for a while, but still not getting the right answer for it. Any help will be appreciated. Thanks Haifeng

2 3

Can't ping windows HVM machine
by Nico Sabbi 15 Oct '07

15 Oct '07

Hi, using fc8-test2 updated with all packages available I could successfully install a windows xp machine, but I can't even ping it (although windows can ping my fedora and internet overall). The ip assigned to windows is in the range 192.168.122.x (assigned to a virtual nic) while the ip assigned to fedora is 192.168.2.102. route -n shows everything correctly set up (192.168.122.0/24 going through the virtual interface (virbr0 IIRC)), but still I can't ping the virtual machine, even if I run $ /etc/init.d/iptables stop What did I do wrong? BTW, since it's well known that in the virtual machine the usb mouse must be replace with a usb tablet to make the pointer work correctly, I guess that this option should be set by default rather than making users swear for hours :) Thanks, Nico

1 0

are dropped packets a sign of network problem?
by progressdll 12 Oct '07

12 Oct '07

in a xen environment , i see a lot op dropped packets via netstat -i Is this a sign of network problems, or is it normal to see this kind of numbers? i'm not sure how to interprete the data. is this normal, bad, critical. What are your stats on this? I guess i have a xen issue of some sort, but not sure what yet # netstat -i Kernel Interface table Iface MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR Flg eth0 1500 0 53999491 0 0 0 6234874 0 0 0 BMRU lo 16436 0 9247734 0 0 0 9247734 0 0 0 LRU peth0 1500 0 135679270 0 0 0 95455766 0 0 0 BORU tap0 1500 0 5411876 0 0 0 35071803 0 0 0 BMRU vif-vmdeb 1500 0 0 0 0 0 35 0 333350 0 BORU vif0.0 1500 0 6234874 0 0 0 53999492 0 0 0 BORU vif6.0 1500 0 55813641 0 0 0 98482968 0 482338 0 BORU vif6.1 1500 0 0 0 0 0 39 0 46875771 0 BORU vif68.0 1500 0 0 0 0 0 0 0 0 0 BOU vif69.0 1500 0 9349348 0 0 0 30571290 0 3282 0 BORU vif69.1 1500 0 0 0 0 0 44 0 18947590 0 BORU vif70.0 1500 0 129605 0 0 0 460256 0 702 0 BORU vif70.1 1500 0 0 0 0 0 35 0 333350 0 BORU xenbr0 1500 0 46298695 0 0 0 0 0 0 0 BORU -- View this message in context: http://www.nabble.com/are-dropped-packets-a-sign-of-network-problem--tf4598… Sent from the Fedora Xen mailing list archive at Nabble.com.

3 2

Install, backup and recover
by Augusto Castelan Carlson 11 Oct '07

11 Oct '07

Hi! I will have to install almost 30 servers with xen. Each server will host two guests. Both host and guests OS will be fedora 7. One hard disk partition will be allocated to each guest, for example: host -> /dev/sda1 guest1 -> /dev/sda2 guest2 -> /dev/sda3 I did some search in the list archives and google, but found what I was looking for. I would like to have some suggestion about the following: 1) The best approach to install all these hosts and guest? Cobbler? I'm not sure if the server configuration will be all the same. 2) If the hard disk fail, which is the best approach to recover the host and guests? Don't care about users data and server config files, as there will be user data and config backup. 3) If only one of the guests fail, which is the best approach to recover this guest? 4) If only the host fail, which is the best approach to recover this host and do not lost access to guest using to such as virt-manager/virsh/xm? I can provide more info if needed. Thanks in advance. Regards, -- Augusto Castelan Carlson

1 0

Re: xen, revisor
by Arno Karner 09 Oct '07

09 Oct '07

Arno, May we know what version of Fedora you are using, and what version/distribution it is you're trying to create? * fc7 with test repos + my repos Arno Karner wrote: > ok I make a new dvd iso image with revisor, modify the isolinux.cfg so > it boots, and installs with my kickstart file. Revisor does that already, so what is it you edit? Further on in the message I see you add another option but also you want to troubleshoot really fast. Manually adding in an option doesn't seem like the most efficient thing to do. * revisor doesn't do the default kickstart file right, corectectly at the moment, and * certainly doesn't do multiple kickstart files boot menu options yet * so i deconstuct the image, copy my kick start files to the iso image * copy my isolinux.cfg menu, reconstruct the image > I look at the anaconda.ks.cfg and none of my rpm packages, which were > installed show up. > Depending on what is in your original kickstart I'm sure the sample kickstart anaconda writes out after install is much different. In particular when any of the packages fail to install. * my kickstart ran too comletion, but none of my packages which were * installed showed up in the anaconda.ks.cfg, I'd really like to * get some kind of feed back loop working so dep packages show up > Now I'm trying to trouble shoot the post install section of a few of my > packages, which didn't seem to be executed, > so is there some where to look. install log didn't show anything useful. > I think the %post scripts of your packages is an OK point to start troubleshooting. * well the first package I'm trouble shooting is not a simple one * its a multiple package spec file unfortanlly Requires: is not a * %command <sub package> format command > How do I get my rpm-keys installed, doing it from an rpm is not possible > because rpm does not support nested transactions, > any Ideas here welcome also. the key files are in the root directory od > the install image, but that is not good enough. > > I looked at the spec file of the package in question and it had a > Requires: <package> <package> but the packages were not installed, > or complained about. Hence other packages that were installed gave > warning messages about user xxx, group yyyy not defined using root:root > not what I wanted........ > Are you saying yum and anaconda are having difficulties eating your packages, or any package? I remember something called PreRequires to indicate the package being required should be actually installed before the package requiring is installed. Did you check any Provides: user(foo), group(bar), against any Requires: user(foo), group(bar)? Are these your very own packages? Have you tested excluding your own packages see how that goes? * PreRequires theres something worth looking into, because * yes I want certain of my packages to be installed before others > I'm reducing the install ks.cfg file, to make trouble shooting this in > xen faster, and including a second menu option for a full install of > what I want. > > Sorry for the cross post but this issue deals with revisor, xen, > anaconda. I'll be looking into the anaconda emailing list next. > I really don't think this has anything to do with xen. * your right I apologize, I left out the parts about how do I get * my virtual machine to reboot off the cd the second, third time I restart it * to test the different install menu items without deleting the virt machine and recreating * how do I change the cd/dvd isoimage to handle the second piece of media > I am using bleeding edge rpms, git fc7.testing revisor:git pull, and I > should expect this. So I thought I'd ask while I try and reduce my loop > time from 4+ hours > build iso image, xen.new virt machine + play free civ, update scripts > and ks.list(s)....repeat > > Ideas, flames, constructive criticizing welcome. Here's an idea. The average desktop floating around any company will beat you to the 4+ hours of composing time (~30 minutes) and another will brighten your day doing installs in 15 minutes or less. * soon that might be an option, but using xen, is where some of this is going * kickstart files for different virt machine configurations, kind of hard to test those on * a desktop machine, without running xen which brings me to where I am. Kind regards, Jeroen van Meeuwen - -kanarip -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFHCe39KN6f2pNCvwgRAkkDAJ42TGT1FVUZP4LNkmXFMv2Gpbap5wCgkCj9 g3SfZdWP+TD5EOnvu2T+DgU= =j7R8 -----END PGP SIGNATURE----- sorry for the poor quoting, but yahoo does lack common features, like quoting replies Arno ____________________________________________________________________________________ Take the Internet to Go: Yahoo!Go puts the Internet in your pocket: mail, news, photos & more. http://mobile.yahoo.com/go?refer=1GNXIC

1 0

xen, revisor
by Arno Karner 08 Oct '07

08 Oct '07

ok I make a new dvd iso image with revisor, modify the isolinux.cfg so it boots, and installs with my kickstart file. I look at the anaconda.ks.cfg and none of my rpm packaes, which were installed show up. Now I'm trying to trouble shoot the post install section of a few of my packages, which didn't seem to be executed, so is there some where to look. install log didn't show anything useful. How do I get my rpm-keys installed, doing it from an rpm is not possible because rpm does not support nested transactions, any Ideas here welcome also. the key files are in the root directory od the install image, but that is not good enough. I looked at the spec file of the package in question and it had a Requires: <package> <package> but the packages were not installed, or complained about. Hence other packages that were installed gave warning messages about user xxx, group yyyy not defined using root:root not what I wanted........ I'm reducing the install ks.cfg file, to make trouble shooting this in xen faster, and including a second menu option for a full install of what I want. Sorry for the cross post but this issue deals with revisor, xen, anaconda. I'll be looking into the anaconda emailing list next. I am using bleeding edge rpms, git fc7.testing revisor:git pull, and I should expect this. So I thought I'd ask while I try and reduce my loop time from 4+ hours build iso image, xen.new virt machine + play free civ, update scripts and ks.list(s)....repeat Ideas, flames, constructive criticizing welcome. Arno ____________________________________________________________________________________ Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games. http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow

2 1

virt-manager "Cannot allocate memory" creating Win2K3 HVM
by Andy Burns 05 Oct '07

05 Oct '07

Configuration of dom0 is as follows 8GB memory 2 x dual core Xeon kernel-xen.x86_64 2.6.20-2931.fc7 xen.x86_64 3.1.0-2.fc7 xen-libs.x86_64 3.1.0-2.fc7 libvirt.x86_64 0.3.2-1.fc7 virt-manager-0.4.0-2.fc7 I'm trying to install a win2K3 32bit HVM domU using /cd.iso file local to dom0 LVM partition /dev/vgmirror/win2k3 as domU disk bridged eth0 network 1024MB memory for domU (have tried with lower memory) Fails with POST operation failed: (xend.err "Error creating domain: (12, 'Cannot allocate memory')") Rebooting the dom0 is to be avoided if possible due to other paravirt domU guests running, any suggestions or extra info that might help? I have managed to successfully create a win2K3 64bit HVM domU on an identical dom0 server.

1 1

Xen Virt-Manager odd reaction
by slbubba＠charter.net 05 Oct '07

05 Oct '07

Folks; I have a small problem here that perhaps someone knows the answer to. I'm running the following: kernel xen 2.6.18-8.1.14.el5 xen 3.0.3-25.0.4.el5 virt-manager-0.2.6-7.0.2.el5 When using virt-manager to add a new VM (in this case Fedora core 6) the installation moves along smoothly and the VM is displayed in the main virt-manager window under dom-0. At the end of the VM installation the screen to remove all disks and click the reboot button appears. Clicking on "reboot" in the VM window starts the reboot process. However, the VM entry in the virt-manager window disappears, the VM console disappears, and the VM never comes back. The VM flat file is there (in my case it's /opt/vm01a) and the correct size but virt-manager does not see the VM, only dom-0. I've gone through this multiple times and now have 3 VM's but no way to access them. Where is the data for the VM kept that virt-manager uses? Why is virt-manager acting this way? Any patches, fixes, different revs? Thanks Warren

2 2

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

xen October 2007