xen October 2006

xen@lists.fedoraproject.org

59 participants
64 discussions

how to use "xm save" and "xm restore" to avoid re-install of guest OSes
by Ranganathan, Shobha 23 Oct '06

23 Oct '06

Hi, Can I use "xm save" and "xm restore" to be able to use guest OSes image between xen builds and avoid re-installing guest OSes ? I am changing Xen 3.0.2 sources. I now keep installing the guest OSes after every rebuild of Xen kernel.. How can I avoid it ? Regards Shobha Ranganathan "The significant challenges we face cannot be resolved at the same level of thinking we were at when we created them." Albert Einstein

1 0

FC6 File-backed domain kernel panic
by Bill Jenkins 23 Oct '06

23 Oct '06

I am having a few problems with Xen on Fedora 6 Test 3. I am attempting to use a file-backed container for the individual operating systems. Everything installs fine, and runs fine...but when I attempt to either xm shutdown or xm destroy a domain, and then reboot it with xm create, it gives a long error, containing, among other things, the following: WARNING: leaking g.e. and page still in use! end_request: I/O error, dev xvda, sector 0 Buffer I/O error on device xvda, logical block 0 WARNING: g.e. still in use! WARNING: leaking g.e. and page still in use! end_request: I/O error, dev xvda, sector 0 Buffer I/O error on device xvda, logical block 0 WARNING: g.e. still in use! WARNING: leaking g.e. and page still in use! end_request: I/O error, dev xvda, sector 0 Buffer I/O error on device xvda, logical block 0 unable to read partition table If I save the domain into a CHK file and restore it, everything is fine. But is this a side-effect of running a file-backed OS? Or could there be something else I am doing wrong? Bill

1 0

Testing report for xen against RHEL5 Beta2 M5
by Zhao, Yunfeng 23 Oct '06

23 Oct '06

Hi. We have finished a full test for xen IA32p/IA32e/IA64 against RHEL5 Beta2 M5. The full test includes guest installation test,smp Linux/Windows guest test,device model test,control panel test, hvm guest stress test and manual testing for Virtual Machine Manager. In the testing we found 20 issues totally. (10 are RHEL5 specific issues,and 10 are common issues.) Following are the issues: RHEL5 specific issues: 1. 'Could not initialize SDL - exiting' error The error happens when I try to create vmx guest on RHEL5. A workround for this issue is: execute "xend stop" and "xend start" before creating vmx guests. https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104369 2. 'Failed vm entry' error when install RHEL5 IA32e into VMX The guest crashed for vm entry failure. (XEN) Failed vm entry (exit reason 0x80000021) caused by invalid guest state (0). https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104283 3. Cannot boot RHEL5 Beta2 M5 SMP VMX guest The guest always stops booting after printing "input: ImExPS/2 Logitech Explorer Mouse as /class/input/input1". https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104292 4. Virtual Machine Manager cannot detect correct total memory size on host machine On my test box the total memory is 1GB, but the tool shows it is 2GB. https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104367 5. Fails to install fully virtualized guests from ISO with using Virutal Machine Manager Virutal Machine Manager is slow to bring up the qemu window, that cause user hasn't chance to choose boot from CDROM. https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104368 6. The network speed between xen0 and hvm guest is slow. https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104855 7. Xen crashed after xen0 rebooted https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&t id=104856 8. On IA64 xen, xenu would crash Xen0 if insmod xennet.ko https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=208062 9. On IA64 xen, xen0 could be rebooted. (This issue has been fixed in latest Rawhide.) 10. On IA64 xen, xenU can not boot up with 256M https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=207241 Common issues that also exist on xen unstable: 11. Failed to install Vista Beta2 IA32 12. Installing win2k with acpi=1 failed with blue screen error. 13. Can not input number from speail number keys (when Open Num Lock) 14. SMP Windows 2003 HVM guest is not stable. After running for a while the guest hanged. 15. Full virtualization and Para virtualization networks cannot coexist on HVM guest. PV network module cannot be loaded when FV network driver running on HVM guest. 16. Xen0 ia32-pae may hang after running some heavy workload on hvm guest for several hours. We met xen0 ia32-pae hang several times in the full functional testing and stress testing. In the full functional testing, the test system was continuing creating and destroying lots of hvm guests. In the stress testing, two hvm guests were running some heavy stress test suites synchronously. 17. On IA64 xen, LTP test causes SMP VTI to hang. 18. On IA64 xen, VTI guest can not boot 3 NICs at the same time. 19. On IA64 xen, "Shut down" in VTI win2k3 will cause VTI reboot 20. On IA64 xen, SMP (8 vcpus) VTI failed to pass 5 hours helltest. (Helltest is an Intel internal stress test for linux ) When RHEL5 beta2 M6 server version is available, we will do a regression test for xen. Thanks Yunfeng

1 0

A different problem with the new xen 3.0.2 package
by Ben 22 Oct '06

22 Oct '06

After upgrading to the new kernel and xen userspace packages, I'm seeing this in my xend.log when I try to start it: [2006-10-21 09:37:24 xend 2674] INFO (SrvDaemon:283) Xend Daemon started [2006-10-21 09:37:24 xend 2674] INFO (SrvDaemon:287) Xend changeset: unavailable . [2006-10-21 09:37:24 xend 2674] ERROR (SrvDaemon:297) Exception starting xend ((111, 'Connection refused')) Traceback (most recent call last): File "/usr/lib64/python2.4/site-packages/xen/xend/server/ SrvDaemon.py", line 291, in run servers = SrvServer.create() File "/usr/lib64/python2.4/site-packages/xen/xend/server/ SrvServer.py", line 108, in create root.putChild('xend', SrvRoot()) File "/usr/lib64/python2.4/site-packages/xen/xend/server/ SrvRoot.py", line 40, in __init__ self.get(name) File "/usr/lib64/python2.4/site-packages/xen/web/SrvDir.py", line 82, in get val = val.getobj() File "/usr/lib64/python2.4/site-packages/xen/web/SrvDir.py", line 52, in getobj self.obj = klassobj() File "/usr/lib64/python2.4/site-packages/xen/xend/server/ SrvDomainDir.py", line 39, in __init__ self.xd = XendDomain.instance() File "/usr/lib64/python2.4/site-packages/xen/xend/XendDomain.py", line 636, in instance inst.init() File "/usr/lib64/python2.4/site-packages/xen/xend/XendDomain.py", line 71, in init xstransact.Mkdir(VMROOT) File "/usr/lib64/python2.4/site-packages/xen/xend/xenstore/ xstransact.py", line 345, in Mkdir complete(path, lambda t: t.mkdir(*args)) File "/usr/lib64/python2.4/site-packages/xen/xend/xenstore/ xstransact.py", line 351, in complete t = xstransact(path) File "/usr/lib64/python2.4/site-packages/xen/xend/xenstore/ xstransact.py", line 20, in __init__ self.transaction = xshandle().transaction_start() File "/usr/lib64/python2.4/site-packages/xen/xend/xenstore/ xsutil.py", line 18, in xshandle xs_handle = xen.lowlevel.xs.xs() Error: (111, 'Connection refused') Interestingly (?) my normal xend processes remain running, even if xend doesn't seem to be working: root 2761 0.0 0.0 105000 5908 ? S 09:45 0:00 python /usr/sbin/xend start root 2762 0.5 0.0 115240 6748 ? Sl 09:45 0:00 python /usr/sbin/xend start [root@johnny xen]# uname -a Linux johnny 2.6.18-1.2200.fc5xen0 #1 SMP Sat Oct 14 17:36:54 EDT 2006 x86_64 x86_64 x86_64 GNU/Linux [root@johnny xen]# rpm -q libvirt libvirt-devel xen libvirt-0.1.7-2.FC5 libvirt-devel-0.1.7-2.FC5 xen-3.0.2-5.fc5

2 1

xenconsole: Could not open tty `/dev/pts/2'
by Paul Clark 21 Oct '06

21 Oct '06

I just got the 2.6.18-1.2200.fc5xen0 running. I'm trying to create my first VM and I get the following error. ['/usr/sbin/xm', 'create', '-c', '/etc/xen/XP64-Dev'] Using config file "/etc/xen/XP64-Dev". Started domain XP64-Dev xenconsole: Could not open tty `/dev/pts/2': No such file or directory XM command exited. Your guest can be restarted by running 'xm create -c XP64-Dev'. Otherwise, you can reconnect to the console with vncviewer or 'xm console' Anyone have an idea about why the pseudo-terminal device is not there.

2 1

RE: [Fedora-xen] Fwd: Error: Error connecting to xend: No such fi le or directory. Is xend running?
by Alex Tomic 20 Oct '06

20 Oct '06

I was encountering the same issue after doing a full yum update earlier today, running Xen 3.0.2-5 and kernel 2.6.18-1.2200. Unfortunately, just booting the previous kernel (2.6.17-1.2187) didn't fix the problem so I had to revert back to xen 3.0.2-3 and libvirt 0.1.1-1 packages that I found on rpmfind.net to get back up and running. -Alex -----Original Message----- From: fedora-xen-bounces(a)redhat.com [mailto:fedora-xen-bounces@redhat.com] On Behalf Of Adrian Chadd Sent: Thursday, October 19, 2006 6:58 PM To: Daniel P. Berrange Cc: fedora-xen(a)redhat.com Subject: Re: [Fedora-xen] Fwd: Error: Error connecting to xend: No such file or directory. Is xend running? On Thu, Oct 19, 2006, Daniel P. Berrange wrote: > On Thu, Oct 19, 2006 at 04:52:25AM -0400, Daniel Veillard wrote: > > On Thu, Oct 19, 2006 at 03:31:01PM +0800, Adrian Chadd wrote: > > > It seems due to a mismatch between the xen kernel version they've rolled in the > > > latest FC5 and the Xen user utilities. > > > > And the solution is to update from xen and libvirt-devel from fedora-testing > > The neccessary packages have now been moved from updates-testing to the > regular 'updates' channel. So a simple 'yum update' should pull in all > the bits needed to get a working Xen install with new kernels. I'm now seeing this, from xend.log: [2006-10-20 09:46:41 xend] ERROR (xmlrpclib2:124) method "xend.domain.getRestartCount" is not supported Traceback (most recent call last): File "/usr/lib/python2.4/site-packages/xen/util/xmlrpclib2.py", line 103, in _marshaled_dispatch class ServerProxy(xmlrpclib.ServerProxy): File "/usr/lib/python2.4/SimpleXMLRPCServer.py", line 408, in _dispatch raise Exception('method "%s" is not supported' % method) Exception: method "xend.domain.getRestartCount" is not supported [root@hosting-2 ~]# rpm -q -a | grep xen kernel-xen0-2.6.18-1.2200.fc5 xen-3.0.2-5.fc5 kernel-xenU-2.6.18-1.2200.fc5 [root@hosting-2 ~]# rpm -q -i xen Name : xen Relocations: (not relocatable) Version : 3.0.2 Vendor: Red Hat, Inc. Release : 5.fc5 Build Date: Tue 03 Oct 2006 09:11:07 AM WST Install Date: Fri 20 Oct 2006 09:44:51 AM WST Build Host: hs20-bc2-2.build.redhat.com Group : Development/Libraries Source RPM: xen-3.0.2-5.fc5.src.rpm Size : 4905848 License: GPL Signature : DSA/SHA1, Thu 19 Oct 2006 10:23:45 PM WST, Key ID b44269d04f2a6fd2 Packager : Red Hat, Inc. <http://bugzilla.redhat.com/bugzilla> URL : http://www.cl.cam.ac.uk/Research/SRG/netos/xen/index.html Summary : Xen is a virtual machine monitor Adrian -- Fedora-xen mailing list Fedora-xen(a)redhat.com https://www.redhat.com/mailman/listinfo/fedora-xen

1 0

[RFC] how to mkinitrd of domU after updating kernel-xen.rpm
by Akio Takebe 20 Oct '06

20 Oct '06

Hi, Current kernel-xen.rpm make only initrd of dom0. So we need to make initrd of domU. But when we make initrd of domU, mkinird command include modules for dom0. (e.g. e1000, mptscsih, ... these are not used by domU) And we need to copy new modlues into guestOS image. How should we make initrd of domU? Or already FC6 have another useful commands? https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=207241 Best Regards, Akio Takebe

1 0

re: [Fedora-xen] kernel-xen0-2.6.18-1.2200.fc5 amusingly unhappy
by Tom Wilkie 18 Oct '06

18 Oct '06

Just to confirm, I'm having the same issues. Tom

1 0

Re: [Fedora-xen] Problem assigning a usb or pci port to a domU
by Pascal Parrot 14 Oct '06

14 Oct '06

Atsushi-san, I tried to re compile the kernel but I ran into problems. In other words, I could not make it work. So I reverted back to the idea of loading the kernel xen0 in the domU instead of compiling a new version of xenU. I think I was able to pass the pci and usb to the domU by doing the following: 1) dom0: install a guest with xenguest-install.py. 2) dom0: Start the guest and log in the guest. 3) domU: Then install kernel-xen0 "yum install xen kernel-xen0". 4) domU: Still in the guest, create a initrd file "mkinitrd --with=xennet --preload=xenblk /boot/initrd-2.6.17-1.2187_FC5xen0-test.img 2.6.17-1.2187_FC5xen0" as explained in http://wiki.xensource.com/xenwiki/Fedora5Xen0DomU. Once this file is created, modify the guest /etc/grub.conf xen0 initrd entry with the file that was newly created. 5) dom0: Modify /etc/modprobe.conf in dom0 as described in http://wiki.xensource.com/xenwiki/Assign_hardware_to_DomU_with_PCIBack_as_m… option 2) 6) dom0: The configuration file of the domU uses the option bootloader="/usr/bin/pygrub" (no kernel parameter, no ramdisk parameter) and specifies the parameter "pci = [ <add the lsusb or lspci values> ]". 7) domU: Reload the dom0 and select the xen0 kernel. The usb and pci should be seen in the domU I have not tested the devices but at least they are now showing up in the guest domain when I do a lspci or lsusb. Also, I made a lot of changes trying to make this work so hopefully, the steps above are complete and accurate. Best regards, Pascal >From: Atsushi SAKAI <sakaia(a)jp.fujitsu.com> >To: Pascal Parrot <pascal_parrot(a)hotmail.com>, fedora-xen(a)redhat.com >Subject: Re: [Fedora-xen] Problem assigning a usb or pci port to a domU >Date: Wed, 11 Oct 2006 15:26:46 +0900 > >Hi, Pascal > > Our method needs kernel recompilation. > >I want to ask you your environment. >1)Do You plan to do it w/ xenguest-install.py? > (We are not succeeded yet since pciback.hide is not passed to xen0) >or >2)Do You plan to do from compile? > (We are just using the original xen's >buildconfigs/linux-defconfig_xenU_x86_32) > and Set > CONFIG_PCI=y > CONFIG_XEN_PCIDEV_FRONTEND=y > for DomU > and > CONFIG_XEN_PCIDEV_BACKEND_PASS=y > for Dom0 > >For more detail, see the original Xen file of >linux-sparse/drivers/xen/pciback/Makefile. > > >Thanks >Atsushi SAKAI > > > Yoshi-san, Thank you for the link. > > > > > > The link suggests to build from the source and set the parameters >CONFIG_PCI and CONFIG_XEN_PCIDEV_FRONTEND to "y". These parameters are set >to "y" in xen0 and the xen documentation says that it is possible to use >xen0 in the domains and it might be easier to do than to build from the >source. quote "The ``0'' suffixed privileged versions can be used to boot >the system, as well as in driver domains and unprivileged domains." > > > > How can I make my domU use the xen0 kernel? > > > > > > I downloaded xen0 in the domU but the domain will not boot from it and >it gives me an error (Error: (22, 'Invalid argument')). When I use >the"kernel" parameter in the configuration file of the domU, it does not >seem to have any effect. I am still prompted for the kernel to use (grub). > > > > Pascal > > > > >

1 0

using the xenguest-install.py
by Matt Herzog 13 Oct '06

13 Oct '06

Hello All. I have been attempting to get a Xen guest to boot on a Dell 2650 running FC5. The machine has a 205GB hardware RAID disk controlled by a Dell PERC 3/Di rev 1. The machine is fully updated via yum to all the stock repos. It's running the Xen package from Fedora.org Here is the fstab file for the host domain (Domain-0): ------------------------------------------------------------------------ /dev/sda2 / ext3 defaults 1 1 /dev/sda1 /boot ext3 defaults 1 2 devpts /dev/pts devpts gid=5,mode=620 0 0 tmpfs /dev/shm tmpfs defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs defaults 0 0 /dev/sda3 swap swap defaults 0 0 ------------------------------------------------------------------------ Here is the output of fdisk -l ------------------------------------------------------------------------ Device Boot Start End Blocks Id System /dev/sda1 * 1 13 104391 83 Linux /dev/sda2 14 1033 8193150 83 Linux /dev/sda3 1034 1294 2096482+ 82 Linux swap / Solaris /dev/sda4 1295 26770 204635970 5 Extended /dev/sda5 1295 26770 204635938+ 83 Linux ------------------------------------------------------------------------ I have attempted to use LVM-backed VBDs as instructed here: http://tx.downloads.xensource.com/downloads/docs/user/#SECTION0333000000000… Here is my config file for the LVM-backed VBD named, psdev0 ------------------------------------------------------------------------ def vmid_check(var, val): val = int(val) if val <= 0: raise ValueError return val xm_vars.var('vmid', use="Virtual machine id. Integer greater than 0.", check=vmid_check) xm_vars.check() kernel = "/boot/vmlinuz-2.6.17-1.2187_FC5xen0" ramdisk = "/boot/initrd-2.6.17-1.2187_FC5xen0.img" memory = 256 name = "VM%d" % vmid vcpus = 2 vif = [ '' ] disk = [ 'phy:vg/psdev0,sda1,w' ] dhcp="dhcp" root = "/dev/sda1" ------------------------------------------------------------------------ "xm info" has this to say about the running Domain-0: ------------------------------------------------------------------------ [root@xiamen xen]# xm info host : xiamen.cinteractive.com release : 2.6.17-1.2187_FC5xen0 version : #1 SMP Mon Sep 11 02:05:21 EDT 2006 machine : i686 nr_cpus : 4 nr_nodes : 1 sockets_per_node : 2 cores_per_socket : 1 threads_per_core : 2 cpu_mhz : 2790 hw_caps : 00000000:00000000:bfebfbff:00000000:00000000:00000080:00004400 total_memory : 3840 free_memory : 3674 xen_major : 3 xen_minor : 0 xen_extra : -unstable xen_caps : xen-3.0-x86_32 platform_params : virt_start=0xfc000000 xen_changeset : unavailable cc_compiler : gcc version 4.1.1 20060525 (Red Hat 4.1.1-1) cc_compile_by : brewbuilder cc_compile_domain : build.redhat.com cc_compile_date : Mon Sep 11 01:13:36 EDT 2006 Here's what I see when I run, "xm create -c psdev0 vmid=1" ---------------------------------------------------------------------------- i8042.c: No controller found. mice: PS/2 mouse device common for all mice md: md driver 0.90.3 MAX_MD_DEVS=256, MD_SB_DISKS=27 md: bitmap version 4.39 TCP bic registered Initializing IPsec netlink socket NET: Registered protocol family 1 NET: Registered protocol family 17 Using IPI No-Shortcut mode XENBUS: Device with no driver: device/vbd/2049 XENBUS: Device with no driver: device/vif/0 md: Autodetecting RAID arrays. md: autorun ... md: ... autorun DONE. 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) No matter what I set the root= option to be, the result is the same. Oh, one last config file: my grub.conf from Domain-0 ---------------------------------------------------------------------------- default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title XenCore (2.6.17-1.2187_FC5xen0) root (hd0,0) kernel /xen.gz-2.6.17-1.2187_FC5 module /vmlinuz-2.6.17-1.2187_FC5xen0 ro root=/dev/sda2 ro console=tty0 module /initrd-2.6.17-1.2187_FC5xen0.img title Fedora Core (2.6.17-1.2187_FC5smp) root (hd0,0) kernel /vmlinuz-2.6.17-1.2187_FC5smp ro root=LABEL=/ initrd /initrd-2.6.17-1.2187_FC5smp.img title Fedora Core (2.6.15-1.2054_FC5smp) root (hd0,0) kernel /vmlinuz-2.6.15-1.2054_FC5smp ro root=LABEL=/ initrd /initrd-2.6.15-1.2054_FC5smp.img Any advice would be welcome.

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

xen October 2006