xen November 2007

xen@lists.fedoraproject.org

39 participants
39 discussions

Fail to install fedora domU
by neohtm 29 Nov '07

29 Nov '07

I just had installed Fedora 7 with virtualization package using installation DVD. I run the virtual manager and try to install a para-virtualize fedora domU. I had followed all the steps which i found in http://www.howtoforge.com/xen_gui_fedora_7_desktop but still it shows that a message that says failed to locate the disk image. I even had 'yum install vnc' and yet the same error still occurs. Do i need to manually create a disk.img file before setting up a domU? Need help on this... Neohtm

1 0

kernel-xen crash dump analysis
by Dale Bewley 28 Nov '07

28 Nov '07

After upgrading to F8 I still had a dom0 crash last night, but didn't catch any output. I think it's still the null pointer dereference in tcp_tso_segment, possibly brought to prominence by traffic shaping in a busy domU. I'd like to catch a crash dump the next time it happens so I can (hopefully) intelligently report this bug somewhere. I'm looking at http://fedoraproject.org/wiki/FC6KdumpKexecHowTo http://www.redhatmagazine.com/2007/08/15/a-quick-overview-of-linux-kernel-c… There is no mention of Xen. I don't see a kernel-xen-debuginfo rpm for f8 or even f7, but there is one for f9 in rawhide it appears. Is kernel-debuginfo sufficient even if running xen? The versions differ, so I would imagine not. # yum --enablerepo=fedora-debuginfo list kernel-xen-debuginfo kernel-debuginfo Available Packages kernel-debuginfo.x86_64 2.6.23.1-49.fc8 updates-debuginf # rpm -q kernel kernel-xen kernel-2.6.23.1-49.fc8 kernel-xen-2.6.21-2950.fc8 Any advice on how capture some useful info from a crashing dom0 kernel? -- Dale Bewley - Unix Administrator - Shields Library - UC Davis GPG: 0xB098A0F3 0D5A 9AEB 43F4 F84C 7EFD 1753 064D 2583 B098 A0F3

2 1

Problem with network install, console late to attach
by John Summerfield 28 Nov '07

28 Nov '07

Complicating my other network install is the problem that by the time the console's displayed, the boot process is all over. The "BIOS" needs to wait longer. -- Cheers John -- spambait 1aaaaaaa(a)coco.merseine.nu Z1aaaaaaa(a)coco.merseine.nu -- Advice http://webfoot.com/advice/email.top.php http://www.catb.org/~esr/faqs/smart-questions.html http://support.microsoft.com/kb/555375 You cannot reply off-list:-)

1 0

Problem with network install, network wrong
by John Summerfield 28 Nov '07

28 Nov '07

I'm using the click-through wizard to try to network-install F8, fully-vertualised, on F8. My dhcpd configuration gives special treatment to PXE clients which it recognises: class "pxeclients" { match if substring (option vendor-class-identifier, 0, 9) = "PXEClient"; option vendor-class-identifier "PXEClient"; # At least one of the vendor-specific option must be set. We set # the MCAST IP address to 0.0.0.0 to tell the bootrom to still use # TFTP (address 0.0.0.0 is considered as "no address") # option PXE.mtftp-ip 192.168.9.1; vendor-option-space PXE; option PXE.mtftp-ip 0.0.0.0; vendor-option-space PXE; } and later, pool { allow members of "pxeclients"; default-lease-time 1800; max-lease-time 3600; range 192.168.9.150 192.168.9.169; filename "/PXE/pxelinux.0"; } I give etherboot clients special (different). treatment too. class "Etherboot" { match if substring (option vendor-class-identifier, 0, 13) = "Etherboot-5.0"; option vendor-encapsulated-options 3c:09:45:74:68:65:72:62:6f:6f:74:ff; option vendor-class-identifier "Etherboot-5.0"; filename "tftpboot.vc"; } and IBM thin clients. Unfortunately, virt-manager tells the user it's using PXE, but something tells the DHCPD server it's etherboot. Here's a bit of 17:57:56.001102 IP (tos 0x0, ttl 60, id 0, offset 0, flags [none], proto 17, length: 576) Possum.demo.lan.bootpc > 255.255.255.255.bootps: [udp sum ok] BOOTP/DHCP, Request from 00:16:3e:3a:0b:74, length: 548, xid:0x3e4202c2, flags: [none] (0x0000) Client Ethernet Address: 00:16:3e:3a:0b:74 Vendor-rfc1048: DHCP:REQUEST SID:ns.demo.lan RQ:Possum.demo.lan MSZ:1500 VC:"Etherboot-5.4" PR:SM+DG+HN+VO+RP+T128+T129+T130+T150 T150:175.5.1.16.236.129.57.177.2.3.0 17:57:56.004481 IP (tos 0x10, ttl 16, id 0, offset 0, flags [none], proto 17, length: 328) ns.demo.lan.bootps > Possum.demo.lan.bootpc: [udp sum ok] BOOTP/DHCP, Reply, length: 300, xid:0x3e4202c2, flags: [none] (0x0000) Your IP: Possum.demo.lan Server IP: ns.demo.lan Client Ethernet Address: 00:16:3e:3a:0b:74 Vendor-rfc1048: DHCP:ACK SID:ns.demo.lan LT:84400 SM:255.255.255.0 DG:ns.demo.lan For this application, the "boot rom" needs to claim it's PXE. btw I can also recognise Macs (good to deal with openfirmware) Anaconda (a shame it won't use the filename I return), debian-installer -- Cheers John -- spambait 1aaaaaaa(a)coco.merseine.nu Z1aaaaaaa(a)coco.merseine.nu -- Advice http://webfoot.com/advice/email.top.php http://www.catb.org/~esr/faqs/smart-questions.html http://support.microsoft.com/kb/555375 You cannot reply off-list:-)

1 0

dom0 F7 crashes tcp_tso_segment oops - new kernel lag time
by Dale Bewley 27 Nov '07

27 Nov '07

Running 2.6.20-2936.fc7xen x86_64 quad CPU, 16G. We have a dom0 with bridges riding on top of a VLAN interface. # brctl show bridge name bridge id STP enabled interfaces br101 8000.00093d139ae9 no eth0 br6 8000.00093d139ae9 no vif2.0 vif1.0 eth0.6 ... Inside a F7 domU on br6 we are running tc (via shorewall) to limit the bandwidth of a mirror server. On Monday I throttled the bandwidth down far below the demand. Since then we are starting to see dom0 crash and reboot with nothing in the log. Crashes happened on Tues around 9am and Thursday (today) around 4am and 9am. I caught the console during the most recent crash: Unable to handle kernel NULL pointer dereference at 0000000000000030 RIP: [<ffffffff803fc79c>] tcp_tso_segment+0x1d8/0x285 PGD 3db8f1067 PUD 3db8f2067 PMD 0 Oops: 0000 [1] SMP last sysfs file: /devices/xen-backend/vbd-2-51712/statistics/wr_sect CPU 1 Modules linked in: loop netbk xenblktap blkbk autofs4 8021q bridge nf_conntrack_netbios_ns ipt_LdPid: 0, comm: swapper Not tainted 2.6.20-2936.fc7xen #1 RIP: e030:[<ffffffff803fc79c>] [<ffffffff803fc79c>] tcp_tso_segment+0x1d8/0x285 RSP: e02b:ffff880002f77950 EFLAGS: 00010216 RAX: 0000000000007a21 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000010130000 RSI: 0000000000000000 RDI: 0000000087e8b626 RBP: 000000000000fa87 R08: 0000000187e8b625 R09: 0000000000000000 R10: 000000009b9a7b25 R11: 0000000000000003 R12: ffff88034ecab034 R13: 0000000017acfe00 R14: 0000000000000020 R15: 00000000ffff0000 FS: 00002aaaab0ff230(0000) GS:ffffffff80580080(0000) knlGS:0000000000000000 We are up to date with patches. After the first 2 crashes and before the 3rd I upgraded from xen-3.1.0-6.fc7 to xen-3.1.0-8.fc7 and rebooted dom0 for good measure. I did some googling and found this: http://lists.openwall.net/netdev/2007/02/09/16 which seems pretty similar. Assuming this is the problem and there is a fix I'm left to ask what is the ETA for a new xen kernel and what is the typical lag time? Current state is 2.6.23.1-21.fc7 vs. 2.6.20-2936.fc7xen -- Dale Bewley - Unix Administrator - Shields Library - UC Davis GPG: 0xB098A0F3 0D5A 9AEB 43F4 F84C 7EFD 1753 064D 2583 B098 A0F3

2 3

Paravirt Gutsy Gibbon on F8?
by Neil Thompson 26 Nov '07

26 Nov '07

Hi all, Before I tear out what little hair I have left, has anyone managed to get Gutsy running in paravirt on F8? If so, please could you tell me how? (In short, simple sentences :-) I've tried just about everything, from installing it under KVM/QEMU and cutting the partitions out, to downloading xen images and nothing works properly - even if I run using the ubuntu supplied kernels. The downloaded from xensource kernels sort of work, until I try supplying an initrd to get consoles and things like that working and then the vm just sits there, eating up CPU. The ubuntu xen kernels also just boot to a particular point and stop - also eating CPU. Logs/dmesg available on request. Unfortunately, some people at work without much taste ;-} have decided they need to use ubuntu, and I have no wish to change my OS while labbing this stuff. TIA -- Cheers! (Relax...have a homebrew) Neil THEOREM: VI is perfect. PROOF: VI in roman numerals is 6. The natural numbers < 6 which divide 6 are 1, 2, and 3. 1+2+3 = 6. So 6 is a perfect number. Therefore, VI is perfect. QED -- Arthur Tateishi

1 0

ppc64 xen support
by Spencer Shimko 26 Nov '07

26 Nov '07

Hi, I was looking for any information people may have on running Fedora 8 Xen on a PPC64 platform. Xen appears to support ppc64 but none of the Fedora pages seem to mention it and Fedora doesn't have a ppc64 Xen kernel. Anyone taken a swing at this or have any thoughts on where I should focus my efforts? Thanks, Spencer

2 2

VMs configuration files
by Patricio A. Bruna 26 Nov '07

26 Nov '07

Hi, How can i define another folder for store the VMs configurations files, for default is /etc/xen Thanks.

2 1

Newbie Virt Questions and comments on wiki/Docs/Fedora8VirtQuickStart
by Aaron Metzger 25 Nov '07

25 Nov '07

I am trying to migrate my virtualization infrastructure from a pure Xen implementation to the more versatile Fedora virt-manager Xen/KVM environment. While doing this, I have hit some snags related to areas in which pure Xen documentation no longer applies due to the presence of lib-virt et al. First a minor comment on the Fedora8VirtQuickStart (draft I know): The section: > Building a Fedora Guest System using `virt-manager` no longer applies because between Fedora 7 and Fedora 8 there no longer appears to be a "New" button or a "File New" option in virt-manager. It would appear that the only way to create a DomU is by way of virt-install. Now, my problem. I have successfully created fully functional guest VMs which work perfectly in every way but, in the context of lib-virt and virsh, I can not figure out how to get the guest machines to be started automatically when the host machine boots. In a pure Xen environment, you symbolically link the "/etc/xen/<vm-config>" file into "/etc/xen/auto". After using virt-install to create the guest, the domain configuration files end up over in "/var/lib/xend/domains". I have tried to symbolically link the whole domain directory to "/etc/xen/auto", I have tried to link the configuration file to "/etc/xen/auto". I have tried to export the configuration as XML using virsh and then put that under "/etc/xen/auto". Nothing works and there is no indication from the xend.log that it is even trying to start the guest domains on bootup. I also tried to use: virsh autostart <domain> and get the error: libvir: error : this function is not supported by the hypervisor: virDomainSetAutostart So, my question is, is it possible to properly manage the start up and shut down of guest domains on host start up and shutdown or is that just not supported yet. If there is a way to do this, it would be a very helpful section to add to the Quick Start document because I would expect anyone migrating from a pure Xen environment to have the same issue. Google reveals that several have asked this question in the past on Fedora 7 etc and I have not seen any answer other than people suggesting to link the configuration file under "/etc/xen/auto". Thanks, Aaron

5 8

Re: unionfs question
by Richard W.M. Jones 25 Nov '07

25 Nov '07

Christoph Höger wrote: > I have a more general question about unionfs: > I want to build a xen machine with a save base-image and a temporary > write image that can be saved or thrown away on demand. So I thought > about using unionfs to mount both these images as /. > The problem is: unionfs does not mount images but directories that are > already mounted. So I would need to mount those two images somewhere > else. Obviuosly this would happen somewhere above / (e.g. /workimage and > /rootimage). Could unionfs handle this? > Or is there even a better way? LVM + snapshots. http://tldp.org/HOWTO/LVM-HOWTO/snapshotintro.html BTW, there is a fedora-xen list for discussion of Xen issues. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

xen November 2007