xen May 2008

xen@lists.stg.fedoraproject.org

29 participants
27 discussions

Guest shutdown problems
by Christian Axelsson 02 Apr '09

02 Apr '09

Hello I have a problem shutting down xen guests. Using xm shutdown guests get shut down but gets stuck in state '---s-d' or sometimes '------'. When trying to clone a domain when in this state (my original purpose of the whole operation) I'll get the error: [root@hydra virtinst--devel]# ./virt-clone -o minimal -n new_img -f /var/lib/xen/images/new_img.img ERROR: virDomainGetXMLDesc() failed failed Xen syscall xenDaemonDomainDumpXMLByID failed to find this domain -490299505 The same errors occurs when for example trying to attach to the console using virsh. I have tried to use 'xm destroy' to kill the guest the hard way but it has no effect - the state remains unchanged. I have also tried this on a few different guest installations with the same result. A thing worth noting is that the output from 'xm list --long' differs, I've attached the out put pre boot, after boot and after shutdown. Note how all the devices in the guests are missing after shutdown. Both the hosts and the guests are fedora 8 installations. Regards, Christian Axelsson smiler(a)lanil.mine.nu [?1034h(domain (domid 0) (on_crash restart) (uuid 00000000-0000-0000-0000-000000000000) (bootloader_args ) (vcpus 2) (name Domain-0) (on_poweroff destroy) (on_reboot restart) (bootloader ) (maxmem 16777215) (memory 1491) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (cpu_time 1644.84369405) (online_vcpus 2) (image (linux (kernel ))) (status 2) (state r-----) ) (domain (domid 2) (on_crash restart) (uuid a7638797-e237-3891-5e64-390f828238ca) (bootloader_args ) (vcpus 1) (name minimal) (on_poweroff destroy) (on_reboot restart) (bootloader /usr/bin/pygrub) (maxmem 512) (memory 512) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (start_time 1206360333.14) (cpu_time 9.753408915) (online_vcpus 1) (image (linux (kernel ) (notes (FEATURES 'writable_page_tables|writable_descriptor_tables|auto_translated_physmap|pae_pgdir_above_4gb|supervisor_mode_kernel' ) (VIRT_BASE 18446744071562067968) (GUEST_VERSION 2.6) (PADDR_OFFSET 18446744071562067968) (GUEST_OS linux) (HYPERCALL_PAGE 18446744071564189696) (LOADER generic) (SUSPEND_CANCEL 1) (ENTRY 18446744071564165120) (XEN_VERSION xen-3.0) ) ) ) (status 2) (state -b----) (store_mfn 196619) (console_mfn 196618) (device (vif (bridge xenbr0) (mac 00:16:3e:3f:93:b8) (script vif-bridge) (uuid 94afd732-920b-2e0b-b3d5-e79174754a80) (backend 0) ) ) (device (vbd (uname file:/var/lib/xen/images/minimal.img) (uuid 8f4f4da3-5f8a-3fee-28e8-41dc49e876cd) (mode w) (dev xvda:disk) (backend 0) (bootable 1) ) ) (device (console (protocol vt100) (location 2) (uuid 0046f2d3-058b-d524-9273-f1dac2ca950b) ) ) ) [?1034h(domain (domid 0) (on_crash restart) (uuid 00000000-0000-0000-0000-000000000000) (bootloader_args ) (vcpus 2) (name Domain-0) (on_poweroff destroy) (on_reboot restart) (bootloader ) (maxmem 16777215) (memory 1491) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (cpu_time 1648.92600832) (online_vcpus 2) (image (linux (kernel ))) (status 2) (state r-----) ) (domain (domid 2) (on_crash restart) (uuid a7638797-e237-3891-5e64-390f828238ca) (bootloader_args ) (vcpus 1) (name minimal) (on_poweroff destroy) (on_reboot restart) (bootloader /usr/bin/pygrub) (maxmem 512) (memory 512) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (start_time 1206360333.14) (cpu_time 13.048743365) (online_vcpus 1) (image (linux (kernel ) (notes (FEATURES 'writable_page_tables|writable_descriptor_tables|auto_translated_physmap|pae_pgdir_above_4gb|supervisor_mode_kernel' ) (VIRT_BASE 18446744071562067968) (GUEST_VERSION 2.6) (PADDR_OFFSET 18446744071562067968) (GUEST_OS linux) (HYPERCALL_PAGE 18446744071564189696) (LOADER generic) (SUSPEND_CANCEL 1) (ENTRY 18446744071564165120) (XEN_VERSION xen-3.0) ) ) ) (status 0) (state ---s-d) (store_mfn 196619) (console_mfn 196618) ) [?1034h(domain (domid 0) (on_crash restart) (uuid 00000000-0000-0000-0000-000000000000) (bootloader_args ) (vcpus 2) (name Domain-0) (on_poweroff destroy) (on_reboot restart) (bootloader ) (maxmem 16777215) (memory 1491) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (cpu_time 1635.21430615) (online_vcpus 2) (image (linux (kernel ))) (status 2) (state r-----) ) (domain (on_crash restart) (uuid a7638797-e237-3891-5e64-390f828238ca) (bootloader_args ) (vcpus 1) (name minimal) (on_poweroff destroy) (on_reboot restart) (bootloader /usr/bin/pygrub) (maxmem 512) (memory 512) (shadow_memory 0) (cpu_weight 256) (cpu_cap 0) (features ) (on_xend_start ignore) (on_xend_stop ignore) (start_time 1206309092.82) (cpu_time 0.0) (image (linux (kernel ) (notes (FEATURES 'writable_page_tables|writable_descriptor_tables|auto_translated_physmap|pae_pgdir_above_4gb|supervisor_mode_kernel' ) (VIRT_BASE 18446744071562067968) (GUEST_VERSION 2.6) (PADDR_OFFSET 18446744071562067968) (GUEST_OS linux) (HYPERCALL_PAGE 18446744071564189696) (LOADER generic) (SUSPEND_CANCEL 1) (ENTRY 18446744071564165120) (XEN_VERSION xen-3.0) ) ) ) (status 0) (device (vif (bridge xenbr0) (mac 00:16:3e:3f:93:b8) (backend 0) (uuid 94afd732-920b-2e0b-b3d5-e79174754a80) (script vif-bridge) ) ) (device (vbd (uuid 8f4f4da3-5f8a-3fee-28e8-41dc49e876cd) (bootable 1) (driver paravirtualised) (dev xvda:disk) (uname file:/var/lib/xen/images/minimal.img) (mode w) (backend 0) ) ) (device (console (protocol vt100) (location 2) (uuid 0046f2d3-058b-d524-9273-f1dac2ca950b) ) ) )

3 3

Cannot start HVM DomU
by Robert Locke 04 Jul '08

04 Jul '08

This has newly started, so I am presuming that it is perhaps related to some recently updated package on F8 in say the last week or so. I have a WindowsXP DomU that when I try to start from virt-manager gives the following error: Error starting domain: virDomainCreate() failed POST operation failed: (xend.err 'int argument required') Details shows: Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/engine.py", line 472, in run_domain vm.startup() File "/usr/share/virt-manager/virtManager/domain.py", line 379, in startup self.vm.create() File "/usr/lib/python2.5/site-packages/libvirt.py", line 240, in create if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) libvirtError: virDomainCreate() failed POST operation failed: (xend.err 'int argument required') I have no idea what other information to supply or where to particularly find it since this has generally "just worked" for the occasional running of a Windows instance that I have needed for the last several months. Thanks for any help, --Rob

2 2

RE: [Fedora-xen] Strange mouse issue (F7 & F8 xen kernel with vncviewer running)
by Dustin Henning 30 May '08

30 May '08

I spoke too soon... I just tried krdc and the same thing happened before I could even try to connect to anything with it (or even determine if it supports vnc, I don't remember). This is a quad core, and I have, in all instances, confirmed that each of the four vcpus in dom0 is tied to a different core. Additionally, I have never had more than one DomU per core on top of Dom0 using each core. Finally, in this latest instance, only one DomU was doing anything, and it could not have been using but 100% of one core, so the other three were, for lack of a better word, bored. -----Original Message----- From: fedora-xen-bounces(a)redhat.com [mailto:fedora-xen-bounces@redhat.com] On Behalf Of Dustin Henning Sent: Friday, May 30, 2008 16:48 To: fedora-xen(a)redhat.com Subject: [Fedora-xen] Strange mouse issue (F7 & F8 xen kernel with vncviewer running) I have been experiencing this (rather strange) issue for as long as I have been a member of this group, and I am curious as to whether anyone else has experienced it. The mouse will continue on in the direction I am moving it after I stop, change direction, or even suddenly reverse direction. Once this happens, the mouse inevitably ends up in one of the four corners of the screen, and will eventually stop trying to travel in that direction, at which point I can control it again. It seems like hitting some keys may cause this to happen sooner. The keys I most often go for are Esc, Alt, Ctrl, Super, Alt+F1, Arrows. Which keys usually depends on what is up in the vncviewer window, where I don't want to issue the wrong command to whatever client I may be working with. I have made the following observations regarding this behavior, though I can't say with absolute certainty that they are all 100% accurate and/or consistent: -keyboard and mouse are PS/2 -in all instances, hvm guests were running, consuming as much as 5 out of 8 GiB of memory on this machine -never experienced in standard kernel on F7 or F8 (even with multiple kvm guests and vncviewers running, though not to consume 5 of 8 GiB) -originally, in F7, vncviewer may not have been running every time it happened -now, in F8, vncviewer has definitely been running every time it happened, but not necessarily connected to anything -same issue was previously experienced on another machine with less memory (and less guests) on F7 Is this a known issue of any sort? It has recently occurred to me that I should try a different client, and I will do that at some point, but I figured it wouldn't hurt to see if there have been any other similar experiences out there in the meantime. Thanks, Dustin -- Fedora-xen mailing list Fedora-xen(a)redhat.com https://www.redhat.com/mailman/listinfo/fedora-xen

1 0

Strange mouse issue (F7 & F8 xen kernel with vncviewer running)
by Dustin Henning 30 May '08

30 May '08

I have been experiencing this (rather strange) issue for as long as I have been a member of this group, and I am curious as to whether anyone else has experienced it. The mouse will continue on in the direction I am moving it after I stop, change direction, or even suddenly reverse direction. Once this happens, the mouse inevitably ends up in one of the four corners of the screen, and will eventually stop trying to travel in that direction, at which point I can control it again. It seems like hitting some keys may cause this to happen sooner. The keys I most often go for are Esc, Alt, Ctrl, Super, Alt+F1, Arrows. Which keys usually depends on what is up in the vncviewer window, where I don't want to issue the wrong command to whatever client I may be working with. I have made the following observations regarding this behavior, though I can't say with absolute certainty that they are all 100% accurate and/or consistent: -keyboard and mouse are PS/2 -in all instances, hvm guests were running, consuming as much as 5 out of 8 GiB of memory on this machine -never experienced in standard kernel on F7 or F8 (even with multiple kvm guests and vncviewers running, though not to consume 5 of 8 GiB) -originally, in F7, vncviewer may not have been running every time it happened -now, in F8, vncviewer has definitely been running every time it happened, but not necessarily connected to anything -same issue was previously experienced on another machine with less memory (and less guests) on F7 Is this a known issue of any sort? It has recently occurred to me that I should try a different client, and I will do that at some point, but I figured it wouldn't hurt to see if there have been any other similar experiences out there in the meantime. Thanks, Dustin

1 1

Xen Documentation
by Gastón Keller 30 May '08

30 May '08

Hello, everybody. I have been struggling with Xen's documentation lately, not finding where to look for accurate and up-to-date info. I remember having read in this list months ago someone saying that the documentation was quite outdated. For example, the xm command _delete_ does not appear in the man pages and many guides still do reference to the guest's config files being stored at /etc/xen (two days ago I discovered that now Xen works in a different way). My objective is not only to learn how to use Xen, but understand how it works. I'm doing my best looking for info in the net and going through docs in the Xen Wiki, but I might be missing something in my search. Therefore, I would really appreciate if someone could point me to some documentation I should read, or whatever else you could think of. Thanks, Gaston -- La única verdad es la realidad.

6 11

Re: [Fedora-xen] Xen Documentation
by Emre Erenoglu 30 May '08

30 May '08

I totally share your view. :) I'm in the same condition. Emre On Fri, May 30, 2008 at 2:12 PM, Dustin Henning <Dustin.Henning(a)prd-inc.com> wrote: > Emre, > As you may already know, I still use the config files in /etc/xen > and don't mess with the virsh stuff. It all certainly worked fine in F7, > and though I haven't fully tested, it appears that it all works fine in F8 > as well. I use them for one, because I learned with them, and for two, > because in the little time I have spent messing with the gui tools, I > haven't found near as much control. That isn't to say that more can be done > with the old way, but instead to say that I can do more with the old way. I > don't know if there are any plans to deprecate that stuff, but I hope that > if there ever are, news makes the list before I end up on a version where I > have no clue how to do anything! > Dustin > > From: fedora-xen-bounces(a)redhat.com [mailto:fedora-xen-bounces@redhat.com] > On Behalf Of Emre ERENOGLU > Sent: Friday, May 30, 2008 04:30 > To: Gastón Keller > Cc: fedora-xen(a)redhat.com > Subject: Re: [Fedora-xen] Xen Documentation > > Hmm, > On Fri, May 30, 2008 at 2:06 AM, Gastón Keller <gastonkeller(a)gmail.com> > wrote: > On Thu, May 29, 2008 at 7:53 PM, Emre ERENOGLU <erenoglu(a)gmail.com> wrote: > > Hi Gaston, > > > > It's true that some documentation is outdated. However, for the > "/etc/xen" > > isssue, my custom Xen 3.2 setup still stores ( and I do store there also) > my > > config files there. > > > > Maybe what you're saying is "Fedora Specific" which I could understand as > > it's fedora mailing list. > _The Fedora team has followed the Xensource model and begun to store > all VM configuration details in a database, referred to as xenstore._ > Source: > http://enterpriselinuxlog.blogs.techtarget.com/2007/06/07/fedora-7-xen-firs… > > OK, I didn't know that. However, I find this xenstore thing a bit > complicated. > > Then, you might be right. I interpreted it was a modification from > Xen, not from Fedora (previously introduced by Xensource). > > So, what is the advantage behind this movement? > > I think it's managegability. > > > Thanks, > Gaston > > However, I never found the flexibility and control > > in the Fedora way of Xen implementation (all those xml stuff etc.). most > > probably, I'm wrong :) > > > > Emre > > > > On Thu, May 29, 2008 at 9:22 PM, Gastón Keller <gastonkeller(a)gmail.com> > > wrote: > >> > >> Hello, everybody. I have been struggling with Xen's documentation > >> lately, not finding where to look for accurate and up-to-date info. I > >> remember having read in this list months ago someone saying that the > >> documentation was quite outdated. For example, the xm command _delete_ > >> does not appear in the man pages and many guides still do reference to > >> the guest's config files being stored at /etc/xen (two days ago I > >> discovered that now Xen works in a different way). > >> > >> My objective is not only to learn how to use Xen, but understand how > >> it works. I'm doing my best looking for info in the net and going > >> through docs in the Xen Wiki, but I might be missing something in my > >> search. Therefore, I would really appreciate if someone could point me > >> to some documentation I should read, or whatever else you could think > >> of. > >> > >> Thanks, > >> Gaston > >> > >> -- > >> La única verdad es la realidad. > >> > >> -- > >> Fedora-xen mailing list > >> Fedora-xen(a)redhat.com > >> https://www.redhat.com/mailman/listinfo/fedora-xen > > > > > > > > -- > > Emre > > > -- > La única verdad es la realidad. > > > > -- > Emre > > > -- Emre

1 0

RE: [Fedora-xen] HVM DomU on F9?
by Dustin Henning 30 May '08

30 May '08

For anyone who might have been following this thread, I accidentally forgot to add fedora-xen to the recipient list before sending my last reply. It is below, along with a response from Avi. I updated my F8 box, and sure enough, kvm-60 was installed. I thought I had just done that and kvm-55 was where it stopped, but I must be crazy or something. Also, I got the gplpv drivers by James Harper installed and functional as of version 0.9.2 (I had no luck with 0.9.0 or any prior version). That said, I probably won't go any further with my quest for kvm, at least not in the near future. The stdvga emulation kvm uses (or at least used on kvm-47) was nice with the standard vesa drivers referenced on the kvm site (went higher than 1024x768), and those drivers don't work for the stdvga=1 option in xen, but I think the potential performance gain from the multiple paravirtual drivers will be more important for me. Also, as I didn't follow up on it, I did try to install CentOS 5.1 on this machine and it does not support the network card, so I am on Fedora 8 as originally recommended by Emre. Dustin -----Original Message----- From: Avi Kivity [mailto:avi@qumranet.com] Sent: Friday, May 30, 2008 03:13 To: Dustin.Henning(a)prd-inc.com Subject: Re: [Fedora-xen] HVM DomU on F9? [please use the list for questions like this] Dustin Henning wrote: > It doesn't look like F8 is likely to get up to kvm-58 either. F8 is up to kvm-60. > I > prefer to stick with the Fedora repositories for the machine I am working > with, as I don't want to run into compatibility issues on account of some > future update. However, I want to try to use the network driver, as I am > hopeful that it might make an apache application I am running perform > better. It is also possible that I really need disk drivers for that. > Would compiling a newer version of kvm on Fedora 8 be risky (in terms of > interoperability problems after some future update)? No. > Also, assuming the > answer to that is no, do I need to remove any certain packages first to make > sure they don't update and overwrite it? No. Do a ./configure --prefix=/opt/kvm-69 and all files will be placed there. > Finally, on the Fedora repo > versions of KVM for F7 and F8, I can't create a VM with 2 GiB of RAM. Was > this a known kvm bug at some point (or worse, is it still?), or could it be > a problem with Fedora (I can do this fine in xen)? Thanks, > For more than 2GB of RAM in a guest, you need a 64-bit hypervisor. Large guest support was added in kvm-47 (2TB guests were booted). -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.

1 0

Is pci-pasthrough enabled for F9 DomU?
by snowcrash+xen＠gmail.com 29 May '08

29 May '08

i've F8 as Dom0. for F8 DomU, pci-passthru devices are visible via lspci *in* DomU. for a new install of F9 DomU, lcpci is (empty). there was discussion that PCI-PassThrough might not make it into F9, http://lists.xensource.com/archives/html/xen-devel/2007-12/msg00454.html "> F9 may well end up lacking features compared to Xen in Fedora 8 & earlier (eg no > PCI device passthrough, or CPU hotplug)" but release notes, http://fedoraproject.org/wiki/Features/XenPvops#head-e15c9305e8d715da9d904a… make no mention one way or other. *is* pci-passthru known/supposed to work in F9 DomU? if so, then i've got a different issue ...

7 22

configure: error: GRUB requires a working absolute objcopy; upgrade your binutils
by Dustin Henning 28 May '08

28 May '08

I hope this isn't off topic since the bug I am trying to resolve only affects the xen kernel... I am trying to compile grub on Fedora 8 because I need to apply the patch for bug 250299. First, I downloaded and installed http://mirrors.kernel.org/fedora/releases/8/Fedora/source/SRPMS/grub-0.97-1… Next, I extracted the included grub-0.97.tar.gz to /usr/src/ At this point, if I run configure, I will get the error mentioned in the subject, as follows: [CODE]checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for gawk... gawk checking whether make sets $(MAKE)... yes checking build system type... x86_64-unknown-linux-gnu checking host system type... x86_64-unknown-linux-gnu checking whether to enable maintainer-specific portions of Makefiles... no checking for gcc... gcc checking for gcc... (cached) gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for style of include used by make... GNU checking dependency style of gcc... gcc3 checking dependency style of gcc... (cached) gcc3 checking for ranlib... ranlib checking whether optimization for size works... yes checking whether gcc has -fno-stack-protector... yes checking whether -Wundef works... yes checking whether -falign-loops works... yes checking for objcopy... objcopy checking if C symbols get an underscore after compilation... no checking whether objcopy works for absolute addresses... no configure: error: GRUB requires a working absolute objcopy; upgrade your binutils[/CODE] However, I am running binutils 2.17.50.0.18 which is the latest in the fedora repo, so I cannot upgrade, and I do not know how to downgrade. I saw another issue where this same error was happening, and someone suggested downloading ftp://alpha.gnu.org/gnu/grub/grub-0.97.tar.gz and using it. I tried this and determined that that is the exact file included in the srpm I had downloaded, and the error still surfaces when I try it. Additionally, if I apply the included grub-fedora-8.patch, the configure file disappears. I can get it from the tar again, but I will still get the same error, so while I need to know how I am supposed to compile after applying that patch (which I apply before the patch for the aforementioned bug), but that point is moot until I can determine how to fix this binutils issue.

1 1

I am now on Refriendz!
by Sameer Gupta 28 May '08

28 May '08

Hi! I would like to invite you to visit my Refriendz page and see my latest photos. In order to visit my space, you must go to: http://www.refriendz.com/?do=Login.Invite&rid=sam9823444029&email=fedora-xe… (If this link does not work, please copy and paste it into your browser or go to www.refriendz.com and enter 'sam9823444029' as Invitation ID to Login to the web site.) P.S. Refriendz is Invitation-Only, so do not miss your chance to visit my page! Cheers! Sameer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Unsubscribe: to opt out of ALL future emails from Refriendz, visit: http://www.refriendz.com/?do=Login.RemoveEntry&email=fedora-xen@redhat.com Please do not reply directly to this email. This mailbox is not monitored and you will not receive a response. Refriendz Limited, PO BOX 1184, Luton, Bedfordshire, LU1 9AT, UK.

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

xen May 2008