From jklein at saugus.k12.ca.us Mon Dec 1 03:02:00 2008 From: jklein at saugus.k12.ca.us (Jim Klein) Date: Sun, 30 Nov 2008 19:02:00 -0800 (PST) Subject: [Fedora-xen] RHEL Update Virt-Manager VNC Problems In-Reply-To: <1199033.1021228100229669.JavaMail.root@do8> Message-ID: <25814968.1101228100520805.JavaMail.root@do8> Just updated all my RHEL4 VMs and RHEL 5 hosts, and virt-manager can no longer connect to the consoles on the vms. Tried recreating VNC adapters in virt-manager to no avail. VMs running kernel 2.6.9-78.0.8.ELxenU. Must be missing something, but not sure where to look. Any assistance would be greatly appreciated. -------------- next part -------------- An HTML attachment was scrubbed... URL: From berrange at redhat.com Mon Dec 1 10:27:22 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 1 Dec 2008 10:27:22 +0000 Subject: [Fedora-xen] RHEL Update Virt-Manager VNC Problems In-Reply-To: <25814968.1101228100520805.JavaMail.root@do8> References: <1199033.1021228100229669.JavaMail.root@do8> <25814968.1101228100520805.JavaMail.root@do8> Message-ID: <20081201102722.GB24079@redhat.com> On Sun, Nov 30, 2008 at 07:02:00PM -0800, Jim Klein wrote: > Just updated all my RHEL4 VMs and RHEL 5 hosts, and virt-manager can no > longer connect to the consoles on the vms. Tried recreating VNC adapters > in virt-manager to no avail. VMs running kernel 2.6.9-78.0.8.ELxenU. Must > be missing something, but not sure where to look. Any assistance would be > greatly appreciated. Use 'virsh dumpxml' to look at the running domain config. It should show you a element like "". If the port number is -1, then its failed to start VNC for some reason. If the graphics tag is missing completely then the config file is wrong. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From jklein at saugus.k12.ca.us Mon Dec 1 17:17:05 2008 From: jklein at saugus.k12.ca.us (Jim Klein) Date: Mon, 1 Dec 2008 09:17:05 -0800 (PST) Subject: [Fedora-xen] RHEL Update Virt-Manager VNC Problems In-Reply-To: <10188762.45591228151469149.JavaMail.root@do8> Message-ID: <4135503.47271228151825734.JavaMail.root@do8> Sorry, I should have mentioned that I did all that. Interestingly, it seems to be related to the subnet I am in (am using ssh to connect to the host.) I am investigating further, but it appears that If I am on the same subnet as the host, it works. I was working before on the other end of an ipsec tunnel, where I got no love. Pretty sure this all worked before, but will need to spend some more time digging to figure it out. ----- Original Message ----- From: "Daniel P. Berrange" To: "Jim Klein" Cc: "fedora-xen" Sent: Monday, December 1, 2008 2:27:22 AM GMT -08:00 US/Canada Pacific Subject: Re: [Fedora-xen] RHEL Update Virt-Manager VNC Problems On Sun, Nov 30, 2008 at 07:02:00PM -0800, Jim Klein wrote: > Just updated all my RHEL4 VMs and RHEL 5 hosts, and virt-manager can no > longer connect to the consoles on the vms. Tried recreating VNC adapters > in virt-manager to no avail. VMs running kernel 2.6.9-78.0.8.ELxenU. Must > be missing something, but not sure where to look. Any assistance would be > greatly appreciated. Use 'virsh dumpxml' to look at the running domain config. It should show you a element like "". If the port number is -1, then its failed to start VNC for some reason. If the graphics tag is missing completely then the config file is wrong. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| -------------- next part -------------- An HTML attachment was scrubbed... URL: From guillaume.chardin at gmail.com Tue Dec 2 17:34:21 2008 From: guillaume.chardin at gmail.com (Guillaume) Date: Tue, 2 Dec 2008 18:34:21 +0100 Subject: [Fedora-xen] Create multiple loopback devices Message-ID: Hi I would like to know if it is possible to create multiple loopback ethernet device. And if yes, how to create new one. The purpose of this is to attach to many vm this interface, and then, share this private network between these vm. Thanks. -- Guillaume From pasik at iki.fi Tue Dec 9 16:24:12 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Tue, 9 Dec 2008 18:24:12 +0200 Subject: [Fedora-xen] Updating out-of-date Fedora XenPvopsDom0 wiki page In-Reply-To: <1227693446.9579.38.camel@blaa> References: <20081125214440.GH15052@edu.joroinen.fi> <1227667152.28887.26.camel@tofu.lib.ucdavis.edu> <20081126091658.GI15052@edu.joroinen.fi> <1227692057.9579.29.camel@blaa> <20081126094823.GJ15052@edu.joroinen.fi> <1227693446.9579.38.camel@blaa> Message-ID: <20081209162412.GX15052@edu.joroinen.fi> On Wed, Nov 26, 2008 at 09:57:26AM +0000, Mark McLoughlin wrote: > > > > My advice to anyone wanting to help this along is to build and test > > > Jeremy's work and iron out any issues with Fedora's xen userland running > > > on it. > > > > > > > Yes, that would be a good idea. > > > > Does someone want to build some testing RPMs? :) > > A volunteer to do this would be very welcome indeed :-) > If I find some extra time I might try building some testing RPMs with Xen pv_ops dom0 patches included.. Should this new unified kernel-package work out of the box (after adding dom0 patches), or should I use the latest kernel-xen as a starting point? I was thinking about grub menu.lst generation etc.. -- Pasi From markmc at redhat.com Wed Dec 10 09:19:14 2008 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 10 Dec 2008 09:19:14 +0000 Subject: [Fedora-xen] Updating out-of-date Fedora XenPvopsDom0 wiki page In-Reply-To: <20081209162412.GX15052@edu.joroinen.fi> References: <20081125214440.GH15052@edu.joroinen.fi> <1227667152.28887.26.camel@tofu.lib.ucdavis.edu> <20081126091658.GI15052@edu.joroinen.fi> <1227692057.9579.29.camel@blaa> <20081126094823.GJ15052@edu.joroinen.fi> <1227693446.9579.38.camel@blaa> <20081209162412.GX15052@edu.joroinen.fi> Message-ID: <1228900754.5384.5.camel@blaa> On Tue, 2008-12-09 at 18:24 +0200, Pasi K?rkk?inen wrote: > If I find some extra time I might try building some testing RPMs with Xen pv_ops > dom0 patches included.. > > Should this new unified kernel-package work out of the box (after adding > dom0 patches), or should I use the latest kernel-xen as a starting point? > > I was thinking about grub menu.lst generation etc.. Well, the two outstanding work items that no-one is looking at are: * Allow the hypervisor to boot bzImage kernels * Make 'new-kernel-pkg' aware of HYPERVISOR setting in /etc/sysconfig/xen Each of these mean that building the standard kernel with the dom0 patches won't be enough, so you could build a kernel-xen-dom0 RPM which a) ships a vmlinuz kernel instead of bzImage and b) adds xen.gz to grub.conf. In the long run, though, what would be most useful would be if someone started working on those two work items. Cheers, Mark. From guillaume.chardin at gmail.com Thu Dec 11 20:04:33 2008 From: guillaume.chardin at gmail.com (Guillaume) Date: Thu, 11 Dec 2008 21:04:33 +0100 Subject: [Fedora-xen] 4gb seg fix error Message-ID: hi, I'im stuck with an error with a fedora 9&10 domU error :this error is common and i can find workaround/solution on forums and ML, but no one help me. I try to rename/delete the /lib/tls directory, echo "hwcap 0 nosegneg" > /etc/ld.conf.so.d/nosegneg.conf, but it does nothing. I read somewhere that installing "libc6-xen" will correct this behavior but i was unable to find this package. Does someone have an idea ? My config is : dom0 = debian etch domU1 = fedora 9 domU2 = fedora 10 The 2 fedora distrib have this error reported in syslog. >>Dec 11 20:22:27 valinor kernel: printk: 646140 messages suppressed. >>Dec 11 20:22:27 valinor kernel: 4gb seg fixup, process bash (pid 3999), cs:ip 73:0022a7e5 thanks. -- Guillaume From paul at xelerance.com Thu Dec 11 20:22:38 2008 From: paul at xelerance.com (Paul Wouters) Date: Thu, 11 Dec 2008 15:22:38 -0500 (EST) Subject: [Fedora-xen] 4gb seg fix error In-Reply-To: References: Message-ID: On Thu, 11 Dec 2008, Guillaume wrote: > I'im stuck with an error with a fedora 9&10 domU error :this error is > common and i can find workaround/solution on forums and ML, but no one > help me. > I try to rename/delete the /lib/tls directory, echo "hwcap 0 > nosegneg" > /etc/ld.conf.so.d/nosegneg.conf, but it does nothing. Did you reboot after this? > I read somewhere that installing "libc6-xen" will correct this > behavior but i was unable to find this package. That was only for some versions of debian/ubuntu. Paul From guillaume.chardin at gmail.com Thu Dec 11 20:53:31 2008 From: guillaume.chardin at gmail.com (Guillaume) Date: Thu, 11 Dec 2008 21:53:31 +0100 Subject: [Fedora-xen] 4gb seg fix error In-Reply-To: References: Message-ID: 2008/12/11 Paul Wouters : > On Thu, 11 Dec 2008, Guillaume wrote: > >> I'im stuck with an error with a fedora 9&10 domU error :this error is >> common and i can find workaround/solution on forums and ML, but no one >> help me. >> I try to rename/delete the /lib/tls directory, echo "hwcap 0 >> nosegneg" > /etc/ld.conf.so.d/nosegneg.conf, but it does nothing. > > Did you reboot after this? yeah ! I found what's wrong, i have to run ldconfig... Don't ask me why but reboot was not sufficient. Then messages disappear thanks -- Guillaume -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Thu Dec 11 23:29:42 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Thu, 11 Dec 2008 18:29:42 -0500 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 Message-ID: <4941A266.4000901@speakeasy.net> Hello, I'm an experienced IBM mainframe programmer with just a little C and Java coding experience. I have no experience in the programming part of virtualization. I currently do have CentOS 5.2 running as a guest under VMWare workstation (an older version which is about a year old, running on a Fedora 7 host. And yes, I know Fedora 7 is long out of date.) I would like to jump in the water and get my feet wet, though. Can someone like me test Dom0 Xen support in Fedora 10? If so, is there a web resource somewhere spelling out how to get started with testing? I will need the steps involved in patching and building a test kernel pointed out to me. I just bought a new laptop, the Dell Latitude E6400 with the Intel P9500 processor. I think this processor has virtualization technology (vmx extensions?) so the hardware for testing is there. Thanks Bob Cochran From sakaia at jp.fujitsu.com Fri Dec 12 00:09:40 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Fri, 12 Dec 2008 09:09:40 +0900 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 In-Reply-To: <4941A266.4000901@speakeasy.net> References: <4941A266.4000901@speakeasy.net> Message-ID: <20081212000944.D717118099@m020.s.css.fujitsu.com> Hello Dom0 is not supported on Fedora 10. It should wait until upstream kernel support. Or you should install upstream-Xen. For more detail(Fedora status) see below. https://fedoraproject.org/wiki/Features/XenPvopsDom0 Dom0 status: awaiting upstream kernel support Thanks Atsushi SAKAI Robert L Cochran wrote: > Hello, > > I'm an experienced IBM mainframe programmer with just a little C and > Java coding experience. I have no experience in the programming part of > virtualization. I currently do have CentOS 5.2 running as a guest under > VMWare workstation (an older version which is about a year old, running > on a Fedora 7 host. And yes, I know Fedora 7 is long out of date.) > > I would like to jump in the water and get my feet wet, though. Can > someone like me test Dom0 Xen support in Fedora 10? If so, is there a > web resource somewhere spelling out how to get started with testing? I > will need the steps involved in patching and building a test kernel > pointed out to me. > > I just bought a new laptop, the Dell Latitude E6400 with the Intel P9500 > processor. I think this processor has virtualization technology (vmx > extensions?) so the hardware for testing is there. > > Thanks > > Bob Cochran > > > -- > Fedora-xen mailing list > Fedora-xen at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-xen From kraxel at redhat.com Fri Dec 12 10:22:45 2008 From: kraxel at redhat.com (Gerd Hoffmann) Date: Fri, 12 Dec 2008 11:22:45 +0100 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 In-Reply-To: <4941A266.4000901@speakeasy.net> References: <4941A266.4000901@speakeasy.net> Message-ID: <49423B75.8040603@redhat.com> Hi, > I would like to jump in the water and get my feet wet, though. Can > someone like me test Dom0 Xen support in Fedora 10? Nothing upstream yet, nothing packaged yet. The dom0 patch queue seems to reach the state where the system boots now though. If you are willing and able to build your own kernels you can watch the xen-devel list, build and test dom0 pv_ops kernels, to help upstream get the job done. If all goes well the first dom0 support bits should be ready for the next merge window, i.e. make it into 2.6.29. If that happens rawhide should have kernel packages with (probably limited) dom0 support ready for testing early next year. fedora10 packages will have to wait until 2.6.29-final is released, which is probably 3-4 months from now. HTH, Gerd From markmc at redhat.com Fri Dec 12 10:36:20 2008 From: markmc at redhat.com (Mark McLoughlin) Date: Fri, 12 Dec 2008 10:36:20 +0000 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 In-Reply-To: <4941A266.4000901@speakeasy.net> References: <4941A266.4000901@speakeasy.net> Message-ID: <1229078180.4041.18.camel@blaa> Hi, On Thu, 2008-12-11 at 18:29 -0500, Robert L Cochran wrote: > Hello, > > I'm an experienced IBM mainframe programmer with just a little C and > Java coding experience. I have no experience in the programming part of > virtualization. I currently do have CentOS 5.2 running as a guest under > VMWare workstation (an older version which is about a year old, running > on a Fedora 7 host. And yes, I know Fedora 7 is long out of date.) > > I would like to jump in the water and get my feet wet, though. Can > someone like me test Dom0 Xen support in Fedora 10? If so, is there a > web resource somewhere spelling out how to get started with testing? I > will need the steps involved in patching and building a test kernel > pointed out to me. > > I just bought a new laptop, the Dell Latitude E6400 with the Intel P9500 > processor. I think this processor has virtualization technology (vmx > extensions?) so the hardware for testing is there. It sounds like your interest is in virtualization generally and not Xen specifically? If that's the case - and your machine has VMX - then perhaps it would be more realistic for you to jump in (slightly shallower water :-) and try using Fedora 10 KVM? I only say that because you might find it difficult to make any headway building a Dom0 kernel. This is a good starting point: http://fedoraproject.org/wiki/Virtualization_Quick_Start Feel free to ask KVM questions on this list too ... it's about time we renamed it to fedora-virt at redhat.com :-) Cheers, Mark. From siddharth.wagh at gmail.com Fri Dec 12 15:11:23 2008 From: siddharth.wagh at gmail.com (Siddharth Wagh) Date: Fri, 12 Dec 2008 10:11:23 -0500 Subject: [Fedora-xen] unable to launch domU (FC 8) Message-ID: THERE ARE NO SPECIFIC ERROR MESSAGES, BUT THE DOMAIN IS NOT GETTING LAUNCHED EITHER. HERE IS THE GENERATED OUTPUT FILE. CAN YOU SPOT AN ERROR IN IT? ANY HELP APPRECIATED. ----------------------------------------------------------- Using config file "./vm1". Started domain domU1 Linux version 2.6.21.7-5.fc8xen (mockbuild at xenbuilder2.fedora.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Thu Aug 7 12:44:49 EDT 2008 BIOS-provided physical RAM map: sanitize start sanitize bail 0 copy_e820_map() start: 0000000000000000 size: 0000000004800000 end: 0000000004800000 type: 1 Xen: 0000000000000000 - 0000000004800000 (usable) 0MB HIGHMEM available. 72MB LOWMEM available. Using x86 segment limits to approximate NX protection Zone PFN ranges: DMA 0 -> 18432 Normal 18432 -> 18432 HighMem 18432 -> 18432 early_node_map[1] active PFN ranges 0: 0 -> 18432 ACPI in unprivileged domain disabled Detected 2784.390 MHz processor. Built 1 zonelists. Total pages: 18288 Kernel command line: root=/dev/VolGroup00/LogVol00 4 Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 CPU 0 irqstacks, hard=c136c000 soft=c134c000 PID hash table entries: 512 (order: 9, 2048 bytes) Xen reported: 2784.310 MHz processor. Console: colour dummy device 80x25 Dentry cache hash table entries: 16384 (order: 4, 65536 bytes) Inode-cache hash table entries: 8192 (order: 3, 32768 bytes) Software IO TLB disabled vmalloc area: c5000000-f4ffe000, maxmem 2d7fe000 Memory: 60372k/73728k available (2071k kernel code, 5096k reserved, 1080k data, 188k init, 0k highmem) virtual kernel memory layout: fixmap : 0xf5315000 - 0xf57fe000 (5028 kB) pkmap : 0xf5000000 - 0xf5200000 (2048 kB) vmalloc : 0xc5000000 - 0xf4ffe000 ( 767 MB) lowmem : 0xc0000000 - 0xc4800000 ( 72 MB) .init : 0xc1319000 - 0xc1348000 ( 188 kB) .data : 0xc1205e6e - 0xc1313fd4 (1080 kB) .text : 0xc1000000 - 0xc1205e6e (2071 kB) Checking if this processor honours the WP bit even in supervisor mode... Ok. Calibrating delay using timer specific routine.. 5572.51 BogoMIPS (lpj=11145020) Security Framework v1.0.0 initialized SELinux: Initializing. selinux_register_security: Registering secondary module capability Capability LSM initialized as secondary Mount-cache hash table entries: 512 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K Checking 'hlt' instruction... OK. SMP alternatives: switching to UP code Freeing SMP alternatives: 13k freed Brought up 1 CPUs NET: Registered protocol family 16 Brought up 1 CPUs PCI: Fatal: No config space access function found PCI: setting up Xen PCI frontend stub Setting up standard PCI resources ACPI: Interpreter disabled. Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI: disabled xen_mem: Initialising balloon driver. usbcore: registered new interface driver usbfs usbcore: registered new interface driver hub usbcore: registered new device driver usb PCI: System does not support PCI PCI: System does not support PCI NetLabel: Initializing NetLabel: domain hash size = 128 NetLabel: protocols = UNLABELED CIPSOv4 NetLabel: unlabeled traffic allowed by default NET: Registered protocol family 2 IP route cache hash table entries: 1024 (order: 0, 4096 bytes) TCP established hash table entries: 4096 (order: 3, 49152 bytes) TCP bind hash table entries: 4096 (order: 3, 32768 bytes) TCP: Hash tables configured (established 4096 bind 4096) TCP reno registered audit: initializing netlink socket (disabled) audit(1228851551.569:1): initialized VFS: Disk quotas dquot_6.5.1 Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) ksign: Installing public key data Loading keyring io scheduler noop registered io scheduler anticipatory registered io scheduler deadline registered io scheduler cfq registered (default) pci_hotplug: PCI Hot Plug PCI Core version: 0.5 rtc: IRQ 8 is not free. Non-volatile memory driver v1.2 Linux agpgart interface v0.102 (c) Dave Jones RAMDISK driver initialized: 16 RAM disks of 16384K size 4096 blocksize input: Macintosh mouse button emulation as /class/input/input0 Xen virtual console successfully installed as xvc0 Linux version 2.6.21.7-5.fc8xen (mockbuild at xenbuilder2.fedora.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Thu Aug 7 12:44:49 EDT 2008 BIOS-provided physical RAM map: sanitize start sanitize bail 0 copy_e820_map() start: 0000000000000000 size: 0000000004800000 end: 0000000004800000 type: 1 Xen: 0000000000000000 - 0000000004800000 (usable) 0MB HIGHMEM available. 72MB LOWMEM available. Using x86 segment limits to approximate NX protection Zone PFN ranges: DMA 0 -> 18432 Normal 18432 -> 18432 HighMem 18432 -> 18432 early_node_map[1] active PFN ranges 0: 0 -> 18432 ACPI in unprivileged domain disabled Detected 2784.390 MHz processor. Built 1 zonelists. Total pages: 18288 Kernel command line: root=/dev/VolGroup00/LogVol00 4 Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 CPU 0 irqstacks, hard=c136c000 soft=c134c000 PID hash table entries: 512 (order: 9, 2048 bytes) Xen reported: 2784.310 MHz processor. Console: colour dummy device 80x25 Dentry cache hash table entries: 16384 (order: 4, 65536 bytes) Inode-cache hash table entries: 8192 (order: 3, 32768 bytes) Software IO TLB disabled vmalloc area: c5000000-f4ffe000, maxmem 2d7fe000 Memory: 60372k/73728k available (2071k kernel code, 5096k reserved, 1080k data, 188k init, 0k highmem) virtual kernel memory layout: fixmap : 0xf5315000 - 0xf57fe000 (5028 kB) pkmap : 0xf5000000 - 0xf5200000 (2048 kB) vmalloc : 0xc5000000 - 0xf4ffe000 ( 767 MB) lowmem : 0xc0000000 - 0xc4800000 ( 72 MB) .init : 0xc1319000 - 0xc1348000 ( 188 kB) .data : 0xc1205e6e - 0xc1313fd4 (1080 kB) .text : 0xc1000000 - 0xc1205e6e (2071 kB) Checking if this processor honours the WP bit even in supervisor mode... Ok. Calibrating delay using timer specific routine.. 5572.51 BogoMIPS (lpj=11145020) Security Framework v1.0.0 initialized SELinux: Initializing. selinux_registe Thanks & Regards, Siddharth Wagh ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ If something is so complicated that you can't explain it in 10 seconds, then it's probably not worth knowing anyway. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From gf_em at hotmail.com Fri Dec 12 20:42:39 2008 From: gf_em at hotmail.com (gregg mogul) Date: Fri, 12 Dec 2008 20:42:39 +0000 Subject: [Fedora-xen] Fedora 10 x_64 Message-ID: I have installed fedora 10 and am trying to get xen working. When I try to load xend I get the following error... xend ERROR Internal error: Could not obtain handle on privileged command interface (2 = No such file or directory) Traceback (most recent call last): File "/usr/sbin/xend", line 16, in from xen.xend.server import SrvDaemon File "/usr/lib64/python2.5/site-packages/xen/xend/server/SrvDaemon.py", line 26, in import relocate File "/usr/lib64/python2.5/site-packages/xen/xend/server/relocate.py", line 28, in from xen.xend import XendDomain File "/usr/lib64/python2.5/site-packages/xen/xend/XendDomain.py", line 35, in from xen.xend import XendOptions, XendCheckpoint, XendDomainInfo File "/usr/lib64/python2.5/site-packages/xen/xend/XendCheckpoint.py", line 20, in from xen.xend import balloon, sxp, image File "/usr/lib64/python2.5/site-packages/xen/xend/image.py", line 44, in xc = xen.lowlevel.xc.xc() xen.lowlevel.xc.Error: (1, 'Internal error', 'Could not obtain handle on privileged command interface (2 = No such file or directory)') SeLinux is disabled. Is there some config file I am missing? Thanks in advance _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From aclark at commsolv.com Fri Dec 12 20:55:49 2008 From: aclark at commsolv.com (Aaron Clark) Date: Fri, 12 Dec 2008 15:55:49 -0500 Subject: [Fedora-xen] Fedora 10 x_64 In-Reply-To: References: Message-ID: <4942CFD5.6000407@commsolv.com> gregg mogul wrote: > I have installed fedora 10 and am trying to get xen working. When I try > to load xend I get the following error... > I'm not sure whether you're trying to use Fedora 10 as a dom0 or domU, but the relevant doc is: http://fedoraproject.org/wiki/Docs/Beats/Virtualization More specifically: http://fedoraproject.org/wiki/Docs/Beats/Virtualization#Xen_Updated_to_3.3.0 The short answer is pretty much use KVM instead or roll-your-own from upstream to get a dom0 that works with Fedora 10 as a domU. Aaron From bill at bfccomputing.com Fri Dec 12 22:41:36 2008 From: bill at bfccomputing.com (Bill McGonigle) Date: Fri, 12 Dec 2008 17:41:36 -0500 Subject: [Fedora-xen] unable to launch domU (FC 8) In-Reply-To: References: Message-ID: <4942E8A0.7000402@bfccomputing.com> Boy, that's not much to go on, and the console often fails to capture all output, but: On 2008-12-12 10:11 AM, Siddharth Wagh wrote: > Linux version 2.6.21.7-5.fc8xen This is the latest f8 xen kernel (at least that my f8 machine has) - have you tried booting an earlier kernel in pygrub? For some reason I don't fully understand kernel-xen updates often fail to include xenblk and xennet modules in the initrd and leave the domU unable to find its root disk. Something like: mkinitrd -v -f --with=xennet --with=xenblk initrd-`uname -r`.img `uname -r` with the correct versions will get you going again if this is the case. Also make sure the menu.lst entry isn't setup for Dom0 after the update, I've seen that too (i.e. xen kernel, linux kernel as a module). -Bill -- Bill McGonigle, Owner Work: 603.448.4440 BFC Computing, LLC Home: 603.448.1668 bill at bfccomputing.com Cell: 603.252.2606 http://www.bfccomputing.com/ Page: 603.442.1833 Blog: http://blog.bfccomputing.com/ VCard: http://bfccomputing.com/vcard/bill.vcf From siddharth.wagh at gmail.com Fri Dec 12 23:57:16 2008 From: siddharth.wagh at gmail.com (Siddharth Wagh) Date: Fri, 12 Dec 2008 18:57:16 -0500 Subject: [Fedora-xen] libvir: Remote error: Connection refused Message-ID: I am trying to run the command: su -c "virsh -c xen:/// list" and getting the error: libvir: Remote error: Connection refused libvir: warning: Failed to find the network: Is the daemon running ? libvir: Remote error: Connection refused Id Name State ------------------------------------------------------------ 0 Domain-0 running What does this mean? Thanks & Regards, Siddharth Wagh ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ If something is so complicated that you can't explain it in 10 seconds, then it's probably not worth knowing anyway. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Sat Dec 13 02:03:09 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Fri, 12 Dec 2008 21:03:09 -0500 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 In-Reply-To: <1229078180.4041.18.camel@blaa> References: <4941A266.4000901@speakeasy.net> <1229078180.4041.18.camel@blaa> Message-ID: <494317DD.1030102@speakeasy.net> Mark McLoughlin wrote: > Hi, > > On Thu, 2008-12-11 at 18:29 -0500, Robert L Cochran wrote: > >> Hello, >> >> I'm an experienced IBM mainframe programmer with just a little C and >> Java coding experience. I have no experience in the programming part of >> virtualization. I currently do have CentOS 5.2 running as a guest under >> VMWare workstation (an older version which is about a year old, running >> on a Fedora 7 host. And yes, I know Fedora 7 is long out of date.) >> >> I would like to jump in the water and get my feet wet, though. Can >> someone like me test Dom0 Xen support in Fedora 10? If so, is there a >> web resource somewhere spelling out how to get started with testing? I >> will need the steps involved in patching and building a test kernel >> pointed out to me. >> >> I just bought a new laptop, the Dell Latitude E6400 with the Intel P9500 >> processor. I think this processor has virtualization technology (vmx >> extensions?) so the hardware for testing is there. >> > > It sounds like your interest is in virtualization generally and not Xen > specifically? > > If that's the case - and your machine has VMX - then perhaps it would be > more realistic for you to jump in (slightly shallower water :-) and try > using Fedora 10 KVM? I only say that because you might find it difficult > to make any headway building a Dom0 kernel. > > This is a good starting point: > > http://fedoraproject.org/wiki/Virtualization_Quick_Start > > Feel free to ask KVM questions on this list too ... it's about time we > renamed it to fedora-virt at redhat.com :-) > > Cheers, > Mark. > > > Thanks for the suggestions! I've compiled Linux kernels several times before but still can't call myself any sort of a kernel hacker. I'm going to install Fedora 10 x86_64 on my new laptop with the virtualization packages. My goal is to install Fedora 10, CentOS 5.2, Windows XP Professional, and Windows Vista as separate virtual machines and ensure each one has access to the hardware. Video, networking, USB access, firewire access, serial port access (through my shiny new Dell E-Port Plus dock) will all be mandatory. I want to do things like bring up a Fedora 10 guest, work with the Arduino development interface (see http://www.greenbeltcomputer.biz/sanguino1.html and be patient as the numerous photos download...my web server sits right behind me and only has 128K upload), and then flash my target Arduino board over USB from the virtual machine. I do have a thousand questions to ask, such as why is a dom0 guest important? I have a hard time visualizing the practical differences between a dom0 and a domU guest. Thanks Bob From +systems.extlists.fedora-xen at dcs.qmul.ac.uk Sat Dec 13 09:30:36 2008 From: +systems.extlists.fedora-xen at dcs.qmul.ac.uk (Matt Bernstein) Date: Sat, 13 Dec 2008 09:30:36 +0000 (GMT) Subject: [Fedora-xen] rawhide domU on RHEL5.2 dom0? In-Reply-To: <1223281197.4182.22.camel@blaa> References: <1223281197.4182.22.camel@blaa> Message-ID: On Oct 6 Mark McLoughlin wrote: > On Sun, 2008-10-05 at 15:25 -0400, Jon Stanley wrote: >> I can't seem to make a rawhide domU work on a RHEL5.2 dom0, because [snip] > Cole has been backporting the patches and 5.3 should be able to be able > to install Fedora 10. Hi, slight thread-hijack here but given: - 5.3 isn't here yet - I have lots of F8 domUs (F8 is very nearly EOL) - F9 domU doesn't balloon - F10 domU doesn't boot on 5.2 ..I'd like to upgrade _only_ the necessary RPMs in 5.3beta to allow F10 to boot as a guest on my hypervisors. Does anyone have such a list? (or even a yum repository?) Somewhat hopefully, Matt From berrange at redhat.com Sat Dec 13 16:46:39 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 13 Dec 2008 16:46:39 +0000 Subject: [Fedora-xen] How To Test Dom0 Xen Support In Fedora 10 In-Reply-To: <494317DD.1030102@speakeasy.net> References: <4941A266.4000901@speakeasy.net> <1229078180.4041.18.camel@blaa> <494317DD.1030102@speakeasy.net> Message-ID: <20081213164639.GA26439@redhat.com> On Fri, Dec 12, 2008 at 09:03:09PM -0500, Robert L Cochran wrote: > > > Mark McLoughlin wrote: > > Hi, > > > > On Thu, 2008-12-11 at 18:29 -0500, Robert L Cochran wrote: > > > >> Hello, > > I do have a thousand questions to ask, such as why is a dom0 guest > important? I have a hard time visualizing the practical differences > between a dom0 and a domU guest. Dom0 is another name for the host operating system. It has access to, and manages all the physical hardware devices, and provides the mgmt of the DomU guest operating systems. The reason getting a working Dom0 is a harder bit of work than DOmU, is precisely because it has to manage all the hardware and work with the hypervisor. The DomU only ever sees virtual hardware devices (ignoring PCI passthrough), so its a much simpler beast. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From berrange at redhat.com Sat Dec 13 16:56:29 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 13 Dec 2008 16:56:29 +0000 Subject: [Fedora-xen] rawhide domU on RHEL5.2 dom0? In-Reply-To: References: <1223281197.4182.22.camel@blaa> Message-ID: <20081213165629.GC26439@redhat.com> On Sat, Dec 13, 2008 at 09:30:36AM +0000, Matt Bernstein wrote: > On Oct 6 Mark McLoughlin wrote: > > >On Sun, 2008-10-05 at 15:25 -0400, Jon Stanley wrote: > >>I can't seem to make a rawhide domU work on a RHEL5.2 dom0, because > [snip] > >Cole has been backporting the patches and 5.3 should be able to be able > >to install Fedora 10. > > Hi, slight thread-hijack here but given: > - 5.3 isn't here yet > - I have lots of F8 domUs (F8 is very nearly EOL) > - F9 domU doesn't balloon > - F10 domU doesn't boot on 5.2 > > ..I'd like to upgrade _only_ the necessary RPMs in 5.3beta to allow F10 to > boot as a guest on my hypervisors. > > Does anyone have such a list? (or even a yum repository?) Cole has rebuilt the RHEL-5.2 Xen and virt-install packages with the neccessary patches to allow Fedora 10 installs: http://crobinso.fedorapeople.org/rhel5/install_f10/ These are totally *unsupported* and not an official part of RHEL, they are just a convenience for people who want to try out F10 by can't wait until 5.3 The official supported packages will arrive in RHEL-5.3. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From cochranb at speakeasy.net Sun Dec 14 03:33:56 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sat, 13 Dec 2008 22:33:56 -0500 Subject: [Fedora-xen] KVM Networking Question Message-ID: <49447EA4.7010200@speakeasy.net> To my pleasure, I was able to get two different virtual machines running pretty quickly. I installed Fedora 10 x86_64 on my new laptop with the Virtualization package group and as little else as possible. Then I updated the system (so now I'm using kvm-74-6-fc10.x86_64), started the graphical Virtual Machine Manager, and built a guest for Windows XP Professional and a second guest for Fedora 10 x86_64. I want the Fedora 10 guest to fetch an IP address of 192.168.1.115 from the dhcp server running on my network. Virtual Machine Manager seems to be creating it's very own dhcp server and assigning network addresses of 192.168.122.2 and higher. So I can't ping the guest I just created from a host on the 192.168.1.0 network. The networking behavior VMWare gives me results in my virtual machines getting their addresses from the dhcp server on the 192.168.1.0 network, so the host IP I'm expecting for a machine is the one that is assigned and used. How do I change the networking in Virtual Machine Manager to simply pass all the dhcp calls to my network? Or is there a good reason to let the Manager do its own 192.168.122.x network? Thanks Bob Cochran From deshantm at gmail.com Sun Dec 14 03:58:12 2008 From: deshantm at gmail.com (Todd Deshane) Date: Sat, 13 Dec 2008 22:58:12 -0500 Subject: [Fedora-xen] KVM Networking Question In-Reply-To: <49447EA4.7010200@speakeasy.net> References: <49447EA4.7010200@speakeasy.net> Message-ID: <1e16a9ed0812131958g5c5fc937na4cfb9aa0f907ee1@mail.gmail.com> On Sat, Dec 13, 2008 at 10:33 PM, Robert L Cochran wrote: > To my pleasure, I was able to get two different virtual machines running > pretty quickly. I installed Fedora 10 x86_64 on my new laptop with the > Virtualization package group and as little else as possible. Then I > updated the system (so now I'm using kvm-74-6-fc10.x86_64), started the > graphical Virtual Machine Manager, and built a guest for Windows XP > Professional and a second guest for Fedora 10 x86_64. > > I want the Fedora 10 guest to fetch an IP address of 192.168.1.115 from > the dhcp server running on my network. Virtual Machine Manager seems to > be creating it's very own dhcp server and assigning network addresses of > 192.168.122.2 and higher. So I can't ping the guest I just created from > a host on the 192.168.1.0 network. > > The networking behavior VMWare gives me results in my virtual machines > getting their addresses from the dhcp server on the 192.168.1.0 network, > so the host IP I'm expecting for a machine is the one that is assigned > and used. > > How do I change the networking in Virtual Machine Manager to simply pass > all the dhcp calls to my network? Or is there a good reason to let the > Manager do its own 192.168.122.x network? > I don't know specifically about Virtual Machine Manager and don't have it handy, but for the basic concepts of KVM networking and also the libvirt style and nomenclature see the following links: http://wiki.libvirt.org/page/Networking http://kvm.qumranet.com/kvmwiki/Networking The Virtual Machine Manager page may have some information that could help you: http://virt-manager.et.redhat.com/screenshots/networking.html Hope that helps. Cheers, Todd -- Todd Deshane http://todddeshane.net http://runningxen.com From cochranb at speakeasy.net Sun Dec 14 12:42:33 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sun, 14 Dec 2008 07:42:33 -0500 Subject: [Fedora-xen] KVM Networking Question In-Reply-To: <1e16a9ed0812131958g5c5fc937na4cfb9aa0f907ee1@mail.gmail.com> References: <49447EA4.7010200@speakeasy.net> <1e16a9ed0812131958g5c5fc937na4cfb9aa0f907ee1@mail.gmail.com> Message-ID: <4944FF39.7070305@speakeasy.net> Todd Deshane wrote: > On Sat, Dec 13, 2008 at 10:33 PM, Robert L Cochran > wrote: > >> To my pleasure, I was able to get two different virtual machines running >> pretty quickly. I installed Fedora 10 x86_64 on my new laptop with the >> Virtualization package group and as little else as possible. Then I >> updated the system (so now I'm using kvm-74-6-fc10.x86_64), started the >> graphical Virtual Machine Manager, and built a guest for Windows XP >> Professional and a second guest for Fedora 10 x86_64. >> >> I want the Fedora 10 guest to fetch an IP address of 192.168.1.115 from >> the dhcp server running on my network. Virtual Machine Manager seems to >> be creating it's very own dhcp server and assigning network addresses of >> 192.168.122.2 and higher. So I can't ping the guest I just created from >> a host on the 192.168.1.0 network. >> >> The networking behavior VMWare gives me results in my virtual machines >> getting their addresses from the dhcp server on the 192.168.1.0 network, >> so the host IP I'm expecting for a machine is the one that is assigned >> and used. >> >> How do I change the networking in Virtual Machine Manager to simply pass >> all the dhcp calls to my network? Or is there a good reason to let the >> Manager do its own 192.168.122.x network? >> >> > > I don't know specifically about Virtual Machine Manager and don't have it > handy, but for the basic concepts of KVM networking and also the libvirt > style and nomenclature see the following links: > > http://wiki.libvirt.org/page/Networking > http://kvm.qumranet.com/kvmwiki/Networking > > The Virtual Machine Manager page may have some information that > could help you: > > http://virt-manager.et.redhat.com/screenshots/networking.html > > Hope that helps. > > Cheers, > Todd > > Todd, Thank you. This is very useful and helpful. It looks like I need to configure bridged networking. From the Fedora 10 release notes, Network Manager does not support virtual devices, so I'll have to disable it and roll my own. Bob From erenoglu at gmail.com Sun Dec 14 16:23:44 2008 From: erenoglu at gmail.com (Emre Erenoglu) Date: Sun, 14 Dec 2008 17:23:44 +0100 Subject: [Fedora-xen] KVM Networking Question In-Reply-To: <49447EA4.7010200@speakeasy.net> References: <49447EA4.7010200@speakeasy.net> Message-ID: hi Robert, On Sun, Dec 14, 2008 at 4:33 AM, Robert L Cochran wrote: > How do I change the networking in Virtual Machine Manager to simply pass > all the dhcp calls to my network? Or is there a good reason to let the > Manager do its own 192.168.122.x network? > I'm not using Fedora 10 but Pardus. Nevertheless, I had a similar problem with bridged networking, I couldn't get it to work with Virtual Machine Manager. Maybe the support was missing from my version. Basically, what you need to do is: 1) Create a network bridge 2) Create an ifup script 3) when starting kvm, start it with tap networking and then use the script (2) to add that tap adapter to the bridge (1). I use qtemu svn as a graphic frontend, until things get easier with Virt-Manager, with this bridged networking and virtio stuff. Emre -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Sun Dec 14 16:55:25 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sun, 14 Dec 2008 11:55:25 -0500 Subject: [Fedora-xen] KVM Networking Question In-Reply-To: <4944FF39.7070305@speakeasy.net> References: <49447EA4.7010200@speakeasy.net> <1e16a9ed0812131958g5c5fc937na4cfb9aa0f907ee1@mail.gmail.com> <4944FF39.7070305@speakeasy.net> Message-ID: <49453A7D.5060907@speakeasy.net> Robert L Cochran wrote: > Todd Deshane wrote: > >> On Sat, Dec 13, 2008 at 10:33 PM, Robert L Cochran >> wrote: >> >> >>> To my pleasure, I was able to get two different virtual machines running >>> pretty quickly. I installed Fedora 10 x86_64 on my new laptop with the >>> Virtualization package group and as little else as possible. Then I >>> updated the system (so now I'm using kvm-74-6-fc10.x86_64), started the >>> graphical Virtual Machine Manager, and built a guest for Windows XP >>> Professional and a second guest for Fedora 10 x86_64. >>> >>> I want the Fedora 10 guest to fetch an IP address of 192.168.1.115 from >>> the dhcp server running on my network. Virtual Machine Manager seems to >>> be creating it's very own dhcp server and assigning network addresses of >>> 192.168.122.2 and higher. So I can't ping the guest I just created from >>> a host on the 192.168.1.0 network. >>> >>> The networking behavior VMWare gives me results in my virtual machines >>> getting their addresses from the dhcp server on the 192.168.1.0 network, >>> so the host IP I'm expecting for a machine is the one that is assigned >>> and used. >>> >>> How do I change the networking in Virtual Machine Manager to simply pass >>> all the dhcp calls to my network? Or is there a good reason to let the >>> Manager do its own 192.168.122.x network? >>> >>> >>> >> I don't know specifically about Virtual Machine Manager and don't have it >> handy, but for the basic concepts of KVM networking and also the libvirt >> style and nomenclature see the following links: >> >> http://wiki.libvirt.org/page/Networking >> http://kvm.qumranet.com/kvmwiki/Networking >> >> The Virtual Machine Manager page may have some information that >> could help you: >> >> http://virt-manager.et.redhat.com/screenshots/networking.html >> >> Hope that helps. >> >> Cheers, >> Todd >> >> >> > Todd, > > Thank you. This is very useful and helpful. It looks like I need to > configure bridged networking. From the Fedora 10 release notes, Network > Manager does not support virtual devices, so I'll have to disable it and > roll my own. > > Bob > I have it working, thanks very much! The XML format stuff was quite helpful ( --> http://libvirt.org/formatdomain.html#elementsNICS ) Bob From +systems.extlists.fedora-xen at dcs.qmul.ac.uk Sun Dec 14 22:19:39 2008 From: +systems.extlists.fedora-xen at dcs.qmul.ac.uk (Matt Bernstein) Date: Sun, 14 Dec 2008 22:19:39 +0000 (GMT) Subject: [Fedora-xen] rawhide domU on RHEL5.2 dom0? In-Reply-To: <20081213165629.GC26439@redhat.com> References: <1223281197.4182.22.camel@blaa> <20081213165629.GC26439@redhat.com> Message-ID: On Dec 13 Daniel P. Berrange wrote: > On Sat, Dec 13, 2008 at 09:30:36AM +0000, Matt Bernstein wrote: >> ..I'd like to upgrade _only_ the necessary RPMs in 5.3beta to allow F10 to >> boot as a guest on my hypervisors. > > http://crobinso.fedorapeople.org/rhel5/install_f10/ Thank you Cole and thank you Daniel--just what the doctor ordered. Matt From m.a.young at durham.ac.uk Tue Dec 16 00:07:30 2008 From: m.a.young at durham.ac.uk (M A Young) Date: Tue, 16 Dec 2008 00:07:30 +0000 (GMT) Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 Message-ID: I have succeeded in getting a fedora based kernel to build with Dom0 patches added. The steps were roughly as follows; 1. Start from kernel-2.6.28-0.106.rc6.git4.fc11.src.rpm 2. Create a patch from http://xenbits.xen.org/paravirt_ops/patches.hg/ changeset 2238 by concatenating those patches listed in the series file (excluding those commented out) together. 3. Edit kernel.spec to revert to plain rc6 (which the pvops patches seem currently to be based on though some patches from rc8 seem to be included in the x86/x86.patch file), add the combined patch, and increase the fuzz factor to 2 so it applies. 4. Make several edits to the source files and configuration options to get it to compile on F10. Some of these, such as the couple I have so far reported on http://bugzilla.redhat.org/ (476456 and 476457), are due to code problems that show up because the F10 compiler seems to be a bit better at checking for errors. If anyone wants to inspect it, the source rpm generated is at http://compsoc.dur.ac.uk/~may/xen/kernel-2.6.28-0.106.rc6.fc10.src.rpm It is completely untested beyond the fact that it compiles for me, so I have no idea if a kernel built from it will actually boot. Michael Young From deshantm at gmail.com Tue Dec 16 00:34:47 2008 From: deshantm at gmail.com (Todd Deshane) Date: Mon, 15 Dec 2008 19:34:47 -0500 Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 In-Reply-To: References: Message-ID: <1e16a9ed0812151634jf130952tfc5f3784df7ce718@mail.gmail.com> On Mon, Dec 15, 2008 at 7:07 PM, M A Young wrote: > I have succeeded in getting a fedora based kernel to build with Dom0 patches > added. The steps were roughly as follows; > Cool. > 1. Start from kernel-2.6.28-0.106.rc6.git4.fc11.src.rpm > 2. Create a patch from http://xenbits.xen.org/paravirt_ops/patches.hg/ > changeset 2238 by concatenating those patches listed in the series file > (excluding those commented out) together. > 3. Edit kernel.spec to revert to plain rc6 (which the pvops patches seem > currently to be based on though some patches from rc8 seem to be included in > the x86/x86.patch file), add the combined patch, and increase the fuzz > factor to 2 so it applies. > 4. Make several edits to the source files and configuration options to get > it to compile on F10. Some of these, such as the couple I have so far > reported on http://bugzilla.redhat.org/ (476456 and 476457), are due to code > problems that show up because the F10 compiler seems to be a bit better at > checking for errors. > Is there any useful information that you could add to the following? http://wiki.xensource.com/xenwiki/XenParavirtOps Thanks, Todd > If anyone wants to inspect it, the source rpm generated is at > http://compsoc.dur.ac.uk/~may/xen/kernel-2.6.28-0.106.rc6.fc10.src.rpm > It is completely untested beyond the fact that it compiles for me, so I have > no idea if a kernel built from it will actually boot. > > Michael Young > > -- > Fedora-xen mailing list > Fedora-xen at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-xen > -- Todd Deshane http://todddeshane.net http://runningxen.com From pasik at iki.fi Tue Dec 16 16:54:24 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Tue, 16 Dec 2008 18:54:24 +0200 Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 In-Reply-To: References: Message-ID: <20081216165424.GU15052@edu.joroinen.fi> On Tue, Dec 16, 2008 at 12:07:30AM +0000, M A Young wrote: > I have succeeded in getting a fedora based kernel to build with Dom0 > patches added. The steps were roughly as follows; > > 1. Start from kernel-2.6.28-0.106.rc6.git4.fc11.src.rpm > 2. Create a patch from http://xenbits.xen.org/paravirt_ops/patches.hg/ > changeset 2238 by concatenating those patches listed in the series file > (excluding those commented out) together. > 3. Edit kernel.spec to revert to plain rc6 (which the pvops patches seem > currently to be based on though some patches from rc8 seem to be included > in the x86/x86.patch file), add the combined patch, and increase the fuzz > factor to 2 so it applies. > 4. Make several edits to the source files and configuration options to get > it to compile on F10. Some of these, such as the couple I have so far > reported on http://bugzilla.redhat.org/ (476456 and 476457), are due to > code problems that show up because the F10 compiler seems to be a bit > better at checking for errors. > > If anyone wants to inspect it, the source rpm generated is at > http://compsoc.dur.ac.uk/~may/xen/kernel-2.6.28-0.106.rc6.fc10.src.rpm > It is completely untested beyond the fact that it compiles for me, so I > have no idea if a kernel built from it will actually boot. > You might also need xen-unstable (3.4), or at least the tools from it. Or backport the pv_ops memory ballooning support into older xen tools version. See: http://lists.xensource.com/archives/html/xen-devel/2008-12/msg00469.html -- Pasi From bill at bfccomputing.com Tue Dec 16 19:53:24 2008 From: bill at bfccomputing.com (Bill McGonigle) Date: Tue, 16 Dec 2008 14:53:24 -0500 Subject: [Fedora-xen] backporting ZFS for pygrub Message-ID: <49480734.3080201@bfccomputing.com> Hi, all, The subject line might be a bit off, I'm looking for confirmation that I'm headed down the right path. Goal: boot a nexenta DomU (ZFS root) on CentOS 5.2 (until Fedora is ready...) I'm not really sure how the xen boot chain is put together. I find these in OpenSolaris: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6690215 http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6540109 which leads me to believe that pygrub uses libfsimage to read its stages/menu.lst and that from those above bugs I could probably extract some patches and see if they'll apply to the source in CentOS repos. Anybody care to say if that sounds correct or likely or if I'm chasing wild geese? Thanks, -Bill -- Bill McGonigle, Owner Work: 603.448.4440 BFC Computing, LLC Home: 603.448.1668 bill at bfccomputing.com Cell: 603.252.2606 http://www.bfccomputing.com/ Page: 603.442.1833 Blog: http://blog.bfccomputing.com/ VCard: http://bfccomputing.com/vcard/bill.vcf From +systems.extlists.fedora-xen at dcs.qmul.ac.uk Tue Dec 16 22:20:34 2008 From: +systems.extlists.fedora-xen at dcs.qmul.ac.uk (Matt Bernstein) Date: Tue, 16 Dec 2008 22:20:34 +0000 (GMT) Subject: [Fedora-xen] rawhide domU on RHEL5.2 dom0? In-Reply-To: References: <1223281197.4182.22.camel@blaa> <20081213165629.GC26439@redhat.com> Message-ID: On Dec 14 Matt Bernstein wrote: > On Dec 13 Daniel P. Berrange wrote: >> http://crobinso.fedorapeople.org/rhel5/install_f10/ > > Thank you Cole and thank you Daniel--just what the doctor ordered. ..and now those RPMs have vanished :( Are there newer ones around, or have they moved somewhere else? From siddharth.wagh at gmail.com Tue Dec 16 23:07:56 2008 From: siddharth.wagh at gmail.com (Siddharth Wagh) Date: Tue, 16 Dec 2008 18:07:56 -0500 Subject: [Fedora-xen] Fedora 8: Empty shared physical device? Message-ID: Hello all, I am in the process of creating a VM and am on the "Connect to host network step". Since my machine is connected via wired ethernet, I wish to choose a physical device to connect to, but am getting an empty shared physical device list. Any hints on how to populate it? Thanks & Regards, Siddharth Wagh ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ If something is so complicated that you can't explain it in 10 seconds, then it's probably not worth knowing anyway. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From siddharth.wagh at gmail.com Wed Dec 17 17:42:09 2008 From: siddharth.wagh at gmail.com (Siddharth Wagh) Date: Wed, 17 Dec 2008 12:42:09 -0500 Subject: [Fedora-xen] Fedora 8: Empty shared physical device? In-Reply-To: References: Message-ID: I havent been getting good solutions online in various archives. One says to check if peth0 network is up. In my case, it is. Am I supposed to manually add a bridge configuration file like Xenbro0? I dont see it created automatically. Any hints? Thanks & Regards, Siddharth Wagh ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ If something is so complicated that you can't explain it in 10 seconds, then it's probably not worth knowing anyway. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ---------- Forwarded message ---------- From: Siddharth Wagh Date: Tue, Dec 16, 2008 at 6:07 PM Subject: Fedora 8: Empty shared physical device? To: xen-users at lists.xensource.com, fedora-xen at redhat.com Hello all, I am in the process of creating a VM and am on the "Connect to host network step". Since my machine is connected via wired ethernet, I wish to choose a physical device to connect to, but am getting an empty shared physical device list. Any hints on how to populate it? Thanks & Regards, Siddharth Wagh ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ If something is so complicated that you can't explain it in 10 seconds, then it's probably not worth knowing anyway. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From mateenaslam at hotmail.com Wed Dec 17 18:43:52 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Wed, 17 Dec 2008 18:43:52 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: I have FC8 running as dom0 and i want to install FC10 as domU. I have installed python-virtinst-0.300.2-5.fc8 from fedora standard repo on FC8. But still i am unable to boot FC10 kernel on FC8. Following error comes when I try to boot F10 with Ramdisk created on FC8/ Dom0. # xm create -c xxx.cfg ........ Creating root device. Mounting root filesystem. mount: could not find filesystem '/dev/root' Setting up other filesystems. Setting up new root fs setuproot: moving /dev failed: No such file or directory no fstab.sys, mounting internal defaults setuproot: error mounting /proc: No such file or directory setuproot: error mounting /sys: No such file or directory Switching to new root and running init. unmounting old /dev unmounting old /proc unmounting old /sys switchroot: mount failed: No such file or directory Booting has failed. .................................................................... Here is my xen configuration file. ----------------------------------------- kernel = "/boot/vmlinuz-2.6.27.7-134.fc10.i686.PAE" ramdisk = "/boot/initrd-2.6.27.7-134.fc10.i686.PAE-DomU.img" memory = 128 name = "xxxx" vif = [ 'ip=xx.xxx.xxxx.xxx' ] disk = ['tap:aio:/var/vm13/root,sda1,w', 'tap:aio:/var/vm13/var,sda2,w', 'tap:aio:/var/vm13/swap,sda3,w'] root = "/dev/sda1 ro" on_reboot = 'restart' on_crash = 'restart' ------------------------------------------ How to boot FC-10 as domU on FC8? I will appreciate help from xen experts. Regards. Mateen _________________________________________________________________ Send e-mail anywhere. No map, no compass. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_anywhere_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From aclark at commsolv.com Wed Dec 17 18:50:48 2008 From: aclark at commsolv.com (Aaron Clark) Date: Wed, 17 Dec 2008 13:50:48 -0500 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: <49494A08.1010703@commsolv.com> MOHAMMAD MATEEN ASLAM wrote: > > I have FC8 running as dom0 and i want to install FC10 as domU. I have > installed python-virtinst-0.300.2-5.fc8 from fedora standard repo on > FC8. But still i am unable to boot FC10 kernel on FC8. > The relevant doc is: http://fedoraproject.org/wiki/Docs/Beats/Virtualization More specifically: http://fedoraproject.org/wiki/Docs/Beats/Virtualization#Xen_Updated_to_3.3.0 Fedora 8 would need pretty much all of its Xen-related packages updated to be able to handle the new F10 pv_ops kernel as a domU. I believe there was discussion about building such packages and leaving them in the testing repo for F8 (i.e. not pushing them out before EOL) but I don't remember what came of it. Search the list for more details. The short answer is pretty much use KVM instead or roll-your-own from upstream to get a dom0 that works with Fedora 10 as a domU. Aaron From mateenaslam at hotmail.com Wed Dec 17 19:49:43 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Wed, 17 Dec 2008 19:49:43 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <49494A08.1010703@commsolv.com> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <49494A08.1010703@commsolv.com> Message-ID: Hi Aaron, Thanks for quick help. Infact my current network architecture dont allow me to upgrade quickly to KVM. My FC8 is fully updated with standard update repo, but not from test repo. Is there any difference between standard repo versions and test repo versions ? My FC8 have the same python-vertinst version which was in test repo. Regards Mohammad Mateen > The relevant doc is: > http://fedoraproject.org/wiki/Docs/Beats/Virtualization > > More specifically: > http://fedoraproject.org/wiki/Docs/Beats/Virtualization#Xen_Updated_to_3.3.0 > > Fedora 8 would need pretty much all of its Xen-related packages updated > to be able to handle the new F10 pv_ops kernel as a domU. I believe > there was discussion about building such packages and leaving them in > the testing repo for F8 (i.e. not pushing them out before EOL) but I > don't remember what came of it. Search the list for more details. > > The short answer is pretty much use KVM instead or roll-your-own from > upstream to get a dom0 that works with Fedora 10 as a domU. > > Aaron _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From pasik at iki.fi Wed Dec 17 20:32:18 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Wed, 17 Dec 2008 22:32:18 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: <20081217203218.GV15052@edu.joroinen.fi> On Wed, Dec 17, 2008 at 06:43:52PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > > I have FC8 running as dom0 and i want to install FC10 as domU. I have installed python-virtinst-0.300.2-5.fc8 from fedora standard repo on FC8. But still i am unable to boot FC10 kernel on FC8. > > Following error comes when I try to boot F10 with Ramdisk created on FC8/ Dom0. > > # xm create -c xxx.cfg > > ........ > Creating root device. > Mounting root filesystem. > mount: could not find filesystem '/dev/root' > Setting up other filesystems. > Setting up new root fs > setuproot: moving /dev failed: No such file or directory > no fstab.sys, mounting internal defaults > setuproot: error mounting /proc: No such file or directory > setuproot: error mounting /sys: No such file or directory > Switching to new root and running init. > unmounting old /dev > unmounting old /proc > unmounting old /sys > switchroot: mount failed: No such file or directory > Booting has failed. > .................................................................... > > Here is my xen configuration file. > > ----------------------------------------- > kernel = "/boot/vmlinuz-2.6.27.7-134.fc10.i686.PAE" > ramdisk = "/boot/initrd-2.6.27.7-134.fc10.i686.PAE-DomU.img" > memory = 128 > name = "xxxx" > vif = [ 'ip=xx.xxx.xxxx.xxx' ] > disk = ['tap:aio:/var/vm13/root,sda1,w', 'tap:aio:/var/vm13/var,sda2,w', 'tap:aio:/var/vm13/swap,sda3,w'] > root = "/dev/sda1 ro" > on_reboot = 'restart' > on_crash = 'restart' > ------------------------------------------ > > How to boot FC-10 as domU on FC8? I will appreciate help from xen experts. > https://bugzilla.redhat.com/show_bug.cgi?id=458164 "python-virtinst-0.300.2-5.fc8 has been pushed to the Fedora 8 stable repository. If problems still persist, please make note of it in this bug report." So the python-virtinst you have should be OK. Are you sure your ramdisk/initrd is correct? based on the output you pasted it seems your initrd image is not working correctly.. Have you tried installing F10 domU with the installer? -- Pasi From mateenaslam at hotmail.com Wed Dec 17 21:45:52 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Wed, 17 Dec 2008 21:45:52 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081217203218.GV15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081217203218.GV15052@edu.joroinen.fi> Message-ID: Hi Pasi, Thanks for helping. Yes, apparently, the problem looks in Ramdisk. I have created Ramdisk with following command but it is unable to boot FC10:- mkinitrd -f --builtin=xenner --preload=xenblk --builtin=ehci-hcd --builtin=ohci-hcd --builtin=uhci-hcd --builtin=ahci --builtin=ata_piix /boot/initrd-2.6.27.7-134.fc10.i686.PAE-domU.img 2.6.27.7-134.fc10.i686.PAE if i dont mention these modules "--builtin=xenner --builtin=ehci-hcd --builtin=ohci-hcd --builtin=uhci-hcd --builtin=ahci --builtin=ata_piix" i got following errors. No module xenblk found for kernel 2.6.27.7-134.fc10.i686, aborting. No module ehci-hcd found for kernel 2.6.27.7-134.fc10.i686, aborting. No module ohci-hcd found for kernel 2.6.27.7-134.fc10.i686, aborting. ..... I have also try to use Fedora 10 Ramdisk instead of re-creating but VM halt few steps earlier giving these errors:- ................................... TCP cubic registered Initializing XFRM netlink socket NET: Registered protocol family 17 Using IPI No-Shortcut mode registered taskstats version 1 XENBUS: Device with no driver: device/vbd/2049 XENBUS: Device with no driver: device/vbd/2050 XENBUS: Device with no driver: device/vbd/2051 XENBUS: Device with no driver: device/vif/0 XENBUS: Device with no driver: device/console/0 Magic number: 1:252:3141 Freeing unused kernel memory: 424k freed Write protecting the kernel text: 2788k Write protecting the kernel read-only data: 1184k Mounting proc filesystem Mounting sysfs filesystem Creating /dev Creating initial device nodes Loading /lib/kbd/keymaps/i386/qwerty/us.map ------------------------------------------------------------ I have live updated my Old DomU (FC8) -> to New DomU (FC10). Plz let me know if there is a batter way for Ramdisk. Regards, Mateen > https://bugzilla.redhat.com/show_bug.cgi?id=458164 > > "python-virtinst-0.300.2-5.fc8 has been pushed to the Fedora 8 stable > repository. If problems still persist, please make note of it in this bug > report." > > So the python-virtinst you have should be OK. > > Are you sure your ramdisk/initrd is correct? based on the output you pasted > it seems your initrd image is not working correctly.. > > Have you tried installing F10 domU with the installer? > > -- Pasi _________________________________________________________________ Send e-mail anywhere. No map, no compass. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_anywhere_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From m.a.young at durham.ac.uk Wed Dec 17 22:09:41 2008 From: m.a.young at durham.ac.uk (M A Young) Date: Wed, 17 Dec 2008 22:09:41 +0000 (GMT) Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: On Wed, 17 Dec 2008, MOHAMMAD MATEEN ASLAM wrote: > I have FC8 running as dom0 and i want to install FC10 as domU. I have > installed python-virtinst-0.300.2-5.fc8 from fedora standard repo on FC8. > But still i am unable to boot FC10 kernel on FC8. ... > kernel = "/boot/vmlinuz-2.6.27.7-134.fc10.i686.PAE" > ramdisk = "/boot/initrd-2.6.27.7-134.fc10.i686.PAE-DomU.img" It worked for me using the vmlinuz-PAE and and initrd-PAE.img files from the DVD image (in the images/pxeboot directory). I suggest you retry with these in case there was some problem with the ramdisk you created. The python-virtinst version shouldn't matter if you are installing using xm and a configuration file, but the xen package does need to be updated to the latest fc8 version. Michael Young From cochranb at speakeasy.net Wed Dec 17 22:39:03 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Wed, 17 Dec 2008 17:39:03 -0500 Subject: [Fedora-xen] Using Wireless Connection In KVM Message-ID: <49497F87.1030302@speakeasy.net> How can I get my Fedora 10 guest to use a wireless connection? Do I need to configure a wireless "bridge" adapter? That is, if the wired adapter bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named ifcfg-wlan0 referencing "br1"? In short: does anyone have wireless working for Fedora 10, KVM guests? If yes, can the configuration details be shared with me? I'm dying to use wireless in my guest. Laptops were never meant for long trailing wires if you ask me... Thanks Bob From stephen.johnston at talktalk.net Wed Dec 17 22:46:33 2008 From: stephen.johnston at talktalk.net (Stephen Johnston) Date: Wed, 17 Dec 2008 22:46:33 +0000 Subject: [Fedora-xen] Using Wireless Connection In KVM In-Reply-To: <49497F87.1030302@speakeasy.net> References: <49497F87.1030302@speakeasy.net> Message-ID: <49498149.1080408@talktalk.net> I have a fed 10 host with wintel guests that work via a wireless connection - they are using a virtual net NAT'ted to wlan0 Robert L Cochran wrote: > How can I get my Fedora 10 guest to use a wireless connection? Do I need > to configure a wireless "bridge" adapter? That is, if the wired adapter > bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named > ifcfg-wlan0 referencing "br1"? > > In short: does anyone have wireless working for Fedora 10, KVM guests? > If yes, can the configuration details be shared with me? I'm dying to > use wireless in my guest. Laptops were never meant for long trailing > wires if you ask me... > > Thanks > > Bob > > -- > Fedora-xen mailing list > Fedora-xen at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-xen > -- Stephen Johnston 8 Lime Grove Tel. +44 (0)1506 437766 Craigshill Mob. 07786 733 150 Livingston West Lothian www.fluffyrodent.co.uk EH54 5FB From erenoglu at gmail.com Wed Dec 17 23:09:21 2008 From: erenoglu at gmail.com (Emre Erenoglu) Date: Thu, 18 Dec 2008 00:09:21 +0100 Subject: [Fedora-xen] Using Wireless Connection In KVM In-Reply-To: <49497F87.1030302@speakeasy.net> References: <49497F87.1030302@speakeasy.net> Message-ID: Hi Robert, On Wed, Dec 17, 2008 at 11:39 PM, Robert L Cochran wrote: > How can I get my Fedora 10 guest to use a wireless connection? Do I need > to configure a wireless "bridge" adapter? That is, if the wired adapter > bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named > ifcfg-wlan0 referencing "br1"? > I don't think bridging works on a wireless network card. maybe you can give a try. The NAT solution shall work though. you connect on the host to the wireless LAN, then NAT this adapter to the virtual network adapter in the guest. -- Emre -------------- next part -------------- An HTML attachment was scrubbed... URL: From pasik at iki.fi Thu Dec 18 09:18:23 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Thu, 18 Dec 2008 11:18:23 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081217203218.GV15052@edu.joroinen.fi> Message-ID: <20081218091822.GW15052@edu.joroinen.fi> On Wed, Dec 17, 2008 at 09:45:52PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > Hi Pasi, > > Thanks for helping. Yes, apparently, the problem looks in Ramdisk. I have created Ramdisk with following command but it is unable to boot FC10:- > > mkinitrd -f --builtin=xenner --preload=xenblk --builtin=ehci-hcd --builtin=ohci-hcd --builtin=uhci-hcd --builtin=ahci > --builtin=ata_piix /boot/initrd-2.6.27.7-134.fc10.i686.PAE-domU.img 2.6.27.7-134.fc10.i686.PAE > if i dont mention these modules "--builtin=xenner --builtin=ehci-hcd --builtin=ohci-hcd --builtin=uhci-hcd --builtin=ahci --builtin=ata_piix" i got following errors. > No module xenblk found for kernel 2.6.27.7-134.fc10.i686, aborting. > No module ehci-hcd found for kernel 2.6.27.7-134.fc10.i686, aborting. > No module ohci-hcd found for kernel 2.6.27.7-134.fc10.i686, aborting. > ..... > xenner is something you use on a KVM host to emulate Xen.. I think you shouldn't need/use xenner with the real Xen. And especially not in domU, xenner is for KVM "dom0". > I have also try to use Fedora 10 Ramdisk instead of re-creating but VM halt few steps earlier giving these errors:- > > ................................... > TCP cubic registered > Initializing XFRM netlink socket > NET: Registered protocol family 17 > Using IPI No-Shortcut mode > registered taskstats version 1 > XENBUS: Device with no driver: device/vbd/2049 > XENBUS: Device with no driver: device/vbd/2050 > XENBUS: Device with no driver: device/vbd/2051 This means xen block/disk device (xvd) driver was not loaded. > XENBUS: Device with no driver: device/vif/0 This means xen network driver was not loaded. Does your kernel contain those modules? or are they built-in in the pv_ops kernels? > XENBUS: Device with no driver: device/console/0 > Magic number: 1:252:3141 > Freeing unused kernel memory: 424k freed > Write protecting the kernel text: 2788k > Write protecting the kernel read-only data: 1184k > Mounting proc filesystem > Mounting sysfs filesystem > Creating /dev > Creating initial device nodes > Loading /lib/kbd/keymaps/i386/qwerty/us.map > ------------------------------------------------------------ > > > I have live updated my Old DomU (FC8) -> to New DomU (FC10). Plz let me know if there is a batter way for Ramdisk. > So.. just to get it correctly: 1. You installed Fedora 8 domU 2. You updated Fedora 8 domU to Fedora 10 (on the fly) with yum 3. You create initrd image for the F10 kernel using the mkinitrd command above That should work.. assuming you get the initrd correct. You could also try installing a fresh F10 domU (with virt-install or virt-manager) and see if that works.. -- Pasi From kraxel at redhat.com Thu Dec 18 10:19:07 2008 From: kraxel at redhat.com (Gerd Hoffmann) Date: Thu, 18 Dec 2008 11:19:07 +0100 Subject: [Fedora-xen] Using Wireless Connection In KVM In-Reply-To: <49497F87.1030302@speakeasy.net> References: <49497F87.1030302@speakeasy.net> Message-ID: <494A239B.50108@redhat.com> Robert L Cochran wrote: > How can I get my Fedora 10 guest to use a wireless connection? Do I need > to configure a wireless "bridge" adapter? That is, if the wired adapter > bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named > ifcfg-wlan0 referencing "br1"? No. wlan adapters can't handle more than one mac address, thus hooking it into the bridge to share it simply isn't going to work. > In short: does anyone have wireless working for Fedora 10, KVM guests? > If yes, can the configuration details be shared with me? I'm dying to > use wireless in my guest. Laptops were never meant for long trailing > wires if you ask me... Use NAT (aka libvirt/virt-manager's 'default' network). cheers, Gerd From berrange at redhat.com Thu Dec 18 10:36:17 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 18 Dec 2008 10:36:17 +0000 Subject: [Fedora-xen] Using Wireless Connection In KVM In-Reply-To: <49497F87.1030302@speakeasy.net> References: <49497F87.1030302@speakeasy.net> Message-ID: <20081218103616.GE23277@redhat.com> On Wed, Dec 17, 2008 at 05:39:03PM -0500, Robert L Cochran wrote: > How can I get my Fedora 10 guest to use a wireless connection? Do I need > to configure a wireless "bridge" adapter? That is, if the wired adapter > bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named > ifcfg-wlan0 referencing "br1"? > > In short: does anyone have wireless working for Fedora 10, KVM guests? > If yes, can the configuration details be shared with me? I'm dying to > use wireless in my guest. Laptops were never meant for long trailing > wires if you ask me... Wireless + bridging == fail. When creating the guest you should use libvirt's 'default' virtual network, which is NAT based and specifically designed to play nicely with NetworkManager http://wiki.libvirt.org/page/Networking So no matter what wifi network/adress your host has, the guest will still get outbound traffic through Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From muzammel.linux at gmail.com Thu Dec 18 13:09:23 2008 From: muzammel.linux at gmail.com (Muzammel Asghar) Date: Thu, 18 Dec 2008 18:09:23 +0500 Subject: [Fedora-xen] Domain-U failed to start when Domain-0 reboot Message-ID: <7756d58d0812180509g72481bdclbb36c537c9feac1f@mail.gmail.com> Hi guys i am facing some strange problem in xen. I used xen version 3 in fedora core 8 in routed mode. I create 5 vm's, the problem is when i reboot mother domain-0 first two vms boot and then immediate hang/shutdown and rest other 3 vm's are boot properly. But after that i manually start 1st and 2nd vm they start fine, i don't understand why they not start automatically when domain-0 boot but manuall start of these vm's are fine here below the logs of /var/log/messages Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. Dec 13 10:28:28 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:28 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. and in xend.log it shows this [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1040) XendDomainInfo.handleShutdownWatch [2008-12-13 10:28:27 2136] DEBUG (DevController:150) Waiting for devices vif. [2008-12-13 10:28:27 2136] DEBUG (DevController:155) Waiting for 0. [2008-12-13 10:28:27 2136] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/vif/1/0/hotplug-status. [2008-12-13 10:28:27 2136] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/vif/1/0/hotplug-status. [2008-12-13 10:28:27 2136] DEBUG (DevController:608) hotplugStatusCallback 2. [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1802) XendDomainInfo.destroy: domid=1 [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1821) XendDomainInfo.destroyDomain(1) kindly help me regarding this matter. -- Regards, Muzammel Asghar -------------- next part -------------- An HTML attachment was scrubbed... URL: From muzammel.linux at gmail.com Thu Dec 18 14:17:46 2008 From: muzammel.linux at gmail.com (Muzammel Asghar) Date: Thu, 18 Dec 2008 19:17:46 +0500 Subject: [Fedora-xen] Domain-U failed to start when Domain-0 reboot Message-ID: <7756d58d0812180617j565b912bvc341eebb9800e6b1@mail.gmail.com> Hi guys i am facing some strange problem in xen. I used xen version 3 in fedora core 8 in routed mode. I create 5 vm's, the problem is when i reboot mother domain-0 first two vms boot and shutdown and rest other 3 vm's are boot properly. But after that i manually start 1st and 2nd vm they start fine, i don't understand why they not start automatically when domain-0 boot but manuall start of these vm's are fine here below the logs of /var/log/messages Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:27 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. Dec 13 10:28:28 test logger: /etc/xen/scripts/vif-route: Dec 13 10:28:28 test logger: /etc/xen/scripts/vif-route: /etc/xen/scripts/vif-route failed; error detected. and in xend.log it shows this [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1040) XendDomainInfo.handleShutdownWatch [2008-12-13 10:28:27 2136] DEBUG (DevController:150) Waiting for devices vif. [2008-12-13 10:28:27 2136] DEBUG (DevController:155) Waiting for 0. [2008-12-13 10:28:27 2136] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/vif/1/0/hotplug-status. [2008-12-13 10:28:27 2136] DEBUG (DevController:594) hotplugStatusCallback /local/domain/0/backend/vif/1/0/hotplug-status. [2008-12-13 10:28:27 2136] DEBUG (DevController:608) hotplugStatusCallback 2. [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1802) XendDomainInfo.destroy: domid=1 [2008-12-13 10:28:27 2136] DEBUG (XendDomainInfo:1821) XendDomainInfo.destroyDomain(1) kindly help me regarding this matter. -- Regards, Muzammel Asghar -------------- next part -------------- An HTML attachment was scrubbed... URL: From mateenaslam at hotmail.com Thu Dec 18 15:27:10 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Thu, 18 Dec 2008 15:27:10 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081218091822.GW15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081217203218.GV15052@edu.joroinen.fi> <20081218091822.GW15052@edu.joroinen.fi> Message-ID: > xenner is something you use on a KVM host to emulate Xen.. I think you > shouldn't need/use xenner with the real Xen. And especially not in domU, > xenner is for KVM "dom0". True, xenner is also required for KVM. I have used xenner module successfully for my old FC8 (DomU) and it worked fine. On your advice I have tried to created RamDisk without xenner but no luck. > > XENBUS: Device with no driver: device/vbd/2049 > > XENBUS: Device with no driver: device/vbd/2050 > > XENBUS: Device with no driver: device/vbd/2051 > This means xen block/disk device (xvd) driver was not loaded. > > > XENBUS: Device with no driver: device/vif/0 > > This means xen network driver was not loaded. > > Does your kernel contain those modules? or are they built-in in the pv_ops > kernels? This seems sound, i am using Latest FC10 kernel -2.6.27.7-134.fc10.i686.PAE as DomU and Latest kernel 2.6.21.7-5.fc8xen on FC8 (Dom0). Ideally FC10 pv_ops kernel should have these modules support, either built-in or as module. > > I have live updated my Old DomU (FC8) -> to New DomU (FC10). Plz let me know if there is a batter way for Ramdisk. > > > > So.. just to get it correctly: > > 1. You installed Fedora 8 domU > 2. You updated Fedora 8 domU to Fedora 10 (on the fly) with yum > 3. You create initrd image for the F10 kernel using the mkinitrd command above > > That should work.. assuming you get the initrd correct. Yes you got exactly right, logically it should work, but its not working. > You could also try installing a fresh F10 domU (with virt-install or > virt-manager) and see if that works.. I am checking fresh installation via virt-install on another box with same setup. I will post results here. Regards, Mateen _________________________________________________________________ You live life online. So we put Windows on the web. http://clk.atdmt.com/MRT/go/127032869/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From paul at xelerance.com Thu Dec 18 15:54:55 2008 From: paul at xelerance.com (Paul Wouters) Date: Thu, 18 Dec 2008 10:54:55 -0500 (EST) Subject: [Fedora-xen] Domain-U failed to start when Domain-0 reboot In-Reply-To: <7756d58d0812180617j565b912bvc341eebb9800e6b1@mail.gmail.com> References: <7756d58d0812180617j565b912bvc341eebb9800e6b1@mail.gmail.com> Message-ID: On Thu, 18 Dec 2008, Muzammel Asghar wrote: > Hi guys i am facing some strange problem in xen. I used xen version 3 > in fedora core 8 in routed mode. I create 5 vm's, the problem is when i > reboot mother > domain-0 first two vms boot and shutdown and rest other 3 vm's > > are boot properly. But after that i manually start 1st and 2nd vm they > start fine, i don't understand why they not start automatically when > domain-0 boot but manuall start of these vm's are fine Did you create softlinks for those VM's in /etc/xen/auto/ ? Does your xendomains service startup at boot in dom0? Are there interdependancies between VM's causing them to fail to boot, eg NFS ? Paul From mateenaslam at hotmail.com Thu Dec 18 17:21:30 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Thu, 18 Dec 2008 17:21:30 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: Hi young, thanks for looking into problem. I have used vmlinuz-PAE and and initrd-PAE.img files from the DVD image (in the images/pxeboot directory). Installation starts, after selecting language, i got this +-----------------+ No driver found +------------------+ | | | Unable to find any devices of the type | | needed for this installation type. | | Would you like to manually select your | | driver or use a driver disk? | | | | +---------------+ +-------------------+ +------+ | | | Select driver | | Use a driver disk | | Back | | | +---------------+ +-------------------+ +------+ | I have tried to used both xen-blkfront, xen-netfront drivers, but installation is unable to proceed further. Do you successfully install and boot FC10 as DomU on FC8 box? > It worked for me using the vmlinuz-PAE and and initrd-PAE.img files from > the DVD image (in the images/pxeboot directory). I suggest you retry > with these in case there was some problem with the ramdisk you created. > The python-virtinst version shouldn't matter if you are installing using > xm and a configuration file, but the xen package does need to be updated > to the latest fc8 version. > > Michael Young _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From pasik at iki.fi Thu Dec 18 18:10:11 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Thu, 18 Dec 2008 20:10:11 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> Message-ID: <20081218181011.GZ15052@edu.joroinen.fi> On Thu, Dec 18, 2008 at 05:21:30PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > Hi young, > > thanks for looking into problem. I have used vmlinuz-PAE and and initrd-PAE.img files from the DVD image (in the images/pxeboot directory). Installation starts, after selecting language, i got this > > +-----------------+ No driver found +------------------+ > | | > | Unable to find any devices of the type | > | needed for this installation type. | > | Would you like to manually select your | > | driver or use a driver disk? | > | | > | +---------------+ +-------------------+ +------+ | > | | Select driver | | Use a driver disk | | Back | | > | +---------------+ +-------------------+ +------+ | > > > I have tried to used both xen-blkfront, xen-netfront drivers, but installation is unable to proceed further. > > Do you successfully install and boot FC10 as DomU on FC8 box? > Can you please post the exact commands and/or cfgfiles you used to start the installation? -- Pasi > > > It worked for me using the vmlinuz-PAE and and initrd-PAE.img files from > > the DVD image (in the images/pxeboot directory). I suggest you retry > > with these in case there was some problem with the ramdisk you created. > > The python-virtinst version shouldn't matter if you are installing using > > xm and a configuration file, but the xen package does need to be updated > > to the latest fc8 version. > > > > Michael Young > From mateenaslam at hotmail.com Thu Dec 18 19:14:50 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Thu, 18 Dec 2008 19:14:50 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081218181011.GZ15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> Message-ID: # cat vm.cfg kernel = "/boot/vmlinuz-PAE" ramdisk = "/boot/initrd-PAE.img" memory = 128 name = "xxxxx" vif = [ 'ip=xx.xx.xx.xx' ] disk = ['tap:aio:/var/xen/vm13/root,sda1,w', 'tap:aio:/var/xen/vm13/var,sda2,w', 'tap:aio:/var/xen/vm13/swap,sda3,w'] root = "/dev/sda1 ro" on_reboot = 'restart' on_crash = 'restart' # xm create -c vm.cfg -> for booting vm > Can you please post the exact commands and/or cfgfiles you used to start the > installation? > > -- Pasi > > > > > > It worked for me using the vmlinuz-PAE and and initrd-PAE.img files from > > > the DVD image (in the images/pxeboot directory). I suggest you retry > > > with these in case there was some problem with the ramdisk you created. > > > The python-virtinst version shouldn't matter if you are installing using > > > xm and a configuration file, but the xen package does need to be updated > > > to the latest fc8 version. > > > > > > Michael Young > > _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From m.a.young at durham.ac.uk Thu Dec 18 19:43:17 2008 From: m.a.young at durham.ac.uk (M A Young) Date: Thu, 18 Dec 2008 19:43:17 +0000 (GMT) Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> Message-ID: On Thu, 18 Dec 2008, MOHAMMAD MATEEN ASLAM wrote: > # cat? vm.cfg > > kernel = "/boot/vmlinuz-PAE" > ramdisk = "/boot/initrd-PAE.img" > > memory = 128 > name = "xxxxx" > vif = [ 'ip=xx.xx.xx.xx' ] > disk = ['tap:aio:/var/xen/vm13/root,sda1,w', > 'tap:aio:/var/xen/vm13/var,sda2,w', 'tap:aio:/var/xen/vm13/swap,sda3,w'] > root = "/dev/sda1 ro" > on_reboot?? = 'restart' > on_crash??? = 'restart' Should that be xvda rather than sda? I think xvda is required in some situations. I am not familar with the tap:aio: prefix either, I would have expected something like file: but as I mostly map partitions to disks or partitions it isn't an area I am expert it. Michael Young From mateenaslam at hotmail.com Thu Dec 18 21:46:21 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Thu, 18 Dec 2008 21:46:21 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> Message-ID: i think it doesn't matter what xenserver block devices is for Guest OS. You can get more details from http://searchservervirtualization.techtarget.com/tip/0,289483,sid94_gci1281401,00.html > Should that be xvda rather than sda? I think xvda is required in some > situations. I am not familar with the tap:aio: prefix either, I would have > expected something like file: but as I mostly map partitions to disks or > partitions it isn't an area I am expert it. > > Michael Young _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/Explore/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From pasik at iki.fi Thu Dec 18 21:47:01 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Thu, 18 Dec 2008 23:47:01 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> Message-ID: <20081218214701.GC15052@edu.joroinen.fi> On Thu, Dec 18, 2008 at 07:14:50PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > # cat vm.cfg > > kernel = "/boot/vmlinuz-PAE" > ramdisk = "/boot/initrd-PAE.img" > > memory = 128 > name = "xxxxx" > vif = [ 'ip=xx.xx.xx.xx' ] > disk = ['tap:aio:/var/xen/vm13/root,sda1,w', 'tap:aio:/var/xen/vm13/var,sda2,w', 'tap:aio:/var/xen/vm13/swap,sda3,w'] > root = "/dev/sda1 ro" > on_reboot = 'restart' > on_crash = 'restart' > > # xm create -c vm.cfg -> for booting vm > Your disk configuration is wrong. You need to have xvda disk for domU. Something like this: disk = ['tap:aio:/var/xen/vm13/xvda,xvda,w'] if 'tap:aio' doesn't work, then try just 'file'. -- Pasi From m.a.young at durham.ac.uk Fri Dec 19 00:25:01 2008 From: m.a.young at durham.ac.uk (M A Young) Date: Fri, 19 Dec 2008 00:25:01 +0000 (GMT) Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 In-Reply-To: <20081216165424.GU15052@edu.joroinen.fi> References: <20081216165424.GU15052@edu.joroinen.fi> Message-ID: On Tue, 16 Dec 2008, Pasi K?rkk?inen wrote: > On Tue, Dec 16, 2008 at 12:07:30AM +0000, M A Young wrote: >> If anyone wants to inspect it, the source rpm generated is at >> http://compsoc.dur.ac.uk/~may/xen/kernel-2.6.28-0.106.rc6.fc10.src.rpm >> It is completely untested beyond the fact that it compiles for me, so I >> have no idea if a kernel built from it will actually boot. >> > > You might also need xen-unstable (3.4), or at least the tools from it. > Or backport the pv_ops memory ballooning support into older xen tools version. Thanks, I have built myself an rpm starting from xen-3.3.0-1.fc10.src.rpm, with the resulting source RPM at http://compsoc.dur.ac.uk/~may/xen/xen-3.4.0-0.fc10.dev18926.0.src.rpm if anyone else is interested. Michael Young From pasik at iki.fi Fri Dec 19 11:58:52 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Fri, 19 Dec 2008 13:58:52 +0200 Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 In-Reply-To: References: <20081216165424.GU15052@edu.joroinen.fi> Message-ID: <20081219115852.GD15052@edu.joroinen.fi> On Fri, Dec 19, 2008 at 12:25:01AM +0000, M A Young wrote: > On Tue, 16 Dec 2008, Pasi K?rkk?inen wrote: > > >On Tue, Dec 16, 2008 at 12:07:30AM +0000, M A Young wrote: > >>If anyone wants to inspect it, the source rpm generated is at > >>http://compsoc.dur.ac.uk/~may/xen/kernel-2.6.28-0.106.rc6.fc10.src.rpm > >>It is completely untested beyond the fact that it compiles for me, so I > >>have no idea if a kernel built from it will actually boot. > >> > > > >You might also need xen-unstable (3.4), or at least the tools from it. > >Or backport the pv_ops memory ballooning support into older xen tools > >version. > > Thanks, I have built myself an rpm starting from xen-3.3.0-1.fc10.src.rpm, > with the resulting source RPM at > http://compsoc.dur.ac.uk/~may/xen/xen-3.4.0-0.fc10.dev18926.0.src.rpm > if anyone else is interested. > >From rawhide buildlog/changelog: xen-3.3.0-2.fc11 ---------------- * Wed Dec 17 17:00:00 2008 Gerd Hoffmann - 3.3.0-2 - build and package stub domains (pvgrub, ioemu). - backport unstable fixes for pv_ops dom0. So you could use that too. -- Pasi From pasik at iki.fi Fri Dec 19 12:08:15 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Fri, 19 Dec 2008 14:08:15 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> Message-ID: <20081219120815.GE15052@edu.joroinen.fi> On Thu, Dec 18, 2008 at 09:46:21PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > > i think it doesn't matter what xenserver block devices is for Guest OS. IIRC Fedora installer expects to find and use xvd* devices only. -- Pasi From m.a.young at durham.ac.uk Fri Dec 19 12:11:39 2008 From: m.a.young at durham.ac.uk (M A Young) Date: Fri, 19 Dec 2008 12:11:39 +0000 (GMT) Subject: [Fedora-xen] Dom0 Xen kernel on Fedora 10 In-Reply-To: <20081219115852.GD15052@edu.joroinen.fi> References: <20081216165424.GU15052@edu.joroinen.fi> <20081219115852.GD15052@edu.joroinen.fi> Message-ID: On Fri, 19 Dec 2008, Pasi K?rkk?inen wrote: > From rawhide buildlog/changelog: > > xen-3.3.0-2.fc11 > ---------------- > * Wed Dec 17 17:00:00 2008 Gerd Hoffmann - 3.3.0-2 > - build and package stub domains (pvgrub, ioemu). > - backport unstable fixes for pv_ops dom0. > > So you could use that too. Yes, I spotted that a few minutes after I had uploaded that SRPM and posted about it, and I have rebuilt that version for F10 on the machine I was going to test it on. Unfortunately the kernel I posted doesn't build for i686 PAE (it does for x86_64), so I might have to change my plans or come up with a revised kernel. Michael Young From pasik at iki.fi Fri Dec 19 12:13:44 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Fri, 19 Dec 2008 14:13:44 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081218214701.GC15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> <20081218214701.GC15052@edu.joroinen.fi> Message-ID: <20081219121344.GF15052@edu.joroinen.fi> On Thu, Dec 18, 2008 at 11:47:01PM +0200, Pasi K?rkk?inen wrote: > On Thu, Dec 18, 2008 at 07:14:50PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > > > # cat vm.cfg > > > > kernel = "/boot/vmlinuz-PAE" > > ramdisk = "/boot/initrd-PAE.img" > > > > memory = 128 > > name = "xxxxx" > > vif = [ 'ip=xx.xx.xx.xx' ] > > disk = ['tap:aio:/var/xen/vm13/root,sda1,w', 'tap:aio:/var/xen/vm13/var,sda2,w', 'tap:aio:/var/xen/vm13/swap,sda3,w'] > > root = "/dev/sda1 ro" > > on_reboot = 'restart' > > on_crash = 'restart' > > > > # xm create -c vm.cfg -> for booting vm > > > > Your disk configuration is wrong. > > You need to have xvda disk for domU. > > Something like this: > > disk = ['tap:aio:/var/xen/vm13/xvda,xvda,w'] > > if 'tap:aio' doesn't work, then try just 'file'. > Have you tried installing with virt-install? or virt-manager? Example how I installed rawhide on CentOS 5.2 dom0: (NOTE: This was earlier, before Fedora 9 was released. I haven't tried it now for F10 or F11 rawhide) $ virt-install -n test -r 512 --vcpus=1 -f /dev/vg0/test_disk1 --vnc -p -l "ftp://ftp.funet.fi/pub/mirrors/fedora.redhat.com/pub/fedora/linux/development/i386/os" Replace the FTP mirror with your local mirror. Does that work? -- Pasi From berrange at redhat.com Fri Dec 19 12:16:16 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 19 Dec 2008 12:16:16 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081219120815.GE15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> <20081219120815.GE15052@edu.joroinen.fi> Message-ID: <20081219121616.GC7284@redhat.com> On Fri, Dec 19, 2008 at 02:08:15PM +0200, Pasi K?rkk?inen wrote: > On Thu, Dec 18, 2008 at 09:46:21PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > > > > > i think it doesn't matter what xenserver block devices is for Guest OS. > > IIRC Fedora installer expects to find and use xvd* devices only. Support for hijacking SCSI/IDE device names was ripped out of the Xen blkfront when it was merged in upstream Linux. Since Fedora 10 guest is based on upstream, you need to use xvda not sda, regardless of installer. We did warn people this would likely happen when upstream got hold of the code... Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From pasik at iki.fi Fri Dec 19 12:17:51 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Fri, 19 Dec 2008 14:17:51 +0200 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081219121344.GF15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> <20081218214701.GC15052@edu.joroinen.fi> <20081219121344.GF15052@edu.joroinen.fi> Message-ID: <20081219121751.GG15052@edu.joroinen.fi> On Fri, Dec 19, 2008 at 02:13:44PM +0200, Pasi K?rkk?inen wrote: > On Thu, Dec 18, 2008 at 11:47:01PM +0200, Pasi K?rkk?inen wrote: > > On Thu, Dec 18, 2008 at 07:14:50PM +0000, MOHAMMAD MATEEN ASLAM wrote: > > > > > > # cat vm.cfg > > > > > > kernel = "/boot/vmlinuz-PAE" > > > ramdisk = "/boot/initrd-PAE.img" > > > > > > memory = 128 > > > name = "xxxxx" > > > vif = [ 'ip=xx.xx.xx.xx' ] > > > disk = ['tap:aio:/var/xen/vm13/root,sda1,w', 'tap:aio:/var/xen/vm13/var,sda2,w', 'tap:aio:/var/xen/vm13/swap,sda3,w'] > > > root = "/dev/sda1 ro" > > > on_reboot = 'restart' > > > on_crash = 'restart' > > > > > > # xm create -c vm.cfg -> for booting vm > > > > > > > Your disk configuration is wrong. > > > > You need to have xvda disk for domU. > > > > Something like this: > > > > disk = ['tap:aio:/var/xen/vm13/xvda,xvda,w'] > > > > if 'tap:aio' doesn't work, then try just 'file'. > > > > Have you tried installing with virt-install? or virt-manager? > > Example how I installed rawhide on CentOS 5.2 dom0: > (NOTE: This was earlier, before Fedora 9 was released. I haven't tried it now for F10 or F11 rawhide) > > $ virt-install -n test -r 512 --vcpus=1 -f /dev/vg0/test_disk1 --vnc -p -l "ftp://ftp.funet.fi/pub/mirrors/fedora.redhat.com/pub/fedora/linux/development/i386/os" > > Replace the FTP mirror with your local mirror. > > Does that work? > .. The reason you upgraded python-virtinst is to be able to use virt-install and/or virt-manager .. -- Pasi From mateenaslam at hotmail.com Fri Dec 19 14:47:11 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Fri, 19 Dec 2008 14:47:11 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081219120815.GE15052@edu.joroinen.fi> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> <20081219120815.GE15052@edu.joroinen.fi> Message-ID: Hi Pasi, I successfully booted FC10 kernel (DomU) on FC8 (Dom0) by changing sda with xvd*. Thanks alot for your and Young help. This is how community works. > IIRC Fedora installer expects to find and use xvd* devices only. > > -- Pasi _________________________________________________________________ Life on your PC is safer, easier, and more enjoyable with Windows Vista?. http://clk.atdmt.com/MRT/go/127032870/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From mateenaslam at hotmail.com Fri Dec 19 14:58:02 2008 From: mateenaslam at hotmail.com (MOHAMMAD MATEEN ASLAM) Date: Fri, 19 Dec 2008 14:58:02 +0000 Subject: [Fedora-xen] RE: Urgent Help Required: FC10 as DomU on FC8 as Dom0 In-Reply-To: <20081219121616.GC7284@redhat.com> References: <1136201589.1030671229535505042.JavaMail.root@zebra.lib.ucdavis.edu> <20081218181011.GZ15052@edu.joroinen.fi> <20081219120815.GE15052@edu.joroinen.fi> <20081219121616.GC7284@redhat.com> Message-ID: Thanks Daniel, my problem resolved by using xvd* instead of sda. Regards, Mateen > Support for hijacking SCSI/IDE device names was ripped out of the > Xen blkfront when it was merged in upstream Linux. Since Fedora 10 > guest is based on upstream, you need to use xvda not sda, regardless > of installer. > > We did warn people this would likely happen when upstream got hold > of the code... > > Daniel _________________________________________________________________ Send e-mail faster without improving your typing skills. http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008 -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Sat Dec 20 02:29:15 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Fri, 19 Dec 2008 21:29:15 -0500 Subject: [Fwd: Re: [Fedora-xen] Using Wireless Connection In KVM] Message-ID: <494C587B.7040002@speakeasy.net> Daniel P. Berrange wrote: > On Wed, Dec 17, 2008 at 05:39:03PM -0500, Robert L Cochran wrote: > >> How can I get my Fedora 10 guest to use a wireless connection? Do I need >> to configure a wireless "bridge" adapter? That is, if the wired adapter >> bridge is ifcfg-eth0 referencing "br0" perhaps I need a second named >> ifcfg-wlan0 referencing "br1"? >> >> In short: does anyone have wireless working for Fedora 10, KVM guests? >> If yes, can the configuration details be shared with me? I'm dying to >> use wireless in my guest. Laptops were never meant for long trailing >> wires if you ask me... >> > > Wireless + bridging == fail. > > When creating the guest you should use libvirt's 'default' virtual > network, which is NAT based and specifically designed to play nicely > with NetworkManager > > http://wiki.libvirt.org/page/Networking > > So no matter what wifi network/adress your host has, the guest will still > get outbound traffic through > > Daniel > Thanks for the advice. I want to be able to use wireless, but still get a specific IP address on my home network in the 192.168.1.x range. The virtual network doesn't use that address range by default. I guess I could kludge this...I'll have to think carefully about the networking that I'm doing. Bob From cochranb at speakeasy.net Sat Dec 20 19:23:56 2008 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sat, 20 Dec 2008 14:23:56 -0500 Subject: [Fedora-xen] Problem Having Fedora 10 Guest See USB Devices (AVC Denied Message, Too...) Message-ID: <494D464C.7070402@speakeasy.net> I have a Fedora 10 x86_64 host running a Fedora 10 x86_64 guest under KVM. I want to be able to plug a USB flash drive in and have the guest able to read and write to that device. According to the libvirt.org XML format suggestions, the way to that is with a container. Here is how I added it to the xml for my guest machine. Note that this doesn't mean I added it correctly, though: fedora10x64 33e7e731-4e18-dd90-222e-b1df83a76cad 2097152 2097152 1 hvm destroy restart restart /usr/bin/qemu-kvm When I launch Virtual Machine Manager, open the Fedora 10 guest, and click the Run button, I get an immediate AVC denied message. Dec 20 13:54:45 deafeng3 setroubleshoot: SELinux is preventing qemu (qemu-kvm) "read" to ./devices (usbfs_t). For complete SELinux messages. run sealert -l 33327e80-28c3-460a-a759-dfae737c863b Here are the `sealert` details: [root at deafeng3 qemu]# sealert -l 33327e80-28c3-460a-a759-dfae737c863b Summary: SELinux is preventing qemu (qemu-kvm) "read" to ./devices (usbfs_t). Detailed Description: SELinux denied qemu access to ./devices. If this is a virtualization image, it has to have a file context label of virt_image_t. The system is setup to label image files in directory./var/lib/libvirt/images correctly. We recommend that you copy your image file to /var/lib/libvirt/images. If you really want to have your qemu image files in the current directory, you can relabel ./devices to be virt_image_t using chcon. You also need to execute semanage fcontext -a -t virt_image_t './devices' to add this new path to the system defaults. If you did not intend to use ./devices as a qemu image it could indicate either a bug or an intrusion attempt. Allowing Access: You can alter the file context by executing chcon -t virt_image_t './devices' You must also change the default file context files on the system in order to preserve them even on a full relabel. "semanage fcontext -a -t virt_image_t './devices'" Fix Command: chcon -t virt_image_t './devices' Additional Information: Source Context system_u:system_r:qemu_t:s0 Target Context system_u:object_r:usbfs_t:s0 Target Objects ./devices [ file ] Source qemu-kvm Source Path /usr/bin/qemu-kvm Port Host deafeng3.signtype.info Source RPM Packages kvm-74-6.fc10 Target RPM Packages Policy RPM selinux-policy-3.5.13-34.fc10 Selinux Enabled True Policy Type targeted MLS Enabled True Enforcing Mode Enforcing Plugin Name qemu_file_image Host Name deafeng3.signtype.info Platform Linux deafeng3.signtype.info 2.6.27.7-134.fc10.x86_64 #1 SMP Mon Dec 1 22:21:35 EST 2008 x86_64 x86_64 Alert Count 3 First Seen Fri Dec 19 11:01:52 2008 Last Seen Sat Dec 20 13:54:45 2008 Local ID 33327e80-28c3-460a-a759-dfae737c863b Line Numbers Raw Audit Messages node=deafeng3.signtype.info type=AVC msg=audit(1229799285.706:69): avc: denied { read } for pid=4276 comm="qemu-kvm" name="devices" dev=usbfs ino=341 scontext=system_u:system_r:qemu_t:s0 tcontext=system_u:object_r:usbfs_t:s0 tclass=file node=deafeng3.signtype.info type=SYSCALL msg=audit(1229799285.706:69): arch=c000003e syscall=2 success=no exit=-13 a0=54c733 a1=0 a2=1b6 a3=7fe48d8d16f0 items=0 ppid=2903 pid=4276 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="qemu-kvm" exe="/usr/bin/qemu-kvm" subj=system_u:system_r:qemu_t:s0 key=(null) Is there a way to fix this so I can have my Fedora 10 guest read and write to USB devices? Thanks Bob Cochran From siddharth.wagh at gmail.com Mon Dec 22 21:41:08 2008 From: siddharth.wagh at gmail.com (Siddharth Wagh) Date: Mon, 22 Dec 2008 16:41:08 -0500 Subject: [Fedora-xen] VFS: Cannot open root device Message-ID: how do i know what the root partition is? i gave the option: root="/dev/hda" in the config file. Thanks & Regards, Siddharth Wagh ----------------------------------------------------------------------------------------- To ensure perfect aim, shoot first and call whatever you hit the target... ----------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill at bfccomputing.com Mon Dec 22 22:54:58 2008 From: bill at bfccomputing.com (Bill McGonigle) Date: Mon, 22 Dec 2008 17:54:58 -0500 Subject: [Fedora-xen] VFS: Cannot open root device In-Reply-To: References: Message-ID: <49501AC2.40800@bfccomputing.com> On 2008-12-22 4:41 PM, Siddharth Wagh wrote: > how do i know what the root partition is? > i gave the option: > root="/dev/hda" > in the config file. Labels are your friend here. E.g. on my machine 'borlaug' I have: kernel /boot/vmlinuz-2.6.26.6-49.fc8 ro root=LABEL=borlaug-root which I set using the e2label command from Dom0 (e2label will also report the current label). This works especially nicely when you have to switch between hd/sd and xvd devices, e.g. installing in HVM mode but running in paravirt, the device changes but the label stays the same. -Bill -- Bill McGonigle, Owner Work: 603.448.4440 BFC Computing, LLC Home: 603.448.1668 bill at bfccomputing.com Cell: 603.252.2606 http://www.bfccomputing.com/ Page: 603.442.1833 Blog: http://blog.bfccomputing.com/ VCard: http://bfccomputing.com/vcard/bill.vcf