From jason at 3dogs.us Tue Jan 1 07:17:53 2008 From: jason at 3dogs.us (Jason Hartley) Date: Tue, 1 Jan 2008 02:17:53 -0500 Subject: [et-mgmt-tools] Koan virt install hanging Message-ID: <113193470712312317q72293054yddfb9a95b29a0c64@mail.gmail.com> Hey, Happy New Years! I am having an issue with doing a koan virt install. I am trying to build a Centos5.1 Xen VM and during the initial boot it hangs at the same spot every time. I am using Cobbler version 0.6.4-2 and Koan version 0.6.3-3. Below is the output from the install during the boot where it hangs. Can some one help direct me on what might be my problem? Regards, Jason Hartley Output: [root at hydra cobbler]# koan --virt --system=sirrush --server=hydra --virt-type=xenpv - using kickstart from cobbler: http://hydra.troy.3dogs.us/cblr/kickstarts_sys/sirrush/ks.cfg libvirt_qemud (pid 3551) is running... downloading initrd initrd.img to /var/lib/xen/initrd.img url=http://hydra/cobbler/images/centos5.1-xen-i386/initrd.img - using kickstart from cobbler: http://hydra/cobbler/images/centos5.1-xen-i386/initrd.img downloading kernel vmlinuz to /var/lib/xen/vmlinuz url=http://hydra/cobbler/images/centos5.1-xen-i386/vmlinuz - using kickstart from cobbler: http://hydra/cobbler/images/centos5.1-xen-i386/vmlinuz use virt-manager or reconnect with virsh console sirrush [root at hydra cobbler]# xm console sirrush Linux version 2.6.18-53.el5xen (mockbuild at builder6.centos.org) (gcc version 4.1.2 20070626 (Red Hat 4.1.2-14)) #1 SMP Mon Nov 12 03:26:12 EST 2007 BIOS-provided physical RAM map: Xen: 0000000000000000 - 0000000010800000 (usable) 0MB HIGHMEM available. 264MB LOWMEM available. NX (Execute Disable) protection: active ACPI in unprivileged domain disabled Built 1 zonelists. Total pages: 67584 Kernel command line: ks=http://hydra.troy.3dogs.us/cblr/kickstarts_sys/sirrush/ks.cfg ksdevice=eth0 lang= kssendmac syslog=hydra.troy.3dogs.us:25150 text Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 CPU 0 irqstacks, hard=c0727000 soft=c0707000 PID hash table entries: 2048 (order: 11, 8192 bytes) Xen reported: 1808.694 MHz processor. Console: colour dummy device 80x25 Dentry cache hash table entries: 65536 (order: 6, 262144 bytes) Inode-cache hash table entries: 32768 (order: 5, 131072 bytes) Software IO TLB disabled vmalloc area: d1000000-f4ffe000, maxmem 2d7fe000 Memory: 247808k/270336k available (2049k kernel code, 14068k reserved, 844k data, 168k init, 0k highmem) Checking if this processor honours the WP bit even in supervisor mode... Ok. Calibrating delay using timer specific routine.. 4523.54 BogoMIPS (lpj=9047089) 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: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) CPU: L2 Cache: 128K (64 bytes/line) Checking 'hlt' instruction... OK. SMP alternatives: switching to UP code Freeing SMP alternatives: 13k freed Brought up 1 CPUs checking if image is initramfs... it is Freeing initrd memory: 6848k freed Grant table initialized NET: Registered protocol family 16 ACPI Exception (utmutex-0262): AE_BAD_PARAMETER, Thread C06D7AA0 could not acquire Mutex [2] [20060707] ACPI Exception (utmutex-0262): AE_BAD_PARAMETER, Thread C06D7AA0 could not acquire Mutex [2] [20060707] Brought up 1 CPUs PCI: setting up Xen PCI frontend stub 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 driver usbfs usbcore: registered new driver hub 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: 4096 (order: 2, 16384 bytes) TCP established hash table entries: 16384 (order: 5, 131072 bytes) TCP bind hash table entries: 8192 (order: 4, 65536 bytes) TCP: Hash tables configured (established 16384 bind 8192) TCP reno registered audit: initializing netlink socket (disabled) audit(1199170836.588:1): initialized VFS: Disk quotas dquot_6.5.1 Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) Initializing Cryptographic API ksign: Installing public key data Loading keyring - Added public key 74432F7696771E6E - User ID: CentOS (Kernel Module GPG key) 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 v0.101 (c) Dave Jones RAMDISK driver initialized: 16 RAM disks of 16384K size 4096 blocksize Xen virtual console successfully installed as xvc0 Event-channel device installed. Console: switching to colour frame buffer device 100x37 input: Xen Virtual Keyboard/Mouse as /class/input/input0 Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 50MHz system bus speed for PIO modes; override with idebus=xx ide-floppy driver 0.99.newide usbcore: registered new driver hiddev usbcore: registered new driver usbhid drivers/usb/input/hid-core.c: v2.6:USB HID core driver PNP: No PS/2 controller found. Probing ports directly. i8042.c: No controller found. mice: PS/2 mouse device common for all mice md: md driver 0.90.3 MAX_MD_DEVS=256, MD_SB_DISKS=27 md: bitmap version 4.39 TCP bic registered Initializing IPsec netlink socket NET: Registered protocol family 1 NET: Registered protocol family 17 Using IPI No-Shortcut mode XENBUS: Device with no driver: device/vbd/51712 XENBUS: Device with no driver: device/vif/0 Freeing unused kernel memory: 168k freed Write protecting the kernel read-only data: 360k From mdehaan at redhat.com Wed Jan 2 01:23:31 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 01 Jan 2008 20:23:31 -0500 Subject: [et-mgmt-tools] cobbler: import of x86_64 repo creates i386 distro In-Reply-To: References: <4766B491.8040600@redhat.com> Message-ID: <477AE793.8010301@redhat.com> Eugene Ventimiglia wrote: > I'm running: > > cobbler import > --path=/mnt/os-install/exported/media/Linux/RH-4.0-ES-U4-AMD64/ > --name=rhel4 > > and cobbler list reports: > > ... > distro rhel4-i386 > profile rhel4-i386 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > It sort of looks like a bug, but it isn't. Cobbler determines the "arch" bits based on the paths you import. As the standard rsync/build trees (well, RHEL really doesn't have them -- Fedora and RHEL derivatives do), typically say x86_64 or i386 in them, it uses the presense of these tokens in the part to ID the distro. If you are importing from a DVD, this shouldn't happen. Basically the names don't mean anything, so you can just do the following and will be fine: cobbler profile rename --name=rhel4-i386 --newname=rhel4-x86_64 --arch=x86 cobbler distro rename --name=rhel4-i386 --newname=rhel4-x86_64 --arch=x86 The arch field really isn't needed except for distinguishing that things are /not/ ia64 boxes, which require different baremetal boot loader setup. --Michael From mdehaan at redhat.com Wed Jan 2 01:29:29 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 01 Jan 2008 20:29:29 -0500 Subject: [et-mgmt-tools] Koan virt install hanging In-Reply-To: <113193470712312317q72293054yddfb9a95b29a0c64@mail.gmail.com> References: <113193470712312317q72293054yddfb9a95b29a0c64@mail.gmail.com> Message-ID: <477AE8F9.6090203@redhat.com> Jason Hartley wrote: > Hey, > > Happy New Years! > > I am having an issue with doing a koan virt install. I am trying to > build a Centos5.1 Xen VM and during the initial boot it hangs at the > same spot every time. I am using Cobbler version 0.6.4-2 and Koan > version 0.6.3-3. Below is the output from the install during the boot > where it hangs. Can some one help direct me on what might be my > problem? > > I think I know what is going on. It's not hanging, what is happening is that the virtualized console no longer becomes useful once the VNC graphics mode engages -- in other words, you should be able to watch the rest of the install with virt-manager. Older versions of koan had a flag to disable VNC support, but this was missing in recent versions (because I was not aware of the virt console issue). In newer versions of koan (such as the test builds... http://cobbler.et.redhat.com/download/testing) there is once again a --nogfx flag. When this flag is used, the installation will happen entirely in the virt console (and is watchable in text mode), though you will not be able to get VNC graphics at the same time. qemu/KVM (another virtualization type cobbler supports) does not support the --nogfx flag. I suspect most folks using VMs in an automated context would want to disable the VNC/graphics bits so that they could use the text consoles. Since koan in the "testing" directory does not contain any new features that require the newer cobbler, I can look at releasing this version (with the --nogfx flag) in a few weeks to Fedora/EPEL. --Michael > Regards, > Jason Hartley > > Output: > [root at hydra cobbler]# koan --virt --system=sirrush --server=hydra > --virt-type=xenpv > - using kickstart from cobbler: > http://hydra.troy.3dogs.us/cblr/kickstarts_sys/sirrush/ks.cfg > libvirt_qemud (pid 3551) is running... > downloading initrd initrd.img to /var/lib/xen/initrd.img > url=http://hydra/cobbler/images/centos5.1-xen-i386/initrd.img > - using kickstart from cobbler: > http://hydra/cobbler/images/centos5.1-xen-i386/initrd.img > downloading kernel vmlinuz to /var/lib/xen/vmlinuz > url=http://hydra/cobbler/images/centos5.1-xen-i386/vmlinuz > - using kickstart from cobbler: > http://hydra/cobbler/images/centos5.1-xen-i386/vmlinuz > use virt-manager or reconnect with virsh console sirrush > [root at hydra cobbler]# xm console sirrush > Linux version 2.6.18-53.el5xen (mockbuild at builder6.centos.org) (gcc > version 4.1.2 20070626 (Red Hat 4.1.2-14)) #1 SMP Mon Nov 12 03:26:12 > EST 2007 > BIOS-provided physical RAM map: > Xen: 0000000000000000 - 0000000010800000 (usable) > 0MB HIGHMEM available. > 264MB LOWMEM available. > NX (Execute Disable) protection: active > ACPI in unprivileged domain disabled > Built 1 zonelists. Total pages: 67584 > Kernel command line: > ks=http://hydra.troy.3dogs.us/cblr/kickstarts_sys/sirrush/ks.cfg > ksdevice=eth0 lang= kssendmac syslog=hydra.troy.3dogs.us:25150 text > Enabling fast FPU save and restore... done. > Enabling unmasked SIMD FPU exception support... done. > Initializing CPU#0 > CPU 0 irqstacks, hard=c0727000 soft=c0707000 > PID hash table entries: 2048 (order: 11, 8192 bytes) > Xen reported: 1808.694 MHz processor. > Console: colour dummy device 80x25 > Dentry cache hash table entries: 65536 (order: 6, 262144 bytes) > Inode-cache hash table entries: 32768 (order: 5, 131072 bytes) > Software IO TLB disabled > vmalloc area: d1000000-f4ffe000, maxmem 2d7fe000 > Memory: 247808k/270336k available (2049k kernel code, 14068k reserved, > 844k data, 168k init, 0k highmem) > Checking if this processor honours the WP bit even in supervisor mode... Ok. > Calibrating delay using timer specific routine.. 4523.54 BogoMIPS (lpj=9047089) > 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: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) > CPU: L2 Cache: 128K (64 bytes/line) > Checking 'hlt' instruction... OK. > SMP alternatives: switching to UP code > Freeing SMP alternatives: 13k freed > Brought up 1 CPUs > checking if image is initramfs... it is > Freeing initrd memory: 6848k freed > Grant table initialized > NET: Registered protocol family 16 > ACPI Exception (utmutex-0262): AE_BAD_PARAMETER, Thread C06D7AA0 could > not acquire Mutex [2] [20060707] > ACPI Exception (utmutex-0262): AE_BAD_PARAMETER, Thread C06D7AA0 could > not acquire Mutex [2] [20060707] > Brought up 1 CPUs > PCI: setting up Xen PCI frontend stub > 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 driver usbfs > usbcore: registered new driver hub > 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: 4096 (order: 2, 16384 bytes) > TCP established hash table entries: 16384 (order: 5, 131072 bytes) > TCP bind hash table entries: 8192 (order: 4, 65536 bytes) > TCP: Hash tables configured (established 16384 bind 8192) > TCP reno registered > audit: initializing netlink socket (disabled) > audit(1199170836.588:1): initialized > VFS: Disk quotas dquot_6.5.1 > Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) > Initializing Cryptographic API > ksign: Installing public key data > Loading keyring > - Added public key 74432F7696771E6E > - User ID: CentOS (Kernel Module GPG key) > 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 v0.101 (c) Dave Jones > RAMDISK driver initialized: 16 RAM disks of 16384K size 4096 blocksize > Xen virtual console successfully installed as xvc0 > Event-channel device installed. > Console: switching to colour frame buffer device 100x37 > input: Xen Virtual Keyboard/Mouse as /class/input/input0 > Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 > ide: Assuming 50MHz system bus speed for PIO modes; override with idebus=xx > ide-floppy driver 0.99.newide > usbcore: registered new driver hiddev > usbcore: registered new driver usbhid > drivers/usb/input/hid-core.c: v2.6:USB HID core driver > PNP: No PS/2 controller found. Probing ports directly. > i8042.c: No controller found. > mice: PS/2 mouse device common for all mice > md: md driver 0.90.3 MAX_MD_DEVS=256, MD_SB_DISKS=27 > md: bitmap version 4.39 > TCP bic registered > Initializing IPsec netlink socket > NET: Registered protocol family 1 > NET: Registered protocol family 17 > Using IPI No-Shortcut mode > XENBUS: Device with no driver: device/vbd/51712 > XENBUS: Device with no driver: device/vif/0 > Freeing unused kernel memory: 168k freed > Write protecting the kernel read-only data: 360k > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From jason at 3dogs.us Wed Jan 2 11:17:04 2008 From: jason at 3dogs.us (Jason Hartley) Date: Wed, 2 Jan 2008 06:17:04 -0500 Subject: [et-mgmt-tools] Koan virt install hanging In-Reply-To: <477AE8F9.6090203@redhat.com> References: <113193470712312317q72293054yddfb9a95b29a0c64@mail.gmail.com> <477AE8F9.6090203@redhat.com> Message-ID: <113193470801020317o6cbbd463p335a1390a96c9c8b@mail.gmail.com> On 1/1/08, Michael DeHaan wrote: > Jason Hartley wrote: > > Hey, > > > > Happy New Years! > > > > I am having an issue with doing a koan virt install. I am trying to > > build a Centos5.1 Xen VM and during the initial boot it hangs at the > > same spot every time. I am using Cobbler version 0.6.4-2 and Koan > > version 0.6.3-3. Below is the output from the install during the boot > > where it hangs. Can some one help direct me on what might be my > > problem? > > > > > I think I know what is going on. > > It's not hanging, what is happening is that the virtualized console no > longer becomes useful once the VNC graphics mode engages -- in other > words, you should be able to watch the rest of the install with > virt-manager. > > Older versions of koan had a flag to disable VNC support, but this was > missing in recent versions (because I was not aware of the virt console > issue). In newer versions of koan (such as the test builds... > http://cobbler.et.redhat.com/download/testing) there is once again a > --nogfx flag. > > When this flag is used, the installation will happen entirely in the > virt console (and is watchable in text mode), though you will not be > able to get VNC graphics at the same time. > > qemu/KVM (another virtualization type cobbler supports) does not support > the --nogfx flag. > > I suspect most folks using VMs in an automated context would want to > disable the VNC/graphics bits so that they could use the text consoles. > > Since koan in the "testing" directory does not contain any new features > that require the newer cobbler, I can look at releasing this version > (with the --nogfx flag) in a few weeks to Fedora/EPEL. > > --Michael > Thanks, for the quick response. Your explanation is spot-on. I played around with virt-install, turning the graphics support on and off, and I got the same behavior with the graphics on. Thanks, Jason Hartley From dpaulo at fedoraproject.org Wed Jan 2 11:24:13 2008 From: dpaulo at fedoraproject.org (Davidson Rodrigues Paulo) Date: Wed, 2 Jan 2008 09:24:13 -0200 Subject: [et-mgmt-tools] What component use to bug report? Message-ID: Hi, There is a bug I want to report on Bugzilla but I don't know for what component I assign the ticket. The problem is with the / (slash) key which don't works when using pt-br VNC keymap to view Qemu and Xen (and probably KVM too) guests. There is no problem in Xen console windows. Thanks, Davidson Paulo http://fedoraproject.org/wiki/DavidsonPaulo From berrange at redhat.com Wed Jan 2 12:22:59 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 2 Jan 2008 12:22:59 +0000 Subject: [et-mgmt-tools] What component use to bug report? In-Reply-To: References: Message-ID: <20080102122259.GB12101@redhat.com> On Wed, Jan 02, 2008 at 09:24:13AM -0200, Davidson Rodrigues Paulo wrote: > Hi, > > There is a bug I want to report on Bugzilla but I don't know for what > component I assign the ticket. The problem is with the / (slash) key > which don't works when using pt-br VNC keymap to view Qemu and Xen > (and probably KVM too) guests. There is no problem in Xen console > windows. If you're installing with virt-manager then use that as the component, otherwise use 'qemu' or 'kvm'. If need be, the maintainer can easily re-assign the bug to a different component later, so don't worry too much. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Wed Jan 2 16:40:25 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 2 Jan 2008 16:40:25 +0000 Subject: [et-mgmt-tools] [PATCH] BZ#251643: Allow to set the cpu pinning for creating domain with virt-manager. In-Reply-To: <200712251537.BCE52130.0HOK49G8@aa.jp.fujitsu.com> References: <200712251537.BCE52130.0HOK49G8@aa.jp.fujitsu.com> Message-ID: <20080102164025.GD1237@redhat.com> On Tue, Dec 25, 2007 at 03:37:58PM +0900, Nobuhiro Itou wrote: > Hi > > As well as virt-install, I want to set the cpu pinning > at the time of creating domain with virt-manager. > And, because virtinst changeset 320 was committed, > I add the allocation setting of pCPU to virt-manager. > > some notes are follows: > - This feature works with changeset of virtinst is 320 or later. > - The specification about the CPU pin parameter follows that of virt-install. > - This function can be used only for xen hypervisor. > - When 1 pCPU in xen hypervisor, this function is disabled. I appreciate your work on this patch, however, IMHO this is not something we should add to the new VM wizard. The need to pin CPUs at install time has a tiny niche scenario of usefullness. For most people pinning CPUs is a bad idea as they will make worse decisions that the hypervisor/OS schedular. The more features we add to the 'new VM wizard', the more complex the user interaction becomes. Our goal for virt-manager is to provide UI which targets the most commonly used features only and explicitly leave out features only used in a niche use cases, thus keeping the app simple to use. IMHO, pinning CPUs at install time is just not a feature which is critical enough to justify in the new VM wizard. I *would* however welcome a patch to the 'Virtual Machine Details' dialog where you can currently adjust the number of vCPUs. It would make sense to be able to have UI for pinning vCPUs of an existing domain in this part of the UI. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Wed Jan 2 19:04:23 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 02 Jan 2008 14:04:23 -0500 Subject: [et-mgmt-tools] Reminder: FUDCon (the Fedora conference) is January 11-13 in Raleigh, and FREE Message-ID: <477BE037.3060206@redhat.com> Details here: http://barcamp.org/FUDConRaleigh2008 Everyone is welcome. The actual conference talks are going to be on Saturday, with hack sessions Friday and Sunday. --Michael From rjones at redhat.com Wed Jan 2 21:00:48 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 02 Jan 2008 21:00:48 +0000 Subject: [et-mgmt-tools] [PATCH] BZ#251643: Allow to set the cpu pinning for creating domain with virt-manager. In-Reply-To: <200712251537.BCE52130.0HOK49G8@aa.jp.fujitsu.com> References: <200712251537.BCE52130.0HOK49G8@aa.jp.fujitsu.com> Message-ID: <477BFB80.70600@redhat.com> Nobuhiro Itou wrote: > Hi > > As well as virt-install, I want to set the cpu pinning > at the time of creating domain with virt-manager. > And, because virtinst changeset 320 was committed, > I add the allocation setting of pCPU to virt-manager. > > some notes are follows: > - This feature works with changeset of virtinst is 320 or later. > - The specification about the CPU pin parameter follows that of virt-install. > - This function can be used only for xen hypervisor. > - When 1 pCPU in xen hypervisor, this function is disabled. Nobuhiro, Did you have any comments on my proposal here? https://www.redhat.com/archives/libvir-list/2007-December/msg00327.html Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From berrange at redhat.com Wed Jan 2 22:57:01 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 2 Jan 2008 22:57:01 +0000 Subject: [et-mgmt-tools] FYI: virt-manager re-instate new & delete buttons Message-ID: <20080102225701.GJ1237@redhat.com> After the mild disaster of including the new & delete buttons inline in the list of VMs / connections, I've reverted the UI back to the old style with new & delete buttons at the bottom of the window. Anyone interested in checking this out the changes are now in the virt-manager HG repo and I will do an update release this week with them in... Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From hbrock at redhat.com Thu Jan 3 14:02:57 2008 From: hbrock at redhat.com (Hugh O. Brock) Date: Thu, 3 Jan 2008 09:02:57 -0500 Subject: [et-mgmt-tools] FYI: virt-manager re-instate new & delete buttons In-Reply-To: <20080102225701.GJ1237@redhat.com> References: <20080102225701.GJ1237@redhat.com> Message-ID: <20080103140253.GG9820@redhat.com> On Wed, Jan 02, 2008 at 10:57:01PM +0000, Daniel P. Berrange wrote: > After the mild disaster of including the new & delete buttons inline in the > list of VMs / connections, I've reverted the UI back to the old style with > new & delete buttons at the bottom of the window. Anyone interested in > checking this out the changes are now in the virt-manager HG repo and I > will do an update release this week with them in... > > Dan. I know Cole Robinson was working on changing the in-line buttons to make them more usable (that is, actually get buttons to show up in-line rather than the less-than-obvious icons); did you check with him at all before committing this? --H From hyclak at math.ohiou.edu Thu Jan 3 15:47:28 2008 From: hyclak at math.ohiou.edu (Matt Hyclak) Date: Thu, 3 Jan 2008 10:47:28 -0500 Subject: [et-mgmt-tools] Fix for Cobbler import running triggers Message-ID: <20080103154728.GD10057@math.ohiou.edu> This is a fix to the patch yesterday to run triggers on cobbler import (changeset f22f6b18149029e073c3db3ff812bb4c06e24036) Matt -- Matt Hyclak Department of Mathematics Department of Social Work Ohio University (740) 593-1263 -------------- next part -------------- diff --git a/cobbler/action_import.py b/cobbler/action_import.py index 952c663..ca78dc1 100644 --- a/cobbler/action_import.py +++ b/cobbler/action_import.py @@ -195,7 +195,7 @@ class Importer: profile.set_kickstart(kickstart_file) self.configure_tree_location(distro) - self.distros.add(distro,save=True) # re-save + self.distros.add(distro,with_copy=True) # re-save self.api.serialize() # -------------------------------------------------------------------- @@ -480,7 +480,7 @@ class Importer: distro.set_initrd(initrd) distro.set_arch(pxe_arch) distro.source_repos = [] - self.distros.add(distro,save=True) + self.distros.add(distro,with_copy=True) self.distros_added.append(distro) existing_profile = self.profiles.find(name=name) @@ -495,7 +495,7 @@ class Importer: profile.set_name(name) profile.set_distro(name) - self.profiles.add(profile,save=True) + self.profiles.add(profile,with_copy=True) self.api.serialize() return distro -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From crobinso at redhat.com Thu Jan 3 15:58:39 2008 From: crobinso at redhat.com (Cole Robinson) Date: Thu, 03 Jan 2008 10:58:39 -0500 Subject: [et-mgmt-tools] FYI: virt-manager re-instate new & delete buttons In-Reply-To: <20080103140253.GG9820@redhat.com> References: <20080102225701.GJ1237@redhat.com> <20080103140253.GG9820@redhat.com> Message-ID: <477D062F.6080401@redhat.com> Hugh O. Brock wrote: > On Wed, Jan 02, 2008 at 10:57:01PM +0000, Daniel P. Berrange wrote: >> After the mild disaster of including the new & delete buttons inline in the >> list of VMs / connections, I've reverted the UI back to the old style with >> new & delete buttons at the bottom of the window. Anyone interested in >> checking this out the changes are now in the virt-manager HG repo and I >> will do an update release this week with them in... >> >> Dan. > > I know Cole Robinson was working on changing the in-line buttons to > make them more usable (that is, actually get buttons to show up > in-line rather than the less-than-obvious icons); did you check with > him at all before committing this? > > --H > I investigated this but haven't made any changes yet, so I don't have anything ready. I can start working on it if salvaging the inline buttons is the way we want to go. Also before we do a release we should make sure the restore button is hooked up :) I hadn't touched it since I thought we were waiting for the storage api to land before we pushed a new release. - Cole From berrange at redhat.com Thu Jan 3 16:26:45 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 3 Jan 2008 16:26:45 +0000 Subject: [et-mgmt-tools] FYI: virt-manager re-instate new & delete buttons In-Reply-To: <477D062F.6080401@redhat.com> References: <20080102225701.GJ1237@redhat.com> <20080103140253.GG9820@redhat.com> <477D062F.6080401@redhat.com> Message-ID: <20080103162645.GC21949@redhat.com> On Thu, Jan 03, 2008 at 10:58:39AM -0500, Cole Robinson wrote: > Hugh O. Brock wrote: > > On Wed, Jan 02, 2008 at 10:57:01PM +0000, Daniel P. Berrange wrote: > >> After the mild disaster of including the new & delete buttons inline in the > >> list of VMs / connections, I've reverted the UI back to the old style with > >> new & delete buttons at the bottom of the window. Anyone interested in > >> checking this out the changes are now in the virt-manager HG repo and I > >> will do an update release this week with them in... > >> > >> Dan. > > > > I know Cole Robinson was working on changing the in-line buttons to > > make them more usable (that is, actually get buttons to show up > > in-line rather than the less-than-obvious icons); did you check with > > him at all before committing this? > I investigated this but haven't made any changes yet, so I don't have anything > ready. I can start working on it if salvaging the inline buttons is the way we > want to go. I don't think we should worry about inline buttons - if we have full button themeing inline I think it'll look rather ugly. Surprisingly to me, hooking up the buttons along the bottom worked fairly well and simplified the code too. > Also before we do a release we should make sure the restore button is hooked > up :) I hadn't touched it since I thought we were waiting for the storage api > to land before we pushed a new release. No, we shouldn't wait for the storage API. We should fix the save/restore menus so that they work for local HV connections, and simply inform the user that its not supported for remote connections yet. When we get the storage API done, we can then hook up save/restore for remote too. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From hyclak at math.ohiou.edu Thu Jan 3 16:27:47 2008 From: hyclak at math.ohiou.edu (Matt Hyclak) Date: Thu, 3 Jan 2008 11:27:47 -0500 Subject: [et-mgmt-tools] Fix for Cobbler import running triggers In-Reply-To: <20080103154728.GD10057@math.ohiou.edu> References: <20080103154728.GD10057@math.ohiou.edu> Message-ID: <20080103162747.GE10057@math.ohiou.edu> On Thu, Jan 03, 2008 at 10:47:28AM -0500, Matt Hyclak enlightened us: > This is a fix to the patch yesterday to run triggers on cobbler import > (changeset f22f6b18149029e073c3db3ff812bb4c06e24036) > > Matt > diff --git a/cobbler/action_import.py b/cobbler/action_import.py > index 952c663..ca78dc1 100644 > --- a/cobbler/action_import.py > +++ b/cobbler/action_import.py > @@ -195,7 +195,7 @@ class Importer: > profile.set_kickstart(kickstart_file) > > self.configure_tree_location(distro) > - self.distros.add(distro,save=True) # re-save > + self.distros.add(distro,with_copy=True) # re-save > self.api.serialize() > > # -------------------------------------------------------------------- > @@ -480,7 +480,7 @@ class Importer: > distro.set_initrd(initrd) > distro.set_arch(pxe_arch) > distro.source_repos = [] > - self.distros.add(distro,save=True) > + self.distros.add(distro,with_copy=True) > self.distros_added.append(distro) > > existing_profile = self.profiles.find(name=name) > @@ -495,7 +495,7 @@ class Importer: > profile.set_name(name) > profile.set_distro(name) > > - self.profiles.add(profile,save=True) > + self.profiles.add(profile,with_copy=True) > self.api.serialize() > > return distro I should clarify - this fixes the patch for running on the 0.6.x branch of cobbler, not needed for the devel branch. Matt -- Matt Hyclak Department of Mathematics Department of Social Work Ohio University (740) 593-1263 From eventi at gridapp.com Thu Jan 3 21:55:26 2008 From: eventi at gridapp.com (Eugene Ventimiglia) Date: Thu, 3 Jan 2008 16:55:26 -0500 Subject: [et-mgmt-tools] cobbler: import of x86_64 repo creates i386 distro In-Reply-To: <477AE793.8010301@redhat.com> References: <4766B491.8040600@redhat.com> <477AE793.8010301@redhat.com> Message-ID: Maybe I shouldn't be doing import at all, or doing a lot more configuration; I tried what you said, and then ran import with the RH4-i386 source and one distro over-wrote the other. I need 2 seperate "distros" one for i386 and one for x86_64 How can I do this? -----Original Message----- From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Michael DeHaan Sent: Tuesday, January 01, 2008 8:24 PM To: Fedora/Linux Management Tools Subject: Re: [et-mgmt-tools] cobbler: import of x86_64 repo creates i386 distro Eugene Ventimiglia wrote: > I'm running: > > cobbler import > --path=/mnt/os-install/exported/media/Linux/RH-4.0-ES-U4-AMD64/ > --name=rhel4 > > and cobbler list reports: > > ... > distro rhel4-i386 > profile rhel4-i386 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > It sort of looks like a bug, but it isn't. Cobbler determines the "arch" bits based on the paths you import. As the standard rsync/build trees (well, RHEL really doesn't have them -- Fedora and RHEL derivatives do), typically say x86_64 or i386 in them, it uses the presense of these tokens in the part to ID the distro. If you are importing from a DVD, this shouldn't happen. Basically the names don't mean anything, so you can just do the following and will be fine: cobbler profile rename --name=rhel4-i386 --newname=rhel4-x86_64 --arch=x86 cobbler distro rename --name=rhel4-i386 --newname=rhel4-x86_64 --arch=x86 The arch field really isn't needed except for distinguishing that things are /not/ ia64 boxes, which require different baremetal boot loader setup. --Michael _______________________________________________ et-mgmt-tools mailing list et-mgmt-tools at redhat.com https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Thu Jan 3 23:16:19 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 03 Jan 2008 18:16:19 -0500 Subject: [et-mgmt-tools] cobbler: import of x86_64 repo creates i386 distro In-Reply-To: References: <4766B491.8040600@redhat.com> <477AE793.8010301@redhat.com> Message-ID: <477D6CC3.7020807@redhat.com> Eugene Ventimiglia wrote: > Maybe I shouldn't be doing import at all, or doing a lot more > configuration; I tried what you said, and then ran import with the > RH4-i386 source and one distro over-wrote the other. > Yeah, it's probably because your local copy has some things named differently than Cobbler would normally expect. It expects to see x86/ or x86_64 actually in a path segment. The names will likely be weird, but if you do the following, you can import both arches in one swoop and then rename the one that gets set to the wrong arch. cobbler import --name=RHEL4 --path=/mnt/os-install/exported/media/Linux That assumes they are both under the same parent. --Michael From mdehaan at redhat.com Thu Jan 3 23:17:05 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 03 Jan 2008 18:17:05 -0500 Subject: [et-mgmt-tools] cobbler: import of x86_64 repo creates i386 distro In-Reply-To: References: <4766B491.8040600@redhat.com> <477AE793.8010301@redhat.com> Message-ID: <477D6CF1.309@redhat.com> Eugene Ventimiglia wrote: > Maybe I shouldn't be doing import at all, or doing a lot more > configuration; I tried what you said, and then ran import with the > RH4-i386 source and one distro over-wrote the other. > > I need 2 seperate "distros" one for i386 and one for x86_64 > > How can I do this? > Quick add: if you have the DVD media around, importing from a mounted DVD works great too. From thestrider at gmail.com Sat Jan 5 05:34:19 2008 From: thestrider at gmail.com (Adam Rosenwald) Date: Sat, 05 Jan 2008 00:34:19 -0500 Subject: [et-mgmt-tools] PATCH: Graphical options for koan Message-ID: <477F16DB.9030402@gmail.com> Michael, I've noticed that the default graphical console display has changed in koan-0.6.3-3 (I had previously been using v0.6.2-2). The default display is now set to "vnc". The default console setting for python-virtinst is None. This is a good choice for a default, since /everyone/ (even those not using X every time they deploy virtuals) can view the console. The following patches push console policy to the user and defaults the console to text in the event a "-g --graphics" switch is not found. ===BEGIN app.py PATCH=== --- app.py.orig 2008-01-05 05:12:15.000000000 +0000 +++ app.py 2008-01-05 05:07:45.000000000 +0000 @@ -148,6 +148,10 @@ p.add_option("-T", "--virt-type", dest="virt_type", help="virtualization install type (xenpv,qemu)") + p.add_option("-g", "--graphics", + dest="is_graphics", + action="store_true", + help="use vnc console instead of text console") (options, args) = p.parse_args() @@ -168,6 +172,7 @@ k.live_cd = options.live_cd k.virt_path = options.virt_path k.virt_type = options.virt_type + k.is_graphics = options.is_graphics if options.virt_name is not None: k.virt_name = options.virt_name @@ -222,6 +227,7 @@ self.virt_name = None self.virt_type = None self.virt_path = None + self.is_graphics = None #--------------------------------------------------- @@ -882,24 +888,26 @@ pd = profile_data self.load_virt_modules() - arch = self.safe_load(pd,'arch','x86') - kextra = self.calc_kernel_args(pd) - (uuid, create_func) = self.virt_choose(pd) - - virtname = self.calc_virt_name(pd) - - ram = self.calc_virt_ram(pd) - - vcpus = self.calc_virt_cpus(pd) - path_list = self.calc_virt_path(pd, virtname) - size_list = self.calc_virt_filesize(pd) - disks = self.merge_disk_data(path_list,size_list) + arch = self.safe_load(pd,'arch','x86') + kextra = self.calc_kernel_args(pd) + (uuid, create_func) = self.virt_choose(pd) + (graphics, create_func) = self.virt_choose(pd) + + virtname = self.calc_virt_name(pd) + + ram = self.calc_virt_ram(pd) + + vcpus = self.calc_virt_cpus(pd) + path_list = self.calc_virt_path(pd, virtname) + size_list = self.calc_virt_filesize(pd) + disks = self.merge_disk_data(path_list,size_list) results = create_func( name = virtname, ram = ram, disks = disks, uuid = uuid, + graphics = graphics, extra = kextra, vcpus = vcpus, profile_data = profile_data, @@ -923,13 +931,14 @@ def virt_choose(self, pd): if self.virt_type == "xenpv": - uuid = self.get_uuid(self.calc_virt_uuid(pd)) + graphics = self.is_graphics + uuid = self.get_uuid(self.calc_virt_uuid(pd)) import xencreate - creator = xencreate.start_paravirt_install + creator = xencreate.start_paravirt_install elif self.virt_type == "qemu": - uuid = None + uuid = None import qcreate - creator = qcreate.start_install + creator = qcreate.start_install else: raise InfoException, "Unspecified virt type: %s" % self.virt_type return (uuid, creator) ===END app.py PATCH=== ===BEGIN xencreate.py PATCH=== --- xencreate.py.orig 2008-01-05 05:12:30.000000000 +0000 +++ xencreate.py 2008-01-05 05:03:20.000000000 +0000 @@ -47,6 +47,7 @@ uuid=None, extra=None, vcpus=None, + graphics=None, profile_data=None, arch=None): @@ -56,7 +57,8 @@ guest.set_name(name) guest.set_memory(ram) guest.set_vcpus(vcpus) -# guest.set_graphics("vnc") + if graphics: + guest.set_graphics("vnc") if uuid is not None: guest.set_uuid(uuid) ===END xencreate.py PATCH=== Regards, - Adam. From thestrider at gmail.com Sat Jan 5 05:53:16 2008 From: thestrider at gmail.com (Adam Rosenwald) Date: Sat, 05 Jan 2008 00:53:16 -0500 Subject: [et-mgmt-tools] Re: PATCH: Graphical options for koan In-Reply-To: <477F16DB.9030402@gmail.com> References: <477F16DB.9030402@gmail.com> Message-ID: <477F1B4C.2060503@gmail.com> Sorry, Michael. I didn't see your thread on --nogfx. My patches were applicable to koan-0.6.3-3 - not GIT (which likely contained your new code). In any event, I would highly recommend that you revert the default console policy to text, since this will work for practically every end-user environment. VNC requires dependencies that not all users will have or /should/ have in order to use koan. Regards, - A. Adam Rosenwald wrote: > Michael, > > I've noticed that the default graphical console display has changed in > koan-0.6.3-3 (I had previously been using v0.6.2-2). The default > display is now set to "vnc". > > The default console setting for python-virtinst is None. This is a > good choice for a default, since /everyone/ (even those not using X > every time they deploy virtuals) can view the console. The following > patches push console policy to the user and defaults the console to > text in the event a "-g --graphics" switch is not found. > > ===BEGIN app.py PATCH=== > --- app.py.orig 2008-01-05 05:12:15.000000000 +0000 > +++ app.py 2008-01-05 05:07:45.000000000 +0000 > @@ -148,6 +148,10 @@ > p.add_option("-T", "--virt-type", > dest="virt_type", > help="virtualization install type (xenpv,qemu)") > + p.add_option("-g", "--graphics", > + dest="is_graphics", > + action="store_true", > + help="use vnc console instead of text console") > > (options, args) = p.parse_args() > > @@ -168,6 +172,7 @@ > k.live_cd = options.live_cd > k.virt_path = options.virt_path > k.virt_type = options.virt_type > + k.is_graphics = options.is_graphics > > if options.virt_name is not None: > k.virt_name = options.virt_name > @@ -222,6 +227,7 @@ > self.virt_name = None > self.virt_type = None > self.virt_path = None > + self.is_graphics = None > > #--------------------------------------------------- > > @@ -882,24 +888,26 @@ > pd = profile_data > self.load_virt_modules() > > - arch = self.safe_load(pd,'arch','x86') > - kextra = self.calc_kernel_args(pd) > - (uuid, create_func) = self.virt_choose(pd) > - > - virtname = self.calc_virt_name(pd) > - > - ram = self.calc_virt_ram(pd) > - > - vcpus = self.calc_virt_cpus(pd) > - path_list = self.calc_virt_path(pd, virtname) > - size_list = self.calc_virt_filesize(pd) > - disks = self.merge_disk_data(path_list,size_list) > + arch = self.safe_load(pd,'arch','x86') > + kextra = self.calc_kernel_args(pd) > + (uuid, create_func) = self.virt_choose(pd) > + (graphics, create_func) = self.virt_choose(pd) > + > + virtname = self.calc_virt_name(pd) > + > + ram = self.calc_virt_ram(pd) > + > + vcpus = self.calc_virt_cpus(pd) > + path_list = self.calc_virt_path(pd, virtname) > + size_list = self.calc_virt_filesize(pd) > + disks = > self.merge_disk_data(path_list,size_list) > > results = create_func( > name = virtname, > ram = ram, > disks = disks, > uuid = uuid, > + graphics = graphics, > extra = kextra, > vcpus = vcpus, > profile_data = profile_data, @@ -923,13 +931,14 @@ > > def virt_choose(self, pd): > if self.virt_type == "xenpv": > - uuid = self.get_uuid(self.calc_virt_uuid(pd)) > + graphics = self.is_graphics > + uuid = self.get_uuid(self.calc_virt_uuid(pd)) > import xencreate > - creator = xencreate.start_paravirt_install > + creator = xencreate.start_paravirt_install > elif self.virt_type == "qemu": > - uuid = None > + uuid = None > import qcreate > - creator = qcreate.start_install > + creator = qcreate.start_install > else: > raise InfoException, "Unspecified virt type: %s" % > self.virt_type > return (uuid, creator) > ===END app.py PATCH=== > > ===BEGIN xencreate.py PATCH=== > --- xencreate.py.orig 2008-01-05 05:12:30.000000000 +0000 > +++ xencreate.py 2008-01-05 05:03:20.000000000 +0000 > @@ -47,6 +47,7 @@ > uuid=None, > extra=None, > vcpus=None, + > graphics=None, > profile_data=None, arch=None): > > > @@ -56,7 +57,8 @@ > guest.set_name(name) > guest.set_memory(ram) > guest.set_vcpus(vcpus) > -# guest.set_graphics("vnc") > + if graphics: > + guest.set_graphics("vnc") > if uuid is not None: > guest.set_uuid(uuid) > ===END xencreate.py PATCH=== > > > Regards, > > - Adam. > From mdehaan at redhat.com Mon Jan 7 15:06:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 07 Jan 2008 10:06:22 -0500 Subject: [et-mgmt-tools] Re: PATCH: Graphical options for koan In-Reply-To: <477F1B4C.2060503@gmail.com> References: <477F16DB.9030402@gmail.com> <477F1B4C.2060503@gmail.com> Message-ID: <47823FEE.3000104@redhat.com> Adam Rosenwald wrote: > Sorry, Michael. I didn't see your thread on --nogfx. My patches were > applicable to koan-0.6.3-3 - not GIT (which likely contained your new > code). > Correct. 0.6.4, which has a release up on the testing section of the download URL, has the --nogfx flag. Since that is not dependent on Cobbler's 0.7.x features that can be released in the coming weeks, so we don't need a 0.6.3 1/2 release. > > Adam Rosenwald wrote: >> Michael, >> >> I've noticed that the default graphical console display has changed >> in koan-0.6.3-3 (I had previously been using v0.6.2-2). The default >> display is now set to "vnc". >> >> The default console setting for python-virtinst is None. This is a >> good choice for a default, since /everyone/ (even those not using X >> every time they deploy virtuals) can view the console. The following >> patches push console policy to the user and defaults the console to >> text in the event a "-g --graphics" switch is not found. >> >> ===BEGIN app.py PATCH=== >> --- app.py.orig 2008-01-05 05:12:15.000000000 +0000 >> +++ app.py 2008-01-05 05:07:45.000000000 +0000 >> @@ -148,6 +148,10 @@ >> p.add_option("-T", "--virt-type", >> dest="virt_type", >> help="virtualization install type (xenpv,qemu)") >> + p.add_option("-g", "--graphics", >> + dest="is_graphics", >> + action="store_true", >> + help="use vnc console instead of text console") >> >> (options, args) = p.parse_args() >> >> @@ -168,6 +172,7 @@ >> k.live_cd = options.live_cd >> k.virt_path = options.virt_path >> k.virt_type = options.virt_type >> + k.is_graphics = options.is_graphics >> >> if options.virt_name is not None: >> k.virt_name = options.virt_name >> @@ -222,6 +227,7 @@ >> self.virt_name = None >> self.virt_type = None >> self.virt_path = None >> + self.is_graphics = None >> >> #--------------------------------------------------- >> >> @@ -882,24 +888,26 @@ >> pd = profile_data >> self.load_virt_modules() >> >> - arch = self.safe_load(pd,'arch','x86') >> - kextra = self.calc_kernel_args(pd) >> - (uuid, create_func) = self.virt_choose(pd) >> - >> - virtname = self.calc_virt_name(pd) >> - >> - ram = self.calc_virt_ram(pd) >> - >> - vcpus = self.calc_virt_cpus(pd) >> - path_list = self.calc_virt_path(pd, virtname) >> - size_list = self.calc_virt_filesize(pd) >> - disks = self.merge_disk_data(path_list,size_list) >> + arch = self.safe_load(pd,'arch','x86') >> + kextra = self.calc_kernel_args(pd) >> + (uuid, create_func) = self.virt_choose(pd) >> + (graphics, create_func) = self.virt_choose(pd) >> + >> + virtname = self.calc_virt_name(pd) >> + >> + ram = self.calc_virt_ram(pd) >> + >> + vcpus = self.calc_virt_cpus(pd) >> + path_list = self.calc_virt_path(pd, virtname) >> + size_list = self.calc_virt_filesize(pd) >> + disks = >> self.merge_disk_data(path_list,size_list) >> >> results = create_func( >> name = virtname, >> ram = ram, >> disks = disks, >> uuid = uuid, >> + graphics = graphics, >> extra = kextra, >> vcpus = vcpus, >> profile_data = profile_data, @@ -923,13 >> +931,14 @@ >> >> def virt_choose(self, pd): >> if self.virt_type == "xenpv": >> - uuid = self.get_uuid(self.calc_virt_uuid(pd)) >> + graphics = self.is_graphics >> + uuid = self.get_uuid(self.calc_virt_uuid(pd)) >> import xencreate >> - creator = xencreate.start_paravirt_install >> + creator = xencreate.start_paravirt_install >> elif self.virt_type == "qemu": >> - uuid = None >> + uuid = None >> import qcreate >> - creator = qcreate.start_install >> + creator = qcreate.start_install >> else: >> raise InfoException, "Unspecified virt type: %s" % >> self.virt_type >> return (uuid, creator) >> ===END app.py PATCH=== >> >> ===BEGIN xencreate.py PATCH=== >> --- xencreate.py.orig 2008-01-05 05:12:30.000000000 +0000 >> +++ xencreate.py 2008-01-05 05:03:20.000000000 +0000 >> @@ -47,6 +47,7 @@ >> uuid=None, >> extra=None, >> vcpus=None, + >> graphics=None, >> profile_data=None, arch=None): >> >> >> @@ -56,7 +57,8 @@ >> guest.set_name(name) >> guest.set_memory(ram) >> guest.set_vcpus(vcpus) >> -# guest.set_graphics("vnc") >> + if graphics: >> + guest.set_graphics("vnc") >> if uuid is not None: >> guest.set_uuid(uuid) >> ===END xencreate.py PATCH=== >> >> >> Regards, >> >> - Adam. >> > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mrpquter at yahoo.com Mon Jan 7 21:07:27 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Mon, 7 Jan 2008 13:07:27 -0800 (PST) Subject: [et-mgmt-tools] new cobbler install Message-ID: <236027.47756.qm@web54401.mail.yahoo.com> Hi, I've just added cobbler to an existing, working pxe server with the expectation that it will handle the ks, dhcp, and tftp configuration with one interface. Am I on the right track understanding what cobbler does ? This pxe server has dhcp / tftp / ks files, as I mentioned. The distro bits come from another server via ftp and http, so distros are not kept on the pxe server. A "cobbler check" comes back with the following: reposync is not installed, need for cobbler reposync, install/upgrade yum-utils? yumdownloader is not installed, needed for cobbler repo add with --rpm-list parameter, install/upgrade yum-utils? In my setup, as described above, do I need to install reposync and yumdownloader ? By name they seem to be useful only if I served the distros from the pxe / cobbler server. Cheers. -Mike ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From mdehaan at redhat.com Mon Jan 7 21:27:57 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 07 Jan 2008 16:27:57 -0500 Subject: [et-mgmt-tools] new cobbler install In-Reply-To: <236027.47756.qm@web54401.mail.yahoo.com> References: <236027.47756.qm@web54401.mail.yahoo.com> Message-ID: <4782995D.1060806@redhat.com> Michael Harrison wrote: > Hi, > > I've just added cobbler to an existing, working pxe server with the > expectation that it will handle the ks, dhcp, and tftp configuration > with one interface. Am I on the right track understanding what cobbler > does ? > Yes. > This pxe server has dhcp / tftp / ks files, as I mentioned. The distro > bits come from another server via ftp and http, so distros are not kept > on the pxe server. > > A "cobbler check" comes back with the following: > > reposync is not installed, need for cobbler reposync, install/upgrade > yum-utils? > > yumdownloader is not installed, needed for cobbler repo add with > --rpm-list parameter, install/upgrade yum-utils? > > In my setup, as described above, do I need to install reposync and > yumdownloader ? By name they seem to be useful only if I served the > distros from the pxe / cobbler server. > Createrepo should be installed if you want to use "cobbler repo add" or "cobbler import". See also: http://cobbler.et.redhat.com/cobbler-repos.php > Cheers. > -Mike > > > > > > ____________________________________________________________________________________ > Looking for last minute shopping deals? > Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mrpquter at yahoo.com Mon Jan 7 22:21:02 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Mon, 7 Jan 2008 14:21:02 -0800 (PST) Subject: [et-mgmt-tools] new cobbler install In-Reply-To: <4782995D.1060806@redhat.com> Message-ID: <625354.27518.qm@web54407.mail.yahoo.com> Michael, thank you for the quick help. So "cobbler repo add" is for the case where the repo is on the cobbler server. But don't I need to "cobbler import --available-as=ftp://server:/path ? Thanks, -Mike --- Michael DeHaan wrote: > Michael Harrison wrote: > > Hi, > > > > I've just added cobbler to an existing, working pxe server with the > > expectation that it will handle the ks, dhcp, and tftp > configuration > > with one interface. Am I on the right track understanding what > cobbler > > does ? > > > Yes. > > This pxe server has dhcp / tftp / ks files, as I mentioned. The > distro > > bits come from another server via ftp and http, so distros are not > kept > > on the pxe server. > > > > A "cobbler check" comes back with the following: > > > > reposync is not installed, need for cobbler reposync, > install/upgrade > > yum-utils? > > > > yumdownloader is not installed, needed for cobbler repo add with > > --rpm-list parameter, install/upgrade yum-utils? > > > > In my setup, as described above, do I need to install reposync and > > yumdownloader ? By name they seem to be useful only if I served the > > distros from the pxe / cobbler server. > > > > Createrepo should be installed if you want to use "cobbler repo add" > or > "cobbler import". > > See also: http://cobbler.et.redhat.com/cobbler-repos.php > > > Cheers. > > -Mike > > > > > > > > > > > > > ____________________________________________________________________________________ > > Looking for last minute shopping deals? > > Find them fast with Yahoo! Search. > http://tools.search.yahoo.com/newsearch/category.php?category=shopping > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From rjones at redhat.com Tue Jan 8 11:18:39 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 08 Jan 2008 11:18:39 +0000 Subject: [et-mgmt-tools] Fwd: OCaml bindings - Windows installer Message-ID: <47835C0F.7080101@redhat.com> Might interest some people on this list ... -------------- next part -------------- An embedded message was scrubbed... From: "Richard W.M. Jones" Subject: [Libvir] OCaml bindings - Windows installer Date: Tue, 08 Jan 2008 11:15:19 +0000 Size: 10270 URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From mrpquter at yahoo.com Tue Jan 8 15:36:02 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Tue, 8 Jan 2008 07:36:02 -0800 (PST) Subject: [et-mgmt-tools] new install with existing repos Message-ID: <821601.52985.qm@web54408.mail.yahoo.com> Hi, What's the high-level configuration overview for the case in which repos exist, and there is no need to replicate them on the cobbler server? I haven't figured out how to tell cobbler "the repo is at this ftp or http server/path". I tried cobbler import --available-as, but that complained about missing a --path argument. There is no path, the distro is on the ftp server. A different angle on the same question, I expect cobbler to need what argument to give the url argument in the kickstart file, which is the ftp/http/nfs path I tried passing as the --available-as argument. It shouldn't need to have a repo local on the cobbler system. Thanks, -Mike ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From mdehaan at redhat.com Tue Jan 8 15:34:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 08 Jan 2008 10:34:59 -0500 Subject: [et-mgmt-tools] new cobbler install In-Reply-To: <625354.27518.qm@web54407.mail.yahoo.com> References: <625354.27518.qm@web54407.mail.yahoo.com> Message-ID: <47839823.5080700@redhat.com> Michael Harrison wrote: > Michael, thank you for the quick help. > > So "cobbler repo add" is for the case where the repo is on the cobbler > server. But don't I need to "cobbler import > --available-as=ftp://server:/path ? > > Thanks, > -Mike > > Not in this case, actually. Import does use createrepo when using rsync, because the /etc/cobbler/rsync.excludes file may remove some packages that are deemed unneccessary -- so we run createrepo to repair what we have rebuilt as well as to make packages in the install tree (that are not yet in an upgrades repo) also available as a yum repository. It probably wouldn't hurt to go ahead and install it in case you might want it later. If using EL4, this package should be available in EPEL or EPEL testing. > --- Michael DeHaan wrote: > > >> Michael Harrison wrote: >> >>> Hi, >>> >>> I've just added cobbler to an existing, working pxe server with the >>> expectation that it will handle the ks, dhcp, and tftp >>> >> configuration >> >>> with one interface. Am I on the right track understanding what >>> >> cobbler >> >>> does ? >>> >>> >> Yes. >> >>> This pxe server has dhcp / tftp / ks files, as I mentioned. The >>> >> distro >> >>> bits come from another server via ftp and http, so distros are not >>> >> kept >> >>> on the pxe server. >>> >>> A "cobbler check" comes back with the following: >>> >>> reposync is not installed, need for cobbler reposync, >>> >> install/upgrade >> >>> yum-utils? >>> >>> yumdownloader is not installed, needed for cobbler repo add with >>> --rpm-list parameter, install/upgrade yum-utils? >>> >>> In my setup, as described above, do I need to install reposync and >>> yumdownloader ? By name they seem to be useful only if I served the >>> distros from the pxe / cobbler server. >>> >>> >> Createrepo should be installed if you want to use "cobbler repo add" >> or >> "cobbler import". >> >> See also: http://cobbler.et.redhat.com/cobbler-repos.php >> >> >>> Cheers. >>> -Mike >>> >>> >>> >>> >>> >>> >>> > ____________________________________________________________________________________ > >>> Looking for last minute shopping deals? >>> Find them fast with Yahoo! Search. >>> > http://tools.search.yahoo.com/newsearch/category.php?category=shopping > >>> _______________________________________________ >>> et-mgmt-tools mailing list >>> et-mgmt-tools at redhat.com >>> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >>> >>> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> >> > > > > ____________________________________________________________________________________ > Never miss a thing. Make Yahoo your home page. > http://www.yahoo.com/r/hs > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mdehaan at redhat.com Tue Jan 8 15:39:50 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 08 Jan 2008 10:39:50 -0500 Subject: [et-mgmt-tools] new install with existing repos In-Reply-To: <821601.52985.qm@web54408.mail.yahoo.com> References: <821601.52985.qm@web54408.mail.yahoo.com> Message-ID: <47839946.80402@redhat.com> Michael Harrison wrote: > Hi, > > What's the high-level configuration overview for the case in which > repos exist, and there is no need to replicate them on the cobbler > server? > Add them to your kickstart file as normal in that case. There is an RFE for being able to represent external repos in the cobbler configuration for easy management (in fact, I think I filed it) but nothing is implemented quite yet. > I haven't figured out how to tell cobbler "the repo is at this ftp or > http server/path". > > cobbler repo add --name=foo --path=http://blah is how this is done for repositories that you want to mirror. For repositories that you do not want to mirror, you can just add them into your kickstart file/template manually. > I tried cobbler import --available-as, but that complained about > missing a --path argument. There is no path, the distro is on the ftp > server. > Cobbler import does require a filesystem path, --available-as was created for doing imports of distribution trees that are also available on external filers, so trees can be scanned over NFS and not duplicated. (Do not confuse install trees with yum repositories.) > A different angle on the same question, I expect cobbler to need what > argument to give the url argument in the kickstart file, which is the > ftp/http/nfs path I tried passing as the --available-as argument. Yes, import sets the url parameter automagically. If you do an import (a DVD import may be easier for starters, if you are having trouble figuring out what --available as is, it's a niche case, as I've said, for external filers) and then cat /var/lib/cobbler/distros, you'll see that there is a metadata variable set for "tree" and a matching "$tree" in the kickstart template where this is filled in. See here: http://cobbler.et.redhat.com/cobbler-import.php > It > shouldn't need to have a repo local on the cobbler system. > > Correct. > Thanks, > -Mike > > > > ____________________________________________________________________________________ > Never miss a thing. Make Yahoo your home page. > http://www.yahoo.com/r/hs > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mrpquter at yahoo.com Tue Jan 8 16:40:48 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Tue, 8 Jan 2008 08:40:48 -0800 (PST) Subject: [et-mgmt-tools] new install with existing repos In-Reply-To: <47839946.80402@redhat.com> Message-ID: <596294.87908.qm@web54408.mail.yahoo.com> Okay, I've decided not to configure cobbler on my working pxe server and put it on new hardware instead. I'll configure it they way it was intended, adding distros locally, and exploring what it does. Thank you for the great help! -Mike --- Michael DeHaan wrote: > Michael Harrison wrote: > > Hi, > > > > What's the high-level configuration overview for the case in which > > repos exist, and there is no need to replicate them on the cobbler > > server? > > > > Add them to your kickstart file as normal in that case. > > There is an RFE for being able to represent external repos in the > cobbler configuration > for easy management (in fact, I think I filed it) but nothing is > implemented quite yet. > > I haven't figured out how to tell cobbler "the repo is at this ftp > or > > http server/path". > > > > > > cobbler repo add --name=foo --path=http://blah is how this is done > for > repositories that you > want to mirror. > > For repositories that you do not want to mirror, you can just add > them > into your kickstart file/template > manually. > > > I tried cobbler import --available-as, but that complained about > > missing a --path argument. There is no path, the distro is on the > ftp > > server. > > > Cobbler import does require a filesystem path, --available-as was > created for doing imports of distribution trees that > are also available on external filers, so trees can be scanned over > NFS > and not duplicated. > > (Do not confuse install trees with yum repositories.) > > > > A different angle on the same question, I expect cobbler to need > what > > argument to give the url argument in the kickstart file, which is > the > > ftp/http/nfs path I tried passing as the --available-as argument. > > Yes, import sets the url parameter automagically. If you do an > import > (a DVD import > may be easier for starters, if you are having trouble figuring out > what > --available as is, it's a niche > case, as I've said, for external filers) and then cat > /var/lib/cobbler/distros, you'll see that there is a metadata > variable set for "tree" and a matching "$tree" in the kickstart > template > where this is filled in. > > See here: > > http://cobbler.et.redhat.com/cobbler-import.php > > > It > > shouldn't need to have a repo local on the cobbler system. > > > > > Correct. > > > Thanks, > > -Mike > > > > > > > > > ____________________________________________________________________________________ > > Never miss a thing. Make Yahoo your home page. > > http://www.yahoo.com/r/hs > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From soren at ubuntu.com Wed Jan 9 15:11:31 2008 From: soren at ubuntu.com (Soren Hansen) Date: Wed, 9 Jan 2008 16:11:31 +0100 Subject: [et-mgmt-tools] Bogus heuristics in virt-manager Message-ID: <20080109151131.GY1707@atlas.linux2go.dk> Hi! I've run into a very annoying issue in virt-manager. I primarily use libvirt to manage kvm sessions at qemu:///system. I do this as my regular user on my system (I have write access to the appropriate libvirtd socket). If I hand craft the XML describing my virtual machine, I can connect to virtual networks, bridge onto the existing LAN, etc. I can also set up new virtual networks and basically anything else that root can do. However, virt-manager still treats me as a lowly user with no particular privileges. This is because virt-manager's code is full of (os.getuid() == 0) checks, which in my case are completely wrong. I see a few solutions to this: a) Clean up all these calls and replace them with a "can_do_fancy_stuff(con)" call and add the necessary logic to that new function. b) Move the logic to libvirt and provide a virConnectIsPrivileged(virConnectPtr *) function that does basically the same thing. c) Take a more granular approach to b) and provide a way to query if specific tasks can be done through the given connection. a) Is quite easy. I could probably have that done in no more than a few hours. b) is almost as easy to do, but I'm not sure if all the existing (or future) vm types have such a simple model (either you can do only basic stuff, or you can do everything). c) smells like the right way to do it, takes slightly longer, but should be relatively future proof. -- Soren Hansen Ubuntu Server Team http://www.ubuntu.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From berrange at redhat.com Wed Jan 9 15:16:42 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 9 Jan 2008 15:16:42 +0000 Subject: [et-mgmt-tools] Bogus heuristics in virt-manager In-Reply-To: <20080109151131.GY1707@atlas.linux2go.dk> References: <20080109151131.GY1707@atlas.linux2go.dk> Message-ID: <20080109151642.GC8732@redhat.com> On Wed, Jan 09, 2008 at 04:11:31PM +0100, Soren Hansen wrote: > Hi! > > I've run into a very annoying issue in virt-manager. > > I primarily use libvirt to manage kvm sessions at qemu:///system. I do > this as my regular user on my system (I have write access to the > appropriate libvirtd socket). If I hand craft the XML describing my > virtual machine, I can connect to virtual networks, bridge onto the > existing LAN, etc. I can also set up new virtual networks and basically > anything else that root can do. However, virt-manager still treats me as > a lowly user with no particular privileges. This is because > virt-manager's code is full of (os.getuid() == 0) checks, which in my > case are completely wrong. Yep, these checks are all targetted for removal. Basically they fall into a couple of categories: - Checking OS state /proc/xen, /sysfs /dev/kvm to see what HV capabilities/properties you can use. These are targetted to replacement with the virConnectGetCapabilities() APIs - Creating/managing storage. These are targetted for replacement with the forthcoming storage management APIs - Enumerating network devices. TBD. I've been doing some preparatory work in virtinst, to get ready for the first point in virt-manager. The storage stuff is active work in progress too. Once these are addressed there should be no use of anything in the python 'os.' module, and thus no getuid() checks anywhere. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From soren at ubuntu.com Wed Jan 9 15:24:32 2008 From: soren at ubuntu.com (Soren Hansen) Date: Wed, 9 Jan 2008 16:24:32 +0100 Subject: [et-mgmt-tools] Bogus heuristics in virt-manager In-Reply-To: <20080109151642.GC8732@redhat.com> References: <20080109151131.GY1707@atlas.linux2go.dk> <20080109151642.GC8732@redhat.com> Message-ID: <20080109152432.GZ1707@atlas.linux2go.dk> On Wed, Jan 09, 2008 at 03:16:42PM +0000, Daniel P. Berrange wrote: > > This is because virt-manager's code is full of (os.getuid() == 0) > > checks, which in my case are completely wrong. > Yep, these checks are all targetted for removal. Basically they fall > into a couple of categories: > > - Checking OS state /proc/xen, /sysfs /dev/kvm to see what HV > capabilities/properties you can use. These are targetted to > replacement with the virConnectGetCapabilities() APIs > > - Creating/managing storage. These are targetted for replacement > with the forthcoming storage management APIs > > - Enumerating network devices. TBD. Sounds fantastic. Is this in mercurial somewhere? Is there a roadmap for the various tools anywhere? -- Soren Hansen Ubuntu Server Team http://www.ubuntu.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From berrange at redhat.com Wed Jan 9 15:43:25 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 9 Jan 2008 15:43:25 +0000 Subject: [et-mgmt-tools] Bogus heuristics in virt-manager In-Reply-To: <20080109152432.GZ1707@atlas.linux2go.dk> References: <20080109151131.GY1707@atlas.linux2go.dk> <20080109151642.GC8732@redhat.com> <20080109152432.GZ1707@atlas.linux2go.dk> Message-ID: <20080109154325.GD8732@redhat.com> On Wed, Jan 09, 2008 at 04:24:32PM +0100, Soren Hansen wrote: > On Wed, Jan 09, 2008 at 03:16:42PM +0000, Daniel P. Berrange wrote: > > > This is because virt-manager's code is full of (os.getuid() == 0) > > > checks, which in my case are completely wrong. > > Yep, these checks are all targetted for removal. Basically they fall > > into a couple of categories: > > > > - Checking OS state /proc/xen, /sysfs /dev/kvm to see what HV > > capabilities/properties you can use. These are targetted to > > replacement with the virConnectGetCapabilities() APIs > > > > - Creating/managing storage. These are targetted for replacement > > with the forthcoming storage management APIs > > > > - Enumerating network devices. TBD. > > > Sounds fantastic. Is this in mercurial somewhere? All the virt tools are here: http://hg.et.redhat.com/virt/ The storage API stuff is in a patch queue on my laptop, which I'll be posting to libvir-list today/tomorrow for another round of review. > Is there a roadmap for the various tools anywhere? There's a (unfortunately out of date) roadmap for virt-manager here http://virt-manager.org/roadmap.html Feel free to start a Wiki page with more up2date ideas / suggestions - a Wiki is more suited to a roadmap than our current static html page. The wiki is here: http://virt-manager.org/page/Main_Page Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From viswanath.t at atc.tcs.com Thu Jan 10 11:37:59 2008 From: viswanath.t at atc.tcs.com (Viswanath T K) Date: Thu, 10 Jan 2008 17:07:59 +0530 Subject: [et-mgmt-tools] Installation problem with cobbler In-Reply-To: <20080109151131.GY1707@atlas.linux2go.dk> References: <20080109151131.GY1707@atlas.linux2go.dk> Message-ID: <1199965079.2958.15.camel@localhost.localdomain> Hi , I am Viswanath T K, I need some help on using cobbler. I have been trying to do a network installation using cobbler. The steps I followed are: 1. On machine A, I have DHCP server, cobbler, tftp-server, apache installed and working. 2. On Machine A, I have created a cobbler distribution, profile and system and edited the content of the var/lib/cobbler/settings. 3. In the /tftpboot/pxelinux.cfg/ directory there is a file by name default. whose contents are as follows: DEFAULT local PROMPT 1 MENU TITLE Cobbler | http://cobbler.et.redhat.com TIMEOUT 200 TOTALTIMEOUT 6000 ONTIMEOUT local LABEL local MENU LABEL (local) MENU DEFAULT LOCALBOOT 0 LABEL fc7xenwebservers kernel /images/fc7xen/vmlinuz-2.6.20-2936.fc7xen MENU LABEL fc7xenwebservers append ksdevice=eth0 lang= kssendmac syslog=192.168.122.1:25150 text initrd=/images/fc7xen/initrd-2.6.20-2936.fc7xen.img ks=http://192.168.122.1/cblr/kickstarts/fc7xenwebservers/ks.cfg 4. Machine B is a raw machine which I boot using the Onboard Network Controller. 5. Machine B boots and then it is assigned an IP by the DHCP server present on Machine A. 6. I can see the boot prompt and there I type menu. I see the screen as below : 7. When I select the I get the error, "Invalid or corrupt kernel image". Can anyone help me ? Contents of the file /var/lib/cobbler/settings bootloaders: ia64: /var/lib/cobbler/elilo-3.6-ia64.efi standard: /usr/lib/syslinux/pxelinux.0 default_kickstart: /etc/cobbler/default.ks default_virt_type: auto dhcpd_bin: /usr/sbin/dhcpd dhcpd_conf: /etc/dhcpd.conf dnsmasq_bin: /usr/sbin/dnsmasq dnsmasq_conf: /etc/dnsmasq.conf httpd_bin: /usr/sbin/httpd kernel_options: ksdevice: eth0 lang: ' ' text: ~ manage_dhcp: 0 manage_dhcp_mode: isc next_server: '172.19.58.120' pxe_just_once: 0 server: '172.19.58.120' snippetsdir: /var/lib/cobbler/snippets syslog_port: 25150 tftpboot: /tftpboot tftpd_bin: /usr/sbin/in.tftpd tftpd_conf: /etc/xinetd.d/tftp webdir: /var/www/cobbler xmlrpc_port: 25151 xmlrpc_rw_enabled:1 xmlrpc_rw_port: 25152 yum_core_mirror_from_server: 0 Thanks & Regards, Viswanath T K From viswanath.t at atc.tcs.com Thu Jan 10 11:42:19 2008 From: viswanath.t at atc.tcs.com (Viswanath T K) Date: Thu, 10 Jan 2008 17:12:19 +0530 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <20080109151131.GY1707@atlas.linux2go.dk> References: <20080109151131.GY1707@atlas.linux2go.dk> Message-ID: <1199965339.2958.17.camel@localhost.localdomain> Hi , I have been trying to do a network installation using cobbler. The steps I followed are: 1. On machine A, I have DHCP server, cobbler, tftp-server, apache installed and working. 2. On Machine A, I have created a cobbler distribution, profile and system and edited the content of the var/lib/cobbler/settings. 3. In the /tftpboot/pxelinux.cfg/ directory there is a file by name default. whose contents are as follows: DEFAULT local PROMPT 1 MENU TITLE Cobbler | http://cobbler.et.redhat.com TIMEOUT 200 TOTALTIMEOUT 6000 ONTIMEOUT local LABEL local MENU LABEL (local) MENU DEFAULT LOCALBOOT 0 LABEL fc7xenwebservers kernel /images/fc7xen/vmlinuz-2.6.20-2936.fc7xen MENU LABEL fc7xenwebservers append ksdevice=eth0 lang= kssendmac syslog=192.168.122.1:25150 text initrd=/images/fc7xen/initrd-2.6.20-2936.fc7xen.img ks=http://192.168.122.1/cblr/kickstarts/fc7xenwebservers/ks.cfg 4. Machine B is a raw machine which I boot using the Onboard Network Controller. 5. Machine B boots and then it is assigned an IP by the DHCP server present on Machine A. 6. I can see the boot prompt and there I type menu. I see the screen as below : I need to select between local & fc7xenwebservers 7. When I select fc7xenwebservers then I get the error, "Invalid or corrupt kernel image". Can anyone help me ? If I select local nothing happens. Contents of the file /var/lib/cobbler/settings bootloaders: ia64: /var/lib/cobbler/elilo-3.6-ia64.efi standard: /usr/lib/syslinux/pxelinux.0 default_kickstart: /etc/cobbler/default.ks default_virt_type: auto dhcpd_bin: /usr/sbin/dhcpd dhcpd_conf: /etc/dhcpd.conf dnsmasq_bin: /usr/sbin/dnsmasq dnsmasq_conf: /etc/dnsmasq.conf httpd_bin: /usr/sbin/httpd kernel_options: ksdevice: eth0 lang: ' ' text: ~ manage_dhcp: 0 manage_dhcp_mode: isc next_server: '172.19.58.120' pxe_just_once: 0 server: '172.19.58.120' snippetsdir: /var/lib/cobbler/snippets syslog_port: 25150 tftpboot: /tftpboot tftpd_bin: /usr/sbin/in.tftpd tftpd_conf: /etc/xinetd.d/tftp webdir: /var/www/cobbler xmlrpc_port: 25151 xmlrpc_rw_enabled:1 xmlrpc_rw_port: 25152 yum_core_mirror_from_server: 0 Thanks & Regards, Viswanath T K From rainer at ultra-secure.de Thu Jan 10 11:46:43 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Thu, 10 Jan 2008 12:46:43 +0100 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <1199965339.2958.17.camel@localhost.localdomain> References: <20080109151131.GY1707@atlas.linux2go.dk> <1199965339.2958.17.camel@localhost.localdomain> Message-ID: <478605A3.2080604@ultra-secure.de> Viswanath T K wrote: > Hi , > > I have been trying to do a network installation using cobbler. > The steps I followed are: > 1. On machine A, I have DHCP server, cobbler, tftp-server, apache > installed and working. > 2. On Machine A, I have created a cobbler distribution, profile and > system and edited the content of the var/lib/cobbler/settings. > 3. In the /tftpboot/pxelinux.cfg/ directory there is a file by name > default. whose contents are as follows: > No. There must be a file named according to the MAC address of the box you want to PXE-boot. Did you sync? cobbler sync cheers, Rainer From mdehaan at redhat.com Thu Jan 10 15:35:58 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 10:35:58 -0500 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <1199965339.2958.17.camel@localhost.localdomain> References: <20080109151131.GY1707@atlas.linux2go.dk> <1199965339.2958.17.camel@localhost.localdomain> Message-ID: <47863B5E.3030800@redhat.com> > > > 7. When I select fc7xenwebservers then I get the error, "Invalid or > corrupt kernel image". Can anyone help me ? > If I select local nothing happens. > You can't PXE boot a Xen kernel/initrd pair. Those are for installing paravirt Xen guests, not hosts. I would suspect if you did "cobbler profile report fc7xenwebservers" you would find that the distro has the name "xen" in it somewhere. --Michael From agx at sigxcpu.org Thu Jan 10 16:33:07 2008 From: agx at sigxcpu.org (Guido Guenther) Date: Thu, 10 Jan 2008 17:33:07 +0100 Subject: [et-mgmt-tools] virt-manager: handling of bonding and bridge devices In-Reply-To: <20071218153414.GA16706@bogon.ms20.nix> References: <20071218153414.GA16706@bogon.ms20.nix> Message-ID: <20080110163307.GA1133@bogon.ms20.nix> On Tue, Dec 18, 2007 at 04:34:14PM +0100, Guido Guenther wrote: > Hi, > currently virt-manager uses some linux specific heuristics to find out > if a device is part of a bridge. This immediatley fails if the > underlying device is not a physical device but e.g. a bonding device. > > Hal doesn't seem to care about either bridges nor bonding devices. I > wonder where to fix this? Add another linux specific bonding hack > to determine if a eth devices is part of a bonding device which might be > part of a bridge? Would be easy enough to do but I wonder if it wouldn't > be better if hal would export this information in the first place? Hmm...no comments - would another hack that treats bridges on top of bond devices be o.k.? Cheers, -- Guido From mdehaan at redhat.com Thu Jan 10 20:59:03 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 15:59:03 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.6.5, Koan 0.6.4 Message-ID: <47868717.5040702@redhat.com> New releases... Nothing major. This is mainly just to update the apps with a few needed features while development on Cobbler 0.7.X continues. Cobbler 0.6.5 changes: - Fix for f8 comps.xml name change - Fix for repo file permissions during cobbler reposync Koan 0.6.4 has a few more features: - Added --nogfx option when creating virtual guests to disable VNC and just use the console - Allow Xen machines to be constructed with no disks (size=0), but not qemu/KVM - Support Xen Fullvirt via PXE (requires Cobbler 0.7.X) - Add new --autonet parameter (see manpage) - Fail if kernel argument length exceeds the 255 cap, as truncation can lead to hard to debug errors. http://cobbler.et.redhat.com/download/ Should be pushed to Fedora/EPEL shortly... --Michael From berrange at redhat.com Thu Jan 10 21:11:08 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 10 Jan 2008 21:11:08 +0000 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.6.5, Koan 0.6.4 In-Reply-To: <47868717.5040702@redhat.com> References: <47868717.5040702@redhat.com> Message-ID: <20080110211108.GI9706@redhat.com> On Thu, Jan 10, 2008 at 03:59:03PM -0500, Michael DeHaan wrote: > New releases... > > Nothing major. This is mainly just to update the apps with a few needed > features while development on Cobbler 0.7.X continues. > > Cobbler 0.6.5 changes: > - Fix for f8 comps.xml name change > - Fix for repo file permissions during cobbler reposync > > Koan 0.6.4 has a few more features: > - Added --nogfx option when creating virtual guests to disable VNC and > just use the console > - Allow Xen machines to be constructed with no disks (size=0), but not > qemu/KVM > - Support Xen Fullvirt via PXE (requires Cobbler 0.7.X) FYI, F9 Xen will (almost certainly) contain support for booting Xen fullvirt guests with a direct kernel + initrd. I've got the patches in rawhide - Xen config syntax is the same as for paravirt guests. Need to get them accepted by upstream Xen, but once that's done I'll enable support in libvirt. So we'll have parity in terms of local boot functionality between Xen PV, Xen HVM and KVM. Just PXE for Xen paravirt to figure out now... Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Thu Jan 10 21:22:15 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 16:22:15 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.6.5, Koan 0.6.4 In-Reply-To: <20080110211108.GI9706@redhat.com> References: <47868717.5040702@redhat.com> <20080110211108.GI9706@redhat.com> Message-ID: <47868C87.50408@redhat.com> Daniel P. Berrange wrote: > On Thu, Jan 10, 2008 at 03:59:03PM -0500, Michael DeHaan wrote: > >> New releases... >> >> Nothing major. This is mainly just to update the apps with a few needed >> features while development on Cobbler 0.7.X continues. >> >> Cobbler 0.6.5 changes: >> - Fix for f8 comps.xml name change >> - Fix for repo file permissions during cobbler reposync >> >> Koan 0.6.4 has a few more features: >> - Added --nogfx option when creating virtual guests to disable VNC and >> just use the console >> - Allow Xen machines to be constructed with no disks (size=0), but not >> qemu/KVM >> - Support Xen Fullvirt via PXE (requires Cobbler 0.7.X) >> > > FYI, F9 Xen will (almost certainly) contain support for booting Xen > fullvirt guests with a direct kernel + initrd. I've got the patches > in rawhide - Xen config syntax is the same as for paravirt guests. Great! PXE is still undoable for a large portion of the install base (because of static IPs or other IT reasons), so that will be great thing to have. > Need > to get them accepted by upstream Xen, but once that's done I'll enable > support in libvirt. So we'll have parity in terms of local boot > functionality between Xen PV, Xen HVM and KVM. Just PXE for Xen paravirt > to figure out now... > Yep, sounds good, let me know when I can add them to koan and I'll get it in there! > Dan. > From mdehaan at redhat.com Thu Jan 10 23:18:57 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 18:18:57 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.6.5, Koan 0.6.4 In-Reply-To: <47868717.5040702@redhat.com> References: <47868717.5040702@redhat.com> Message-ID: <4786A7E1.7060108@redhat.com> Michael DeHaan wrote: > New releases... FYI ... the builders appear to be running slowly, so if these updates do not show up in Fedora/EPEL shortly, I will submit them next week. Until then, feel free to use the uploaded RPMs. The versions in source control differ /slightly/ in that they are buildable in F-9, but no other changes. --Michael From sakaia at jp.fujitsu.com Fri Jan 11 00:55:42 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Fri, 11 Jan 2008 09:55:42 +0900 Subject: [et-mgmt-tools] [patch][virt-viewer] typo Message-ID: <200801110055.m0B0tqTO004091@fjmscan503.ms.jp.fujitsu.com> Fix typos on virt-viewer. since Viewer Viewer is impressive. Anyway if this ML is wrong for virt-viewer, please suggest me Thanks Atsushi SAKAI -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-viewer-typo.patch Type: application/octet-stream Size: 2953 bytes Desc: not available URL: From berrange at redhat.com Fri Jan 11 01:20:10 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 11 Jan 2008 01:20:10 +0000 Subject: [et-mgmt-tools] [patch][virt-viewer] typo In-Reply-To: <200801110055.m0B0tqTO004091@fjmscan503.ms.jp.fujitsu.com> References: <200801110055.m0B0tqTO004091@fjmscan503.ms.jp.fujitsu.com> Message-ID: <20080111012010.GB26351@redhat.com> On Fri, Jan 11, 2008 at 09:55:42AM +0900, Atsushi SAKAI wrote: > Fix typos on virt-viewer. > since Viewer Viewer is impressive. > > Anyway if this ML is wrong for virt-viewer, > please suggest me This is the right mailing list - I have applied your patch - thanks ! Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Fri Jan 11 02:05:10 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 11 Jan 2008 02:05:10 +0000 Subject: [et-mgmt-tools] ANNOUNCE: virt-manager 0.5.3 and virtinst 0.300.2 releases Message-ID: <20080111020510.GC26351@redhat.com> I'm happy to announce the availability of virt-manager 0.5.3 and virtinst 0.300.2 bug fix releases on the download page: http://virt-manager.org/download.html The virt-manager-0.5.3 release: http://virt-manager.org/download/sources/virt-manager/virt-manager-0.5.3.tar.gz This is a bug fix release. The sizing of the VNC window is fixed for screens where the physical size is less than the guest screen size. The 'new vm' button is switched back to its old (more obvious style/ placement). Restore of VMs is working again for local connections. A menu for sending special key sequences to the guest is added. Lots of other misc bug fixes The virtinst 0.300.2 release: http://virt-manager.org/download/sources/virtinst/virtinst-0.300.2.tar.gz This release fixes a couple of minor bugs. XML attributes are correctly escaped in the XML. vCPUs can be pinned to pCPUs at install time. Windows guests will be automatically configured to use localtime. Improved input validation. Thanks to everyone who has contributed to these releases via patches, bug reports & fixes, new features, and general testing & comments / feedback Regards, Dan -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From smooge at gmail.com Fri Jan 11 03:28:52 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Thu, 10 Jan 2008 20:28:52 -0700 Subject: [et-mgmt-tools] ANNOUNCE: virt-manager 0.5.3 and virtinst 0.300.2 releases In-Reply-To: <20080111020510.GC26351@redhat.com> References: <20080111020510.GC26351@redhat.com> Message-ID: <80d7e4090801101928j13d2cebcob726055eafd3f828@mail.gmail.com> On Jan 10, 2008 7:05 PM, Daniel P. Berrange wrote: > I'm happy to announce the availability of virt-manager 0.5.3 and > virtinst 0.300.2 bug fix releases on the download page: > > http://virt-manager.org/download.html > > The virt-manager-0.5.3 release: > > http://virt-manager.org/download/sources/virt-manager/virt-manager-0.5.3.tar.gz > > This is a bug fix release. The sizing of the VNC window is fixed for > screens where the physical size is less than the guest screen size. > The 'new vm' button is switched back to its old (more obvious style/ > placement). Restore of VMs is working again for local connections. A > menu for sending special key sequences to the guest is added. Lots > of other misc bug fixes > What is the status of these working with RHEL-5? I am mostly trying to gauge if these are for Fedora 9 or could be used with CentOS/RHEL? -- Stephen J Smoogen. -- CSIRT/Linux System Administrator How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From berrange at redhat.com Fri Jan 11 03:37:42 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 11 Jan 2008 03:37:42 +0000 Subject: [et-mgmt-tools] ANNOUNCE: virt-manager 0.5.3 and virtinst 0.300.2 releases In-Reply-To: <80d7e4090801101928j13d2cebcob726055eafd3f828@mail.gmail.com> References: <20080111020510.GC26351@redhat.com> <80d7e4090801101928j13d2cebcob726055eafd3f828@mail.gmail.com> Message-ID: <20080111033742.GA25544@redhat.com> On Thu, Jan 10, 2008 at 08:28:52PM -0700, Stephen John Smoogen wrote: > On Jan 10, 2008 7:05 PM, Daniel P. Berrange wrote: > > I'm happy to announce the availability of virt-manager 0.5.3 and > > virtinst 0.300.2 bug fix releases on the download page: > > > > http://virt-manager.org/download.html > > > > The virt-manager-0.5.3 release: > > > > http://virt-manager.org/download/sources/virt-manager/virt-manager-0.5.3.tar.gz > > > > This is a bug fix release. The sizing of the VNC window is fixed for > > screens where the physical size is less than the guest screen size. > > The 'new vm' button is switched back to its old (more obvious style/ > > placement). Restore of VMs is working again for local connections. A > > menu for sending special key sequences to the guest is added. Lots > > of other misc bug fixes > > > > What is the status of these working with RHEL-5? I am mostly trying to > gauge if these are for Fedora 9 or could be used with CentOS/RHEL? They won't work on RHEL-5.1 because they require a newer version of libvirt than is present, and remote management also requires newer support for VNC encryption in Xen. They should work on F-7 or later,and I will be pushing updates to Fedora soon Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From viswanath.t at atc.tcs.com Fri Jan 11 10:16:19 2008 From: viswanath.t at atc.tcs.com (Viswanath T K) Date: Fri, 11 Jan 2008 15:46:19 +0530 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <47863B5E.3030800@redhat.com> References: <20080109151131.GY1707@atlas.linux2go.dk> <1199965339.2958.17.camel@localhost.localdomain> <47863B5E.3030800@redhat.com> Message-ID: <1200046579.2958.31.camel@localhost.localdomain> Hi, I did try using a Fedora 7 distribution and cobbler did work. But it didnt work for the a kernel with Xen support. cant we do that ? the output of cobbler profile report fc7xenwebservers is as below : profile : fc7xenwebservers distro : fc7xen kickstart : /etc/cobbler/default.ks kernel options : {} ks metadata : {} virt file size : 5 virt ram : 512 virt type : auto virt path : virt bridge : xenbr0 virt cpus : 1 repos : [] dhcp tag : default server : <> Thanks & Regards, Viswanath T K On Thu, 2008-01-10 at 10:35 -0500, Michael DeHaan wrote: > > > > > > 7. When I select fc7xenwebservers then I get the error, "Invalid or > > corrupt kernel image". Can anyone help me ? > > If I select local nothing happens. > > > > You can't PXE boot a Xen kernel/initrd pair. > > Those are for installing paravirt Xen guests, not hosts. > > I would suspect if you did "cobbler profile report fc7xenwebservers" you > would find > that the distro has the name "xen" in it somewhere. > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From chris.sarginson at ukfast.net Fri Jan 11 10:34:50 2008 From: chris.sarginson at ukfast.net (Chris Sarginson) Date: Fri, 11 Jan 2008 10:34:50 +0000 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <1200046579.2958.31.camel@localhost.localdomain> References: <20080109151131.GY1707@atlas.linux2go.dk> <1199965339.2958.17.camel@localhost.localdomain> <47863B5E.3030800@redhat.com> <1200046579.2958.31.camel@localhost.localdomain> Message-ID: <4787464A.4090107@ukfast.net> Viswanath T K wrote: > Hi, > > I did try using a Fedora 7 distribution and cobbler did work. But it > didnt work for the a kernel with Xen support. > > cant we do that ? I think what you will need to do is use the Fedora 7 distro to do the install, however in your kickstart file you should include kernel-xen so that the xen kernel is installed if you are going to be setting up xen vm's on the server Chris > > the output of cobbler profile report fc7xenwebservers is as below : > profile : fc7xenwebservers > distro : fc7xen > kickstart : /etc/cobbler/default.ks > kernel options : {} > ks metadata : {} > virt file size : 5 > virt ram : 512 > virt type : auto > virt path : > virt bridge : xenbr0 > virt cpus : 1 > repos : [] > dhcp tag : default > server : <> > > > Thanks & Regards, > Viswanath T K > > On Thu, 2008-01-10 at 10:35 -0500, Michael DeHaan wrote: >>> >>> 7. When I select fc7xenwebservers then I get the error, "Invalid or >>> corrupt kernel image". Can anyone help me ? >>> If I select local nothing happens. >>> >> You can't PXE boot a Xen kernel/initrd pair. >> >> Those are for installing paravirt Xen guests, not hosts. >> >> I would suspect if you did "cobbler profile report fc7xenwebservers" you >> would find >> that the distro has the name "xen" in it somewhere. >> >> --Michael >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools -- Kind regards Chris Sarginson Technical Support UKFast.Net Ltd (t) 0870 111 8866 (f) 0870 458 4545 "The UK's Best Hosting Provider" ISPA Awards 2007, 2006 and 2005 Dedicated Servers - Managed Hosting - Domain Names- http://www.ukfast.net UKFast.Net Ltd, City Tower, Piccadilly Plaza, Manchester, M1 4BT Registered in England. Number 384 5616 From smooge at gmail.com Fri Jan 11 17:46:12 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Fri, 11 Jan 2008 10:46:12 -0700 Subject: [et-mgmt-tools] ANNOUNCE: virt-manager 0.5.3 and virtinst 0.300.2 releases In-Reply-To: <20080111033742.GA25544@redhat.com> References: <20080111020510.GC26351@redhat.com> <80d7e4090801101928j13d2cebcob726055eafd3f828@mail.gmail.com> <20080111033742.GA25544@redhat.com> Message-ID: <80d7e4090801110946h2310950bh8c3b160e0e9672be@mail.gmail.com> On Jan 10, 2008 8:37 PM, Daniel P. Berrange wrote: > On Thu, Jan 10, 2008 at 08:28:52PM -0700, Stephen John Smoogen wrote: > > On Jan 10, 2008 7:05 PM, Daniel P. Berrange wrote: > > > I'm happy to announce the availability of virt-manager 0.5.3 and > > > virtinst 0.300.2 bug fix releases on the download page: > > > > > > http://virt-manager.org/download.html > > > > > > The virt-manager-0.5.3 release: > > > > > > http://virt-manager.org/download/sources/virt-manager/virt-manager-0.5.3.tar.gz > > > > > > This is a bug fix release. The sizing of the VNC window is fixed for > > > screens where the physical size is less than the guest screen size. > > > The 'new vm' button is switched back to its old (more obvious style/ > > > placement). Restore of VMs is working again for local connections. A > > > menu for sending special key sequences to the guest is added. Lots > > > of other misc bug fixes > > > > > > > What is the status of these working with RHEL-5? I am mostly trying to > > gauge if these are for Fedora 9 or could be used with CentOS/RHEL? > > They won't work on RHEL-5.1 because they require a newer version of libvirt > than is present, and remote management also requires newer support for VNC > encryption in Xen. They should work on F-7 or later,and I will be pushing > updates to Fedora soon > Foo. Sigh.. ok thanks. the libvirt I could see updating.. but the Xen changes would be a massive item. -- Stephen J Smoogen. -- CSIRT/Linux System Administrator How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From mdehaan at redhat.com Fri Jan 11 18:09:30 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 11 Jan 2008 13:09:30 -0500 Subject: [et-mgmt-tools] Installation issue with cobbler In-Reply-To: <4787464A.4090107@ukfast.net> References: <20080109151131.GY1707@atlas.linux2go.dk> <1199965339.2958.17.camel@localhost.localdomain> <47863B5E.3030800@redhat.com> <1200046579.2958.31.camel@localhost.localdomain> <4787464A.4090107@ukfast.net> Message-ID: <4787B0DA.5070709@redhat.com> Chris Sarginson wrote: > Viswanath T K wrote: >> Hi, >> >> I did try using a Fedora 7 distribution and cobbler did work. But it >> didnt work for the a kernel with Xen support. >> >> cant we do that ? > > I think what you will need to do is use the Fedora 7 distro to do the > install, however in your kickstart file you should include kernel-xen > so that the xen kernel is installed if you are going to be setting up > xen vm's on the server > > Chris > Yes, if you are installing on real metal that is what you should do. >> >> the output of cobbler profile report fc7xenwebservers is as below : >> profile : fc7xenwebservers >> distro : fc7xen >> kickstart : /etc/cobbler/default.ks >> kernel options : {} >> ks metadata : {} >> virt file size : 5 >> virt ram : 512 >> virt type : auto >> virt path : virt bridge : xenbr0 >> virt cpus : 1 >> repos : [] >> dhcp tag : default >> server : <> >> >> >> Thanks & Regards, >> Viswanath T K >> On Thu, 2008-01-10 at 10:35 -0500, Michael DeHaan wrote: >>>> >>>> 7. When I select fc7xenwebservers then I get the error, "Invalid or >>>> corrupt kernel image". Can anyone help me ? >>>> If I select local nothing happens. >>>> >>> You can't PXE boot a Xen kernel/initrd pair. >>> >>> Those are for installing paravirt Xen guests, not hosts. >>> >>> I would suspect if you did "cobbler profile report fc7xenwebservers" >>> you would find >>> that the distro has the name "xen" in it somewhere. >>> >>> --Michael >>> >>> _______________________________________________ >>> et-mgmt-tools mailing list >>> et-mgmt-tools at redhat.com >>> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From lfarkas at bppiac.hu Fri Jan 11 20:42:45 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Fri, 11 Jan 2008 21:42:45 +0100 Subject: [et-mgmt-tools] ANNOUNCE: virt-manager 0.5.3 and virtinst 0.300.2 releases In-Reply-To: <80d7e4090801110946h2310950bh8c3b160e0e9672be@mail.gmail.com> References: <20080111020510.GC26351@redhat.com> <80d7e4090801101928j13d2cebcob726055eafd3f828@mail.gmail.com> <20080111033742.GA25544@redhat.com> <80d7e4090801110946h2310950bh8c3b160e0e9672be@mail.gmail.com> Message-ID: <4787D4C5.1030608@bppiac.hu> Stephen John Smoogen wrote: > On Jan 10, 2008 8:37 PM, Daniel P. Berrange wrote: >> On Thu, Jan 10, 2008 at 08:28:52PM -0700, Stephen John Smoogen wrote: >>> On Jan 10, 2008 7:05 PM, Daniel P. Berrange wrote: >>>> I'm happy to announce the availability of virt-manager 0.5.3 and >>>> virtinst 0.300.2 bug fix releases on the download page: >>>> >>>> http://virt-manager.org/download.html >>>> >>>> The virt-manager-0.5.3 release: >>>> >>>> http://virt-manager.org/download/sources/virt-manager/virt-manager-0.5.3.tar.gz >>>> >>>> This is a bug fix release. The sizing of the VNC window is fixed for >>>> screens where the physical size is less than the guest screen size. >>>> The 'new vm' button is switched back to its old (more obvious style/ >>>> placement). Restore of VMs is working again for local connections. A >>>> menu for sending special key sequences to the guest is added. Lots >>>> of other misc bug fixes >>>> >>> What is the status of these working with RHEL-5? I am mostly trying to >>> gauge if these are for Fedora 9 or could be used with CentOS/RHEL? >> They won't work on RHEL-5.1 because they require a newer version of libvirt >> than is present, and remote management also requires newer support for VNC >> encryption in Xen. They should work on F-7 or later,and I will be pushing >> updates to Fedora soon >> > > Foo. Sigh.. ok thanks. the libvirt I could see updating.. but the Xen > changes would be a massive item. hmm, although i already recompile all the required packages and we use them since months without problems. the rpms can be found here: http://www.lfarkas.org/linux/packages/redhat/5/ -- Levente "Si vis pacem para bellum!" From jason at 3dogs.us Sat Jan 12 12:43:33 2008 From: jason at 3dogs.us (Jason Hartley) Date: Sat, 12 Jan 2008 07:43:33 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.6.5, Koan 0.6.4 In-Reply-To: <47868717.5040702@redhat.com> References: <47868717.5040702@redhat.com> Message-ID: <113193470801120443j2a2d7dc5medbc7bccedb82e06@mail.gmail.com> On 1/10/08, Michael DeHaan wrote: > New releases... > > Nothing major. This is mainly just to update the apps with a few needed > features while development on Cobbler 0.7.X continues. > > Cobbler 0.6.5 changes: > - Fix for f8 comps.xml name change > - Fix for repo file permissions during cobbler reposync > > Koan 0.6.4 has a few more features: > - Added --nogfx option when creating virtual guests to disable VNC and > just use the console > - Allow Xen machines to be constructed with no disks (size=0), but not > qemu/KVM > - Support Xen Fullvirt via PXE (requires Cobbler 0.7.X) > - Add new --autonet parameter (see manpage) > - Fail if kernel argument length exceeds the 255 cap, as truncation can > lead to hard to debug errors. > > http://cobbler.et.redhat.com/download/ > > Should be pushed to Fedora/EPEL shortly... > > --Michael > Michael- Thanks! The new koan option "--nogfx" works like a champ. Regards, Jason Hartley From mrpquter at yahoo.com Sat Jan 12 19:43:06 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Sat, 12 Jan 2008 11:43:06 -0800 (PST) Subject: [et-mgmt-tools] vnc option Message-ID: <703510.49583.qm@web54408.mail.yahoo.com> I'd like to be able to watch the installation display during the initial configuring / testing / debugging cycle of kickstarts and templates and whatnot. What is the recommended way of configuring vnc for this purpose? /var/lib/cobbler/settings has a kernel_options stanza which seems the logical place for it. Is the 'text' variable meant for arbitrary stuff to be passed to the kernel, so vnc options could be added there? Thanks, -Mike ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From berrange at redhat.com Sat Jan 12 21:12:47 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 12 Jan 2008 21:12:47 +0000 Subject: [et-mgmt-tools] vnc option In-Reply-To: <703510.49583.qm@web54408.mail.yahoo.com> References: <703510.49583.qm@web54408.mail.yahoo.com> Message-ID: <20080112211247.GB21560@redhat.com> On Sat, Jan 12, 2008 at 11:43:06AM -0800, Michael Harrison wrote: > I'd like to be able to watch the installation display during the > initial configuring / testing / debugging cycle of kickstarts and > templates and whatnot. What is the recommended way of configuring vnc > for this purpose? > > /var/lib/cobbler/settings has a kernel_options stanza which seems the > logical place for it. Is the 'text' variable meant for arbitrary stuff > to be passed to the kernel, so vnc options could be added there? You don't really want to use the guest OS installer's VNC options. All KVM or Xen guests have ability to setup a graphical display an expose that securely as a VNC server. So whether the guest is Linux, or Windows or something else you can access it via VNC. A regular VNC viewer is one option, but if you want secure access use simple 'virt-viewer' app or the full blown 'virt-manager' app. I can't remember what the Cobbler option for turning on VNC is, but I'm sure someone will point it out if its not in the man page... Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mrpquter at yahoo.com Sun Jan 13 01:49:23 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Sat, 12 Jan 2008 17:49:23 -0800 (PST) Subject: [et-mgmt-tools] vnc option In-Reply-To: <20080112211247.GB21560@redhat.com> Message-ID: <596658.15611.qm@web54404.mail.yahoo.com> I probably should have made it clear that I have in mind regular installs, not virtual machines. Thanks! -Mike --- "Daniel P. Berrange" wrote: > On Sat, Jan 12, 2008 at 11:43:06AM -0800, Michael Harrison wrote: > > I'd like to be able to watch the installation display during the > > initial configuring / testing / debugging cycle of kickstarts and > > templates and whatnot. What is the recommended way of configuring > vnc > > for this purpose? > > > > /var/lib/cobbler/settings has a kernel_options stanza which seems > the > > logical place for it. Is the 'text' variable meant for arbitrary > stuff > > to be passed to the kernel, so vnc options could be added there? > > You don't really want to use the guest OS installer's VNC options. > All > KVM or Xen guests have ability to setup a graphical display an expose > that securely as a VNC server. So whether the guest is Linux, or > Windows > or something else you can access it via VNC. A regular VNC viewer is > one option, but if you want secure access use simple 'virt-viewer' > app > or the full blown 'virt-manager' app. I can't remember what the > Cobbler > option for turning on VNC is, but I'm sure someone will point it out > if > its not in the man page... > > Dan. > -- > |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 > 2496 -=| > |=- Perl modules: http://search.cpan.org/~danberr/ > -=| > |=- Projects: http://freshmeat.net/~danielpb/ > -=| > |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B > 9505 -=| > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From mdehaan at redhat.com Sun Jan 13 15:11:18 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Sun, 13 Jan 2008 10:11:18 -0500 Subject: [et-mgmt-tools] vnc option In-Reply-To: <596658.15611.qm@web54404.mail.yahoo.com> References: <596658.15611.qm@web54404.mail.yahoo.com> Message-ID: <478A2A16.9060208@redhat.com> Michael Harrison wrote: > I probably should have made it clear that I have in mind regular > installs, not virtual machines. > Some VNC options are listed here: http://fedoraproject.org/wiki/Anaconda/Options From mrpquter at yahoo.com Mon Jan 14 17:02:55 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Mon, 14 Jan 2008 09:02:55 -0800 (PST) Subject: [et-mgmt-tools] Built-in variables Message-ID: <60277.83829.qm@web54402.mail.yahoo.com> I'm currently using some of the template variables that are shown as examples on the wiki https://fedorahosted.org/cobbler/wiki/KickstartTemplating The "Built In Variables" section recommends looking at the contents of any file in /var/lib/cobbler/profiles/ or /var/lib/cobbler/systems/. On my cobbler-0.6.5-3.el5.noarch install, profiles and systems are files, not directories, and so there are no samples to look at. Where may I find out what are the built-in cobbler variables for use with templates ? Thanks, -Mike ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From mdehaan at redhat.com Mon Jan 14 17:21:01 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 14 Jan 2008 12:21:01 -0500 Subject: [et-mgmt-tools] Built-in variables In-Reply-To: <60277.83829.qm@web54402.mail.yahoo.com> References: <60277.83829.qm@web54402.mail.yahoo.com> Message-ID: <478B99FD.7070409@redhat.com> Michael Harrison wrote: > I'm currently using some of the template variables that are shown as > examples on the wiki > https://fedorahosted.org/cobbler/wiki/KickstartTemplating > > The "Built In Variables" section recommends looking at the contents of > any file in /var/lib/cobbler/profiles/ or /var/lib/cobbler/systems/. > That should read /var/www/cobbler/profiles and /systems, so change the lib to a www. I'll fix the Wiki. > On my cobbler-0.6.5-3.el5.noarch install, profiles and systems are > files, not directories, and so there are no samples to look at. > > Where may I find out what are the built-in cobbler variables for use > with templates ? > > Thanks, > -Mike > > > > ____________________________________________________________________________________ > Be a better friend, newshound, and > know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From khelenek at hotmail.com Mon Jan 14 18:04:49 2008 From: khelenek at hotmail.com (Kris Helenek) Date: Mon, 14 Jan 2008 13:04:49 -0500 Subject: [et-mgmt-tools] Open and Details buttons do nothing Message-ID: Hi, I have installed FC8 and Xen, booted it with xen kernel, and added a 2nd domain so that I now have 2 domains running properly on my machine. I then open your tool (Virtual Machine Manager) and can accurately view a list of my domains and their basic info. But if I select one of my domains and click the Details or Open buttons (or select Edit->Details, or right click and select Details) nothing happens. I *can* open the Edit-Preferences, and Edit->Host Details, but just not the machine details. I know this isn't much to go on, but if this is known behavior or if there is advice on what I can do to debug what is going on it would be much appreciated. I checked the few logs I could think of on the machine but I have no idea what is going on. This is a fresh install today. Thanks, Kris ------------- # uname -a Linux xenqa.studentsonly.com 2.6.21-2952.fc8xen #1 SMP Mon Nov 19 07:06:36 EST 2007 x86_64 x86_64 x86_64 GNU/Linux From berrange at redhat.com Mon Jan 14 18:08:13 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 14 Jan 2008 18:08:13 +0000 Subject: [et-mgmt-tools] Open and Details buttons do nothing In-Reply-To: References: Message-ID: <20080114180813.GS920@redhat.com> On Mon, Jan 14, 2008 at 01:04:49PM -0500, Kris Helenek wrote: > Hi, > > I have installed FC8 and Xen, booted it with xen kernel, and added a 2nd > domain so that I now have 2 domains running properly on my machine. I then > open your tool (Virtual Machine Manager) and can accurately view a list of > my domains and their basic info. But if I select one of my domains and > click the Details or Open buttons (or select Edit->Details, or right click > and select Details) nothing happens. I *can* open the Edit-Preferences, > and Edit->Host Details, but just not the machine details. > > I know this isn't much to go on, but if this is known behavior or if there > is advice on what I can do to debug what is going on it would be much > appreciated. I checked the few logs I could think of on the machine but I > have no idea what is going on. This is a fresh install today. Install the latest 'python-virtinst' and 'virt-manager' from updates-testing F8 repository and this should probably be resolved. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From khelenek at hotmail.com Mon Jan 14 18:35:12 2008 From: khelenek at hotmail.com (Kris Helenek) Date: Mon, 14 Jan 2008 13:35:12 -0500 Subject: [et-mgmt-tools] Open and Details buttons do nothing References: Message-ID: Woops - probably the wrong place for this. Just found the Fedora bugzilla. I found a bug there that is probably related. ----- Original Message ----- From: "Kris Helenek" To: Sent: Monday, January 14, 2008 1:04 PM Subject: [et-mgmt-tools] Open and Details buttons do nothing > Hi, > > I have installed FC8 and Xen, booted it with xen kernel, and added a 2nd > domain so that I now have 2 domains running properly on my machine. I > then open your tool (Virtual Machine Manager) and can accurately view a > list of my domains and their basic info. But if I select one of my domains > and click the Details or Open buttons (or select Edit->Details, or right > click and select Details) nothing happens. I *can* open the > Edit-Preferences, and Edit->Host Details, but just not the machine > details. > > I know this isn't much to go on, but if this is known behavior or if there > is advice on what I can do to debug what is going on it would be much > appreciated. I checked the few logs I could think of on the machine but I > have no idea what is going on. This is a fresh install today. > > > > Thanks, > > Kris > > ------------- > > # uname -a > > Linux xenqa.studentsonly.com 2.6.21-2952.fc8xen #1 SMP Mon Nov 19 07:06:36 > EST 2007 x86_64 x86_64 x86_64 GNU/Linux > > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From astokes at redhat.com Mon Jan 14 20:57:30 2008 From: astokes at redhat.com (Adam Stokes) Date: Mon, 14 Jan 2008 15:57:30 -0500 Subject: [et-mgmt-tools] virt-manager 0.4.0 - paravirt xen guests Message-ID: <20080114205730.GH1467@battlemidget.rdu.redhat.com> Attempting to increase the max and current allocation for a non running guests fails to set both values. The failure occurs within this function on the conditional (virtManager/details.py->virtManager/domain.py) : def set_memory(self, memory): memory = int(memory) if (memory > self.maximum_memory()): logging.warning("Requested memory " + str(memory) + " over maximum " + str(self.maximum_memory())) memory = self.maximum_memory() self.vm.setMemory(memory) >From my investigation self.maximum_memory() seems to pull older information from the record list resulting in the above failure. Now the strange thing is once I set the max memory, click apply, then set current allocation the memory will set as expected. Is this expected behavior? If not, I thought perhaps pulling the information from the max memory spin button and testing against that as well would either fix or workaround this issue but I'm not sure that is the best approach. Thanks! -- Adam Stokes (astokes at redhat.com) From herlo1 at gmail.com Tue Jan 15 01:04:50 2008 From: herlo1 at gmail.com (Clint Savage) Date: Mon, 14 Jan 2008 18:04:50 -0700 Subject: [et-mgmt-tools] Cobbler issues on importing from a copied DVD Message-ID: Per a request from mpdehaan on #cobbler, I'm submitting a request here about an issue I'm currently running into: I've got a pastebin of my main problems. The problem is around lines 27-31 of the pastebin[1]. I was asked if I had the 0.6.5 version of cobbler and the latest (it seems) version of createrepo installed. What else could it be? I've spent a bit of time hunting on the lazyweb with no luck, so any help is greatly appreciated. 1 - http://cobbler.pastebin.com/m299040d6 Other useful output: # rpm -q cobbler cobbler-0.6.5-3.fc8 # rpm -q createrepo createrepo-0.4.11-1.fc8 # mount /dev/sdb1 on /media/distros type jfs (rw,nosuid,nodev,uhelper=hal) /media/distros/FEDORA/Fedora-8-i386-DVD.iso on /mnt type iso9660 (rw,loop=/dev/loop0) Thanks in advance. Cheers, Clint (aka herlo) From soren at ubuntu.com Tue Jan 15 13:20:32 2008 From: soren at ubuntu.com (Soren Hansen) Date: Tue, 15 Jan 2008 14:20:32 +0100 Subject: [et-mgmt-tools] [PATCH] Connect to remote consoles as correct user when using +ssh style url's Message-ID: <20080115132032.GW1707@atlas.linux2go.dk> Hi! virt-manager was hardcoded to always make ssh connections as root when connecting to remote consoles. This patch fixes it. -- Soren Hansen Ubuntu Server Team http://www.ubuntu.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: ssh_tunnel_username.patch Type: text/x-diff Size: 4725 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From berrange at redhat.com Tue Jan 15 14:03:18 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 15 Jan 2008 14:03:18 +0000 Subject: [et-mgmt-tools] [PATCH] Connect to remote consoles as correct user when using +ssh style url's In-Reply-To: <20080115132032.GW1707@atlas.linux2go.dk> References: <20080115132032.GW1707@atlas.linux2go.dk> Message-ID: <20080115140317.GC17783@redhat.com> On Tue, Jan 15, 2008 at 02:20:32PM +0100, Soren Hansen wrote: > Hi! > > virt-manager was hardcoded to always make ssh connections as root when > connecting to remote consoles. This patch fixes it. Thanks, applied to HG. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Tue Jan 15 14:10:09 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 15 Jan 2008 14:10:09 +0000 Subject: [et-mgmt-tools] virt-manager: handling of bonding and bridge devices In-Reply-To: <20080110163307.GA1133@bogon.ms20.nix> References: <20071218153414.GA16706@bogon.ms20.nix> <20080110163307.GA1133@bogon.ms20.nix> Message-ID: <20080115141009.GE17783@redhat.com> On Thu, Jan 10, 2008 at 05:33:07PM +0100, Guido Guenther wrote: > On Tue, Dec 18, 2007 at 04:34:14PM +0100, Guido Guenther wrote: > > Hi, > > currently virt-manager uses some linux specific heuristics to find out > > if a device is part of a bridge. This immediatley fails if the > > underlying device is not a physical device but e.g. a bonding device. > > > > Hal doesn't seem to care about either bridges nor bonding devices. I > > wonder where to fix this? Add another linux specific bonding hack > > to determine if a eth devices is part of a bonding device which might be > > part of a bridge? Would be easy enough to do but I wonder if it wouldn't > > be better if hal would export this information in the first place? > Hmm...no comments - would another hack that treats bridges on top of > bond devices be o.k.? The way virt-manager enumerates network devices only deals with physical NICs which are optionally part of a bridge. It is possible to extend this to deal with NICs that have then had a number of VLANs setup, or are part of a bonded pair, but I don't have any machines with such setups myself so not in a situation to write & test any code. If someone wishes to add support for this, the place to look is the file src/virtManager/connection.py. In particular the _device_added method is called once for each physical NIC on the machine. You'd have to extend that to also look for associated bonds, vlans. So patches welcomed... Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From crobinso at redhat.com Tue Jan 15 15:33:27 2008 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 15 Jan 2008 10:33:27 -0500 Subject: [et-mgmt-tools] virt-manager 0.4.0 - paravirt xen guests In-Reply-To: <20080114205730.GH1467@battlemidget.rdu.redhat.com> References: <20080114205730.GH1467@battlemidget.rdu.redhat.com> Message-ID: <478CD247.4060009@redhat.com> Adam Stokes wrote: > Attempting to increase the max and current allocation for a non running > guests fails to set both values. > > The failure occurs within this function on the conditional > (virtManager/details.py->virtManager/domain.py) : > > def set_memory(self, memory): > memory = int(memory) > if (memory > self.maximum_memory()): > logging.warning("Requested memory " + str(memory) + " over > maximum " + str(self.maximum_memory())) > memory = self.maximum_memory() > self.vm.setMemory(memory) > >>From my investigation self.maximum_memory() seems to pull older > information from the record list resulting in the above failure. > > Now the strange thing is once I set the max memory, click apply, then > set current allocation the memory will set as expected. > > Is this expected behavior? If not, I thought perhaps pulling the > information from the max memory spin button and testing against that as > well would either fix or workaround this issue but I'm not sure that is > the best approach. > > Thanks! > This definitely is not expected behavior, and is still a bug in current upstream. Would you mind submitting a bug against rawhide for this? Thanks, Cole From mdehaan at redhat.com Tue Jan 15 18:58:04 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 15 Jan 2008 13:58:04 -0500 Subject: [et-mgmt-tools] Cobbler issues on importing from a copied DVD In-Reply-To: References: Message-ID: <478D023C.8090707@redhat.com> Clint Savage wrote: > Per a request from mpdehaan on #cobbler, I'm submitting a request here > about an issue I'm currently running into: > > I've got a pastebin of my main problems. The problem is around lines > 27-31 of the pastebin[1]. > > I was asked if I had the 0.6.5 version of cobbler and the latest (it > seems) version of createrepo installed. > > What else could it be? I've spent a bit of time hunting on the > lazyweb with no luck, so any help is greatly appreciated. > > 1 - http://cobbler.pastebin.com/m299040d6 > From talking on IRC, (repeating what I just said on IRC now), it looks like ks_mirror/config was deleted, so recreating it should fix that problem. My fault for suggesting you clean out ks_mirror/*, when I should have said, clean out ks_mirror/* and leave config :) --Michael From mrpquter at yahoo.com Tue Jan 15 20:54:17 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Tue, 15 Jan 2008 12:54:17 -0800 (PST) Subject: [et-mgmt-tools] koan on RHEL3 Message-ID: <190384.63402.qm@web54406.mail.yahoo.com> I installed koan-0.6.4-1.el4.noarch.rpm from EPEL on an old RHEL3 box to rebuild it. I seem to have read that koan runs on RHEL3. Here's what I got: koan --server=server.example.com --list-profiles Traceback (most recent call last): File "/usr/bin/koan", line 18, in ? import koan.app as app ImportError: No module named koan.app Any ideas? Meanwhile I'll try rebuilding the SRPM on the RHEL3 box and see if that fixes it. -Mike ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From mdehaan at redhat.com Tue Jan 15 20:55:06 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 15 Jan 2008 15:55:06 -0500 Subject: [et-mgmt-tools] koan on RHEL3 In-Reply-To: <190384.63402.qm@web54406.mail.yahoo.com> References: <190384.63402.qm@web54406.mail.yahoo.com> Message-ID: <478D1DAA.2010904@redhat.com> Michael Harrison wrote: > I installed koan-0.6.4-1.el4.noarch.rpm from EPEL on an old RHEL3 box > to rebuild it. I seem to have read that koan runs on RHEL3. Here's what > I got: > > koan --server=server.example.com --list-profiles > Traceback (most recent call last): > File "/usr/bin/koan", line 18, in ? > import koan.app as app > ImportError: No module named koan.app > > Any ideas? Meanwhile I'll try rebuilding the SRPM on the RHEL3 box and > see if that fixes it. > > -Mike > Ah, yes. You will want to do that. There is no EPEL 3 build system, so you have to rpmbuild --rebuild the source RPM. Should that not work, let me know. --Michael From mrpquter at yahoo.com Tue Jan 15 21:04:32 2008 From: mrpquter at yahoo.com (Michael Harrison) Date: Tue, 15 Jan 2008 13:04:32 -0800 (PST) Subject: [et-mgmt-tools] koan on RHEL3 In-Reply-To: <478D1DAA.2010904@redhat.com> Message-ID: <295790.42061.qm@web54409.mail.yahoo.com> Yep, that worked. Thank you! -Mike --- Michael DeHaan wrote: > Michael Harrison wrote: > > I installed koan-0.6.4-1.el4.noarch.rpm from EPEL on an old RHEL3 > box > > to rebuild it. I seem to have read that koan runs on RHEL3. Here's > what > > I got: > > > > koan --server=server.example.com --list-profiles > > Traceback (most recent call last): > > File "/usr/bin/koan", line 18, in ? > > import koan.app as app > > ImportError: No module named koan.app > > > > Any ideas? Meanwhile I'll try rebuilding the SRPM on the RHEL3 box > and > > see if that fixes it. > > > > -Mike > > > Ah, yes. You will want to do that. There is no EPEL 3 build > system, > so you have to rpmbuild --rebuild the source RPM. > > Should that not work, let me know. > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From soren at ubuntu.com Wed Jan 16 09:31:27 2008 From: soren at ubuntu.com (Soren Hansen) Date: Wed, 16 Jan 2008 10:31:27 +0100 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types Message-ID: <20080116093126.GE1707@atlas.linux2go.dk> It's perfectly reasonable for privileged users to want to use usermode networking, but currently, there's no way for them to do that. The attached patch fixes this. -- Soren Hansen Ubuntu Server Team http://www.ubuntu.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: add_usermode_network.patch Type: text/x-diff Size: 22203 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From berrange at redhat.com Wed Jan 16 12:45:36 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 16 Jan 2008 12:45:36 +0000 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types In-Reply-To: <20080116093126.GE1707@atlas.linux2go.dk> References: <20080116093126.GE1707@atlas.linux2go.dk> Message-ID: <20080116124536.GC16624@redhat.com> On Wed, Jan 16, 2008 at 10:31:27AM +0100, Soren Hansen wrote: > It's perfectly reasonable for privileged users to want to use usermode > networking, but currently, there's no way for them to do that. The > attached patch fixes this. NACK, it offers nothing of benefit over the existing virtual networking which already provides a NAT based isolated network. Usermode networking is a crude hack which is frankly utter rubbish - the only option you have if you're an unprivileged user, but no privleged user should be using it. Usermode networking is also not 64-bit cleanup & will often corrupt / drop traffic. Just say no if you're privileged. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From crobinso at redhat.com Wed Jan 16 16:43:23 2008 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 16 Jan 2008 11:43:23 -0500 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types In-Reply-To: <20080116124536.GC16624@redhat.com> References: <20080116093126.GE1707@atlas.linux2go.dk> <20080116124536.GC16624@redhat.com> Message-ID: <478E342B.5070209@redhat.com> Daniel P. Berrange wrote: > On Wed, Jan 16, 2008 at 10:31:27AM +0100, Soren Hansen wrote: >> It's perfectly reasonable for privileged users to want to use usermode >> networking, but currently, there's no way for them to do that. The >> attached patch fixes this. > > NACK, it offers nothing of benefit over the existing virtual networking > which already provides a NAT based isolated network. Usermode networking > is a crude hack which is frankly utter rubbish - the only option you > have if you're an unprivileged user, but no privleged user should be > using it. Usermode networking is also not 64-bit cleanup & will often > corrupt / drop traffic. Just say no if you're privileged. > > Dan. Regardless of the NACK, there are still some issues we may want to consider. By not exposing usermode as an option and just setting it up by default, users can't specify a mac address for the interface, and if they remove it from the guest, there is no option to add it back. The latter could be hacked up easily I suppose, so there really aren't dealbreakers, just things to consider. - Cole From berrange at redhat.com Wed Jan 16 16:50:15 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 16 Jan 2008 16:50:15 +0000 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types In-Reply-To: <478E342B.5070209@redhat.com> References: <20080116093126.GE1707@atlas.linux2go.dk> <20080116124536.GC16624@redhat.com> <478E342B.5070209@redhat.com> Message-ID: <20080116165015.GB27604@redhat.com> On Wed, Jan 16, 2008 at 11:43:23AM -0500, Cole Robinson wrote: > Daniel P. Berrange wrote: > > On Wed, Jan 16, 2008 at 10:31:27AM +0100, Soren Hansen wrote: > >> It's perfectly reasonable for privileged users to want to use usermode > >> networking, but currently, there's no way for them to do that. The > >> attached patch fixes this. > > > > NACK, it offers nothing of benefit over the existing virtual networking > > which already provides a NAT based isolated network. Usermode networking > > is a crude hack which is frankly utter rubbish - the only option you > > have if you're an unprivileged user, but no privleged user should be > > using it. Usermode networking is also not 64-bit cleanup & will often > > corrupt / drop traffic. Just say no if you're privileged. > > > > Dan. > > Regardless of the NACK, there are still some issues we may want to consider. > By not exposing usermode as an option and just setting it up by default, users > can't specify a mac address for the interface, and if they remove it from the > guest, there is no option to add it back. The latter could be hacked up easily > I suppose, so there really aren't dealbreakers, just things to consider. There is no point choosing a MAC address if you're using user mode network because you don't have any control over usermode network stack. It always give you the same IP addr, all outgoing traffic gets sent out with the MAC of the host because of the way Slirp works. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From soren at ubuntu.com Wed Jan 16 17:18:56 2008 From: soren at ubuntu.com (Soren Hansen) Date: Wed, 16 Jan 2008 18:18:56 +0100 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types In-Reply-To: <20080116165015.GB27604@redhat.com> References: <20080116093126.GE1707@atlas.linux2go.dk> <20080116124536.GC16624@redhat.com> <478E342B.5070209@redhat.com> <20080116165015.GB27604@redhat.com> Message-ID: <20080116171856.GG1707@atlas.linux2go.dk> On Wed, Jan 16, 2008 at 04:50:15PM +0000, Daniel P. Berrange wrote: > There is no point choosing a MAC address if you're using user mode > network because you don't have any control over usermode network > stack. Not true. At least some OS's (Ubuntu to name one) assigns persistent names to NIC's based on their MAC. It might be convenient for you to be able to switch between virtual, bridged and usermode networking, but keep the same MAC (and hence interface name). -- Soren Hansen Ubuntu Server Team http://www.ubuntu.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From berrange at redhat.com Wed Jan 16 17:59:00 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 16 Jan 2008 17:59:00 +0000 Subject: [et-mgmt-tools] [PATCH] Add usermode networking to the list of available network types In-Reply-To: <20080116171856.GG1707@atlas.linux2go.dk> References: <20080116093126.GE1707@atlas.linux2go.dk> <20080116124536.GC16624@redhat.com> <478E342B.5070209@redhat.com> <20080116165015.GB27604@redhat.com> <20080116171856.GG1707@atlas.linux2go.dk> Message-ID: <20080116175900.GA15623@redhat.com> On Wed, Jan 16, 2008 at 06:18:56PM +0100, Soren Hansen wrote: > On Wed, Jan 16, 2008 at 04:50:15PM +0000, Daniel P. Berrange wrote: > > There is no point choosing a MAC address if you're using user mode > > network because you don't have any control over usermode network > > stack. > > Not true. At least some OS's (Ubuntu to name one) assigns persistent > names to NIC's based on their MAC. It might be convenient for you to be > able to switch between virtual, bridged and usermode networking, but > keep the same MAC (and hence interface name). Again this goes back to the fact that switching between virtual & user mode networking is nonsensical. If you have virtual networking or bridged available, user networking is utterly pointless. If you don't have virtual/bridged available and user networking is your only option then there is no need for persistent MAC because you've got nothing to switch between. The ideal long term plan is to make an implementation of the virtual networking APIs which can be used by unprivileged users too, thus killing any need to use user mode networking at all. User mode networking is a dead end and I've no intention to expose dead end functionality in the UI. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From sakaia at jp.fujitsu.com Thu Jan 17 10:09:19 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Thu, 17 Jan 2008 19:09:19 +0900 Subject: [et-mgmt-tools] Question about vmm-modify-file-storage.glade Message-ID: <200801171009.m0HA9Wio014339@fjmscan501.ms.jp.fujitsu.com> Hi, I have a question about "vmm-modify-file-storage.glade". The file does not seem to use in virt-manager. Is there any plan to expand to use this file? Thanks Atsushi SAKAI From crobinso at redhat.com Thu Jan 17 14:47:45 2008 From: crobinso at redhat.com (Cole Robinson) Date: Thu, 17 Jan 2008 09:47:45 -0500 Subject: [et-mgmt-tools] Question about vmm-modify-file-storage.glade In-Reply-To: <200801171009.m0HA9Wio014339@fjmscan501.ms.jp.fujitsu.com> References: <200801171009.m0HA9Wio014339@fjmscan501.ms.jp.fujitsu.com> Message-ID: <478F6A91.2060306@redhat.com> Atsushi SAKAI wrote: > Hi, > > I have a question about "vmm-modify-file-storage.glade". > The file does not seem to use in virt-manager. > Is there any plan to expand to use this file? > > Thanks > Atsushi SAKAI > > > > Yeah I noticed that a while ago as well but never looked into it. Turns out it was created during some refactoring but doesn't seem to have been used since. I'll remove it. Thanks, Cole From lionel.kernux at gmail.com Mon Jan 21 19:18:09 2008 From: lionel.kernux at gmail.com (Lionel Kernux) Date: Mon, 21 Jan 2008 14:18:09 -0500 Subject: [et-mgmt-tools] Why doesn't my VM show up in virt-manager? Message-ID: Hi, I'm running RHEL 5.1 and created a xen guest (RHEL4) manually by detting everything up in a .hvm file. I have started and stopped thge VM via xm create and it runs just fine. What steps are needed to allow it to be managed from the virt-manager gui? It doesn't show up in the list in the gui. Only VM's created from within the gui seem to show. I undersand that virt-manager uses xml files and a database to store VM info. Is there soem way to "import" a .hvm file into virt-manager? Any Help ?? Thanks Miguel From mdehaan at redhat.com Mon Jan 21 19:16:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 21 Jan 2008 14:16:59 -0500 Subject: [et-mgmt-tools] Why doesn't my VM show up in virt-manager? In-Reply-To: References: Message-ID: <4794EFAB.6010800@redhat.com> Lionel Kernux wrote: > Hi, > > I'm running RHEL 5.1 and created a xen guest (RHEL4) manually by > detting everything up in a .hvm file. I have started and stopped thge > VM via xm create and it runs just fine. I'm sure someone else could provide specifics on how to make it visible, though I should point out that things created by things using the virt-install library (like virt-install (the CLI tool), koan (works with cobbler), and of course virt-manager) show up automatically. If you can use those, they are the easiest way to go. --Michael From berrange at redhat.com Mon Jan 21 19:32:19 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 21 Jan 2008 19:32:19 +0000 Subject: [et-mgmt-tools] Why doesn't my VM show up in virt-manager? In-Reply-To: References: Message-ID: <20080121193219.GI12259@redhat.com> On Mon, Jan 21, 2008 at 02:18:09PM -0500, Lionel Kernux wrote: > Hi, > > I'm running RHEL 5.1 and created a xen guest (RHEL4) manually by > detting everything up in a .hvm file. I have started and stopped thge > VM via xm create and it runs just fine. What steps are needed to allow > it to be managed from the virt-manager gui? It doesn't show up in the > list in the gui. Only VM's created from within the gui seem to show. I > undersand that virt-manager uses xml files and a database to store VM > info. Is there soem way to "import" a .hvm file into virt-manager? I've no idea what a '.hvm' file is. In RHEL-5, virt-manager provides access to inactive guests defined in /etc/xen. The file names in that directory are required to match the name of the guest - ie if your guest is called 'demo', the file must be /etc/xen/demo. Any other filename is unsupported. Do not add file extensions. Do not add other non-guest config files in that directory Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mnielsen at redhat.com Mon Jan 21 19:41:45 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Mon, 21 Jan 2008 14:41:45 -0500 Subject: [et-mgmt-tools] Why doesn't my VM show up in virt-manager? In-Reply-To: References: Message-ID: <4794F579.4040301@redhat.com> Lionel Kernux wrote: > Hi, > > I'm running RHEL 5.1 and created a xen guest (RHEL4) manually by > detting everything up in a .hvm file. in addition to what Dan's response was, I would be concerned by your manual editing. Did you change the "name" and (more importantly) "UUID" lines? If you have the same UUID for 2 VMs, only 1 shows in virt-manager. There is a useful command "virsh domuuid " to get a proper UUID. > I have started and stopped thge > VM via xm create and it runs just fine. What steps are needed to allow > it to be managed from the virt-manager gui? It doesn't show up in the > list in the gui. Only VM's created from within the gui seem to show. I > undersand that virt-manager uses xml files and a database to store VM > info. Is there soem way to "import" a .hvm file into virt-manager? > > Any Help ?? > > Thanks > > Miguel > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -------------- next part -------------- A non-text attachment was scrubbed... Name: mnielsen.vcf Type: text/x-vcard Size: 177 bytes Desc: not available URL: From lionel.kernux at gmail.com Mon Jan 21 20:21:16 2008 From: lionel.kernux at gmail.com (Lionel Kernux) Date: Mon, 21 Jan 2008 15:21:16 -0500 Subject: [et-mgmt-tools] Why doesn't my VM show up in virt-manager? In-Reply-To: <20080121193219.GI12259@redhat.com> References: <20080121193219.GI12259@redhat.com> Message-ID: Hi, Thanks for the replies. I attempted to use the virt-manager gui to create a new guest. The process created the img file the spun its wheels for the rest of the night. I had to manually kill the virt-manager. I poked around and found that virt-manager had finished creating the .img file but had not created the /etc/xen/vigilance config file. So I then copied the "/etc/xen/xmexample.hvm" that shipped with the xen rpm and modified it to meet my needs and pointed it to the vigilance.img file that virt-manager created. I set the boot parameter in the config to boot from CD and used xm create /etc/xen/vigilance to start it up. Everything went fine and I installed the guest OS and can run it fine using xm. Why can't virt-manager see it? Thanks Miguel On Jan 21, 2008 2:32 PM, Daniel P. Berrange wrote: > > On Mon, Jan 21, 2008 at 02:18:09PM -0500, Lionel Kernux wrote: > > Hi, > > > > I'm running RHEL 5.1 and created a xen guest (RHEL4) manually by > > detting everything up in a .hvm file. I have started and stopped thge > > VM via xm create and it runs just fine. What steps are needed to allow > > it to be managed from the virt-manager gui? It doesn't show up in the > > list in the gui. Only VM's created from within the gui seem to show. I > > undersand that virt-manager uses xml files and a database to store VM > > info. Is there soem way to "import" a .hvm file into virt-manager? > > I've no idea what a '.hvm' file is. In RHEL-5, virt-manager provides access > to inactive guests defined in /etc/xen. The file names in that directory > are required to match the name of the guest - ie if your guest is called > 'demo', the file must be /etc/xen/demo. Any other filename is unsupported. > Do not add file extensions. Do not add other non-guest config files in > that directory > > Dan. > -- > |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| > |=- Perl modules: http://search.cpan.org/~danberr/ -=| > |=- Projects: http://freshmeat.net/~danielpb/ -=| > |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From caszt_j at gmx.at Tue Jan 22 16:56:06 2008 From: caszt_j at gmx.at (caszt_j at gmx.at) Date: Tue, 22 Jan 2008 17:56:06 +0100 Subject: [et-mgmt-tools] mounting imagefile created with virtual machine manager 0.3.1 Message-ID: <20080122165606.90740@gmx.net> Hello! Is it possible to mount an image file, which was created with the virtual machine manager 0.3.1? Which filesystem type have i to choose? I'll be very happy about some hints! Thanks! Caszt! -- Psssst! Schon vom neuen GMX MultiMessenger geh?rt? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger From berrange at redhat.com Tue Jan 22 17:12:14 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 22 Jan 2008 17:12:14 +0000 Subject: [et-mgmt-tools] mounting imagefile created with virtual machine manager 0.3.1 In-Reply-To: <20080122165606.90740@gmx.net> References: <20080122165606.90740@gmx.net> Message-ID: <20080122171214.GK29260@redhat.com> On Tue, Jan 22, 2008 at 05:56:06PM +0100, caszt_j at gmx.at wrote: > Hello! > > Is it possible to mount an image file, which was created with the virtual machine manager 0.3.1? > Which filesystem type have i to choose? The image is a full disk image, not a filesystem. It has many partitions and LVMs volumes within it, so you can't mount it in the host directly. See 'Accessing data on a guest image' in this wiki page for instructions on how to get access to it http://fedoraproject.org/wiki/Docs/Fedora8VirtQuickStart Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From jim at rossberry.com Wed Jan 23 02:13:42 2008 From: jim at rossberry.com (Jim Wildman) Date: Tue, 22 Jan 2008 21:13:42 -0500 (EST) Subject: [et-mgmt-tools] Built-in variables In-Reply-To: <478B99FD.7070409@redhat.com> References: <60277.83829.qm@web54402.mail.yahoo.com> <478B99FD.7070409@redhat.com> Message-ID: On Mon, 14 Jan 2008, Michael DeHaan wrote: > That should read /var/www/cobbler/profiles and /systems, so change the lib to > a www. > > I'll fix the Wiki. > > Here's another typo.. "which actually like in /var/www/cobbler" ^^^^ live ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim at rossberry.com http://www.rossberry.com "Society in every state is a blessing, but Government, even in its best state, is a necessary evil; in its worst state, an intolerable one." Thomas Paine From caszt_j at gmx.at Wed Jan 23 20:24:57 2008 From: caszt_j at gmx.at (caszt) Date: Wed, 23 Jan 2008 21:24:57 +0100 Subject: [et-mgmt-tools] mounting imagefile created with virtual machine manager 0.3.1 In-Reply-To: <20080122171214.GK29260@redhat.com> References: <20080122165606.90740@gmx.net> <20080122171214.GK29260@redhat.com> Message-ID: <20080123202457.311280@gmx.net> Hello Dan! Thank you very much for your help! Caszt -------- Original-Nachricht -------- > Datum: Tue, 22 Jan 2008 17:12:14 +0000 > Von: "Daniel P. Berrange" > An: Fedora/Linux Management Tools > Betreff: Re: [et-mgmt-tools] mounting imagefile created with virtual machine manager 0.3.1 > On Tue, Jan 22, 2008 at 05:56:06PM +0100, caszt_j at gmx.at wrote: > > Hello! > > > > Is it possible to mount an image file, which was created with the > virtual machine manager 0.3.1? > > Which filesystem type have i to choose? > > The image is a full disk image, not a filesystem. It has many partitions > and LVMs volumes within it, so you can't mount it in the host directly. > > See 'Accessing data on a guest image' in this wiki page for instructions > on how to get access to it > > http://fedoraproject.org/wiki/Docs/Fedora8VirtQuickStart > > Regards, > Dan. > -- > |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 > -=| > |=- Perl modules: http://search.cpan.org/~danberr/ > -=| > |=- Projects: http://freshmeat.net/~danielpb/ > -=| > |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 > -=| > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail From shonorio at gmail.com Wed Jan 23 21:24:56 2008 From: shonorio at gmail.com (Solli Honorio) Date: Wed, 23 Jan 2008 19:24:56 -0200 Subject: [et-mgmt-tools] Getting error when i trying to connect on xen Message-ID: <12d1f4ee0801231324s58e4cfa7wb79a4b4d47609ae3@mail.gmail.com> Hi, I'm sure if here is the right place to ask this but let me go ... I'm running a Xen server (version 3.1 compiled) on Linux Debian Lenny that I'd install virt-manager by package of lenny repository. I've installed 'libvirt-bin, libvirt0 and python-libvirt' version 0.4.0-2. At '/etc/xen/xend-config.sxp' I had : #(xend-http-server no) (xend-unix-server yes) #(xend-tcp-xmlrpc-server no) (xend-unix-xmlrpc-server yes) #(xend-relocation-server no) (xend-relocation-server yes) When I try to connect to xen server, on localhost, I get back the error bellow. My question is if I have to enable something on xen server to get this working. Unable to open connection to hypervisor URI 'xen:///': libvirt.libvirtError virConnectOpenReadOnly() failed could not connect to xen:/// Traceback (most recent call last): File "/usr/local/share/virt-manager/virtManager/connection.py", line 305, in _open_thread self.vmm = libvirt.openReadOnly(self.uri) File "/usr/local/lib/python2.4/site-packages/libvirt.py", line 132, in openReadOnly if ret is None:raise libvirtError('virConnectOpenReadOnly() failed') libvirtError: virConnectOpenReadOnly() failed could not connect to xen:/// Looking at /root/.virt-manager/virt-manager.log I get these mensagens: [Wed, 23 Jan 2008 19:06:50 virt-manager 5070] INFO (virt-manager:126) Application startup [Wed, 23 Jan 2008 19:06:52 virt-manager 5071] DEBUG (engine:74) About to connect to uris ['xen:///'] [Wed, 23 Jan 2008 19:06:52 virt-manager 5071] WARNING (virt-manager:282) Could not get connection to session bus, disabling DBus service dbus.exceptions.DBusException org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken. [Wed, 23 Jan 2008 19:07:00 virt-manager 5071] DEBUG (connection:292) Scheduling background open thread for xen:/// [Wed, 23 Jan 2008 19:07:00 virt-manager 5071] DEBUG (connection:299) Background thread is running [Wed, 23 Jan 2008 19:07:00 virt-manager 5071] ERROR (connection:324) Unable to open connection to hypervisor URI 'xen:///': libvirt.libvirtError virConnectOpenReadOnly() failed could not connect to xen:/// Traceback (most recent call last): File "/usr/local/share/virt-manager/virtManager/connection.py", line 306, in _open_thread self.vmm = libvirt.openReadOnly(self.uri) File "/usr/local/lib/python2.4/site-packages/libvirt.py", line 132, in openReadOnly if ret is None:raise libvirtError('virConnectOpenReadOnly() failed') libvirtError: virConnectOpenReadOnly() failed could not connect to xen:/// [Wed, 23 Jan 2008 19:07:00 virt-manager 5071] DEBUG (connection:328) Background open thread complete, scheduling notify [Wed, 23 Jan 2008 19:07:00 virt-manager 5071] DEBUG (connection:337) Notifying open result -- "o animal satisfeito dorme". - Guimar?es Rosa -------------- next part -------------- An HTML attachment was scrubbed... URL: From rainer at ultra-secure.de Fri Jan 25 12:58:45 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 25 Jan 2008 13:58:45 +0100 Subject: [et-mgmt-tools] Is there a way to manipulate the "repo"-directives on/after install? Message-ID: <4799DD05.70101@ultra-secure.de> Hi, When installing RHEL5 via cobbler, I do this via a special VLAN with IPs in the RFC1918-space. So, after activating "yum_post_install_mirror: 1", it adds the location of the files in /etc/yum.repos.d. However, later on, this VLAN is most likely no longer present and configured. All the files are available via a "public" IP, though, that is reachable by most systems after the network has been brought up. Should I just go and add a %post section in the template that seds over the files in /etc/yum.repos.d/ and replace the IPs? Or what would be the recommended solution? Thanks in advance. cheers, Rainer From rjones at redhat.com Fri Jan 25 15:29:44 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 25 Jan 2008 15:29:44 +0000 Subject: [et-mgmt-tools] virt-viewer as a browser plugin (patch for discussion only) Message-ID: <479A0068.5060805@redhat.com> I've been trying to get virt-viewer to run as a browser plugin, based on similar work for building a Gtk-VNC plugin. Attached is a patch for discussion. It doesn't work, but I'm not entirely sure why. The earlier Gtk-VNC plugin works, and this code does basically the same thing but doesn't work. It connects to the VNC server, gets framebuffer updates and everything, it just doesn't draw anything -- I'm sure it's something very simple. Anyway ... The idea is that you can put the following code into your web page, and that will display the corresponding console: The full parameters are uri & name (required), waitvnc & direct (optional). Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-viewer-no-worky.patch Type: text/x-patch Size: 51518 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From berrange at redhat.com Fri Jan 25 15:43:57 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 25 Jan 2008 15:43:57 +0000 Subject: [et-mgmt-tools] virt-viewer as a browser plugin (patch for discussion only) In-Reply-To: <479A0068.5060805@redhat.com> References: <479A0068.5060805@redhat.com> Message-ID: <20080125154357.GE7237@redhat.com> On Fri, Jan 25, 2008 at 03:29:44PM +0000, Richard W.M. Jones wrote: > I've been trying to get virt-viewer to run as a browser plugin, based on > similar work for building a Gtk-VNC plugin. > > Attached is a patch for discussion. It doesn't work, but I'm not > entirely sure why. The earlier Gtk-VNC plugin works, and this code does > basically the same thing but doesn't work. It connects to the VNC > server, gets framebuffer updates and everything, it just doesn't draw > anything -- I'm sure it's something very simple. Hmm, I don't see a call to gtk_widget_show_all() there - only a realize() call. It could be that a regular window is visible by default, while a plug needs to be explicitly shown ? Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From rjones at redhat.com Fri Jan 25 16:12:20 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 25 Jan 2008 16:12:20 +0000 Subject: [et-mgmt-tools] [PATCH 1/2] Virt-viewer: Abstract viewer_start entrypoint Message-ID: <479A0A64.7060801@redhat.com> This patch adjusts src/main.c so that there is an entrypoint called viewer_start which will be used by the plugin. Note the header file ("viewer.h") and the function is private and internal to virt-viewer. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-viewer-1.patch Type: text/x-patch Size: 10879 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From mdehaan at redhat.com Fri Jan 25 16:19:14 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 25 Jan 2008 11:19:14 -0500 Subject: [et-mgmt-tools] Is there a way to manipulate the "repo"-directives on/after install? In-Reply-To: <4799DD05.70101@ultra-secure.de> References: <4799DD05.70101@ultra-secure.de> Message-ID: <479A0C02.1070108@redhat.com> Rainer Duffner wrote: > Hi, > > When installing RHEL5 via cobbler, I do this via a special VLAN with > IPs in the RFC1918-space. > So, after activating "yum_post_install_mirror: 1", it adds the location > of the files in /etc/yum.repos.d. > However, later on, this VLAN is most likely no longer present and > configured. > All the files are available via a "public" IP, though, that is reachable > by most systems after the network has been brought up. > > Should I just go and add a %post section in the template that seds over > the files in /etc/yum.repos.d/ and replace the IPs? > Or what would be the recommended solution? > There should be easier ways, but your configuration might not allow them. Details: If you know where the systems are going to finally reside, recent cobbler has a parameter on each profile and/or system called --server-override which will override all IP's/server addresses in settings for anything using that profile/system record. This was originally added in 0.6.3 and is probably more thoroughly implemented in 0.6.5. If you have multiple profiles for multiple locations, you could perhaps use inherited profiles to make a version that had a different server parameter, like this: cobbler profile add --name=foo ... cobbler profile add --name=datacenter2-foo --parent=foo --server-override=cobbler2.example.com Does that make sense? The problem however is if they are provisioned on one VLAN and then moved to another, in which case you need to turn off yum_post_install_mirror altogether and just use something like a config management system to push out the repo definition files *OR* just add some special stuff in %post of your kickstart to create those files in /etc/yum.repos.d ... your sed trick there would probably work fine. --Michael From rjones at redhat.com Fri Jan 25 16:22:21 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 25 Jan 2008 16:22:21 +0000 Subject: [et-mgmt-tools] [PATCH 2/2] Virt-viewer browser plugin Message-ID: <479A0CBD.9000508@redhat.com> This adds a virt-viewer browser plugin which works on Firefox / Mozilla and probably other NPAPI-based browsers on Unix-like systems. The plugin is placed in the plugin/ subdirectory. Libtool is required (since the plugin is a shared library). The plugin is disabled by default. To enable it you have to do './configure --enable-plugin'. To use the plugin, place code such as this in a web page: The width & height are required, and should be at least as large as the virtual machine's console. (Because Gecko + Gtk plugins are really buggy we are unable to place scrollbars or a menubar around the VNC widget). 'uri' (libvirt URI) and 'name' (domain name/ID/UUID) are required. Optionally you can also have waitvnc="1" and direct="1" which have the same effect as the equivalent virt-viewer command line options. Screenshot: http://annexia.org/tmp/Screenshot-SeaMonkey.png Known bugs, at least 4 at the moment: (1) There's a strange race condition where Gecko renders the plugin before rendering the surrounding page. This results in a segfault because the browser sends uninitialized private data to the plugin. (2) The plugin cannot grab the mouse, so mouse handling is erratic. (3) Virt-viewer menubar doesn't render correctly, so in this version it is disabled. Appears to be a general problem with Gecko + Gtk. (4) There are various rendering artifacts on framebuffer updates, but they only affect movement/animation, eg. dragging windows. When the animation has finished the rendering artifacts disappear. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-viewer-2.patch Type: text/x-patch Size: 39807 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From berrange at redhat.com Fri Jan 25 21:30:49 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 25 Jan 2008 21:30:49 +0000 Subject: [et-mgmt-tools] [PATCH 1/2] Virt-viewer: Abstract viewer_start entrypoint In-Reply-To: <479A0A64.7060801@redhat.com> References: <479A0A64.7060801@redhat.com> Message-ID: <20080125213049.GB20240@redhat.com> On Fri, Jan 25, 2008 at 04:12:20PM +0000, Richard W.M. Jones wrote: > This patch adjusts src/main.c so that there is an entrypoint called > viewer_start which will be used by the plugin. > > Note the header file ("viewer.h") and the function is private and > internal to virt-viewer. Thanks have applied this patch. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Fri Jan 25 21:34:26 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 25 Jan 2008 21:34:26 +0000 Subject: [et-mgmt-tools] [PATCH 2/2] Virt-viewer browser plugin In-Reply-To: <479A0CBD.9000508@redhat.com> References: <479A0CBD.9000508@redhat.com> Message-ID: <20080125213426.GC20240@redhat.com> On Fri, Jan 25, 2008 at 04:22:21PM +0000, Richard W.M. Jones wrote: > This adds a virt-viewer browser plugin which works on Firefox / Mozilla > and probably other NPAPI-based browsers on Unix-like systems. > > The plugin is placed in the plugin/ subdirectory. > > Libtool is required (since the plugin is a shared library). > > The plugin is disabled by default. To enable it you have to do > './configure --enable-plugin'. > > To use the plugin, place code such as this in a web page: > > width="800" > height="600" > uri="qemu:///system" > name="1"> > > > The width & height are required, and should be at least as large as the > virtual machine's console. (Because Gecko + Gtk plugins are really > buggy we are unable to place scrollbars or a menubar around the VNC widget). > > 'uri' (libvirt URI) and 'name' (domain name/ID/UUID) are required. > > Optionally you can also have waitvnc="1" and direct="1" which have the > same effect as the equivalent virt-viewer command line options. > > Screenshot: http://annexia.org/tmp/Screenshot-SeaMonkey.png > > Known bugs, at least 4 at the moment: > > (1) There's a strange race condition where Gecko renders the plugin > before rendering the surrounding page. This results in a segfault > because the browser sends uninitialized private data to the plugin. > > (2) The plugin cannot grab the mouse, so mouse handling is erratic. Hmm, that should be fixable - X allows grabs to be constrained to arbitrary windows, not just top level, so the browser widget ought to be able to grab it. > (3) Virt-viewer menubar doesn't render correctly, so in this version it > is disabled. Appears to be a general problem with Gecko + Gtk. Could perhaps use a combo of Ctrl+Alt + Right click to display a popup menu instead of the menu bar. Just use a plain right click of course in case a remote server wants it. > > (4) There are various rendering artifacts on framebuffer updates, but > they only affect movement/animation, eg. dragging windows. When the > animation has finished the rendering artifacts disappear. This one's very odd. Did the plain GTK-VNC plugin show the same thing or is it just the virt-viewer variant ? Despite these issues, I applied the patch - since it is disabled by default it won't hurt to have it in the repo & saves carrying around separate patches. Dan -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From fekete at melkor.dnp.fmph.uniba.sk Mon Jan 28 09:40:34 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Mon, 28 Jan 2008 10:40:34 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine Message-ID: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> Hi *, I need to create an isntalltion server with support of network boot. Friends told me about cobbler koan and kickstart. I choosed to start with cobbler. Problem is than in every documentation/quick guide the first step is: cobbler check (run as root). When I do that, the result is: Traceback (most recent call last): File "/usr/lib/python2.3/site-packages/cobbler/cobbler.py", line 719, in main BootCLI(sys.argv).run() File "/usr/lib/python2.3/site-packages/cobbler/cobbler.py", line 42, in __init__ self.api = api.BootAPI() File "/usr/lib/python2.3/site-packages/cobbler/api.py", line 40, in __init__ self.deserialize() File "/usr/lib/python2.3/site-packages/cobbler/api.py", line 170, in deserialize return self._config.deserialize() File "/usr/lib/python2.3/site-packages/cobbler/config.py", line 169, in deserialize if not serializer.deserialize(x,topological=True): File "/usr/lib/python2.3/site-packages/cobbler/serializer.py", line 79, in deserialize datastruct.sort(cmp=__depth_cmp) TypeError: sort() takes no keyword arguments Machine runs on Scientific Linux 4.0 (RedHat Enterprise Linux) uname: 2.6.9-55.EL.cern #1 Thu May 10 17:40:14 CEST 2007 x86_64 x86_64 x86_64 GNU/Linux package version is cobbler-0.5.0-1.el4.rf Could anybody help me how to solve this problem ? Many Thanks. V. From hyclak at math.ohiou.edu Mon Jan 28 12:12:22 2008 From: hyclak at math.ohiou.edu (Matt Hyclak) Date: Mon, 28 Jan 2008 07:12:22 -0500 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> Message-ID: <20080128121221.GG29571@math.ohiou.edu> On Mon, Jan 28, 2008 at 10:40:34AM +0100, Vladimir Fekete enlightened us: > I need to create an isntalltion server with support of network boot. > Friends told me about cobbler koan and kickstart. I choosed to start with > cobbler. Problem is than in every documentation/quick guide the first step > is: cobbler check (run as root). When I do that, the result is: > > > Traceback (most recent call last): > File "/usr/lib/python2.3/site-packages/cobbler/cobbler.py", line 719, in > main > BootCLI(sys.argv).run() > File "/usr/lib/python2.3/site-packages/cobbler/cobbler.py", line 42, in > __init__ > self.api = api.BootAPI() > File "/usr/lib/python2.3/site-packages/cobbler/api.py", line 40, in > __init__ > self.deserialize() > File "/usr/lib/python2.3/site-packages/cobbler/api.py", line 170, in > deserialize > return self._config.deserialize() > File "/usr/lib/python2.3/site-packages/cobbler/config.py", line 169, in > deserialize > if not serializer.deserialize(x,topological=True): > File "/usr/lib/python2.3/site-packages/cobbler/serializer.py", line 79, in > deserialize > datastruct.sort(cmp=__depth_cmp) > TypeError: sort() takes no keyword arguments > > > > Machine runs on Scientific Linux 4.0 (RedHat Enterprise Linux) > uname: > 2.6.9-55.EL.cern #1 Thu May 10 17:40:14 CEST 2007 x86_64 x86_64 x86_64 GNU/Linux > package version is cobbler-0.5.0-1.el4.rf > > Could anybody help me how to solve this problem ? Many Thanks. > > 0.5.0 is very old. You should upgrade to 0.6.5 and see if the problem persists. Matt -- Matt Hyclak Department of Mathematics Department of Social Work Ohio University (740) 593-1263 From swissslinky at gmail.com Mon Jan 28 12:59:09 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Mon, 28 Jan 2008 13:59:09 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <20080128121221.GG29571@math.ohiou.edu> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> Message-ID: where are people even finding 0.5.0? -------------- next part -------------- An HTML attachment was scrubbed... URL: From fekete at melkor.dnp.fmph.uniba.sk Mon Jan 28 14:53:18 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Mon, 28 Jan 2008 15:53:18 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> Message-ID: <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> CERN V. On Mon, Jan 28, 2008 at 01:59:09PM +0100, Dan Dengate wrote: > where are people even finding 0.5.0? > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From swissslinky at gmail.com Mon Jan 28 15:04:23 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Mon, 28 Jan 2008 16:04:23 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> Message-ID: Sorry Vladimir, I'll give the repos a kick =) On 28/01/2008, Vladimir Fekete wrote: > > CERN > > V. > > On Mon, Jan 28, 2008 at 01:59:09PM +0100, Dan Dengate wrote: > > where are people even finding 0.5.0? > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- MT Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fekete at melkor.dnp.fmph.uniba.sk Mon Jan 28 15:23:24 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Mon, 28 Jan 2008 16:23:24 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> Message-ID: <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> That would be nice. Thanks! V. On Mon, Jan 28, 2008 at 04:04:23PM +0100, Dan Dengate wrote: > Sorry Vladimir, I'll give the repos a kick =) > > > > > > On 28/01/2008, Vladimir Fekete wrote: > > > > CERN > > > > V. > > > > On Mon, Jan 28, 2008 at 01:59:09PM +0100, Dan Dengate wrote: > > > where are people even finding 0.5.0? > > > > > _______________________________________________ > > > et-mgmt-tools mailing list > > > et-mgmt-tools at redhat.com > > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > > > -- > MT > Dan > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From swissslinky at gmail.com Mon Jan 28 15:52:50 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Mon, 28 Jan 2008 16:52:50 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> Message-ID: Not us. Blame dag. On 28/01/2008, Vladimir Fekete wrote: > > That would be nice. Thanks! > > V. > > > On Mon, Jan 28, 2008 at 04:04:23PM +0100, Dan Dengate wrote: > > Sorry Vladimir, I'll give the repos a kick =) > > > > > > > > > > > > On 28/01/2008, Vladimir Fekete wrote: > > > > > > CERN > > > > > > V. > > > > > > On Mon, Jan 28, 2008 at 01:59:09PM +0100, Dan Dengate wrote: > > > > where are people even finding 0.5.0? > > > > > > > _______________________________________________ > > > > et-mgmt-tools mailing list > > > > et-mgmt-tools at redhat.com > > > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > > > _______________________________________________ > > > et-mgmt-tools mailing list > > > et-mgmt-tools at redhat.com > > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > > > > > > > > -- > > MT > > Dan > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- MT Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Mon Jan 28 15:56:43 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 28 Jan 2008 10:56:43 -0500 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> Message-ID: <479DFB3B.6050809@redhat.com> Dan Dengate wrote: > Not us. Blame dag. > You might want to consider adding EPEL to your yum configuration -- it will always have the newest version (I'm the EPEL owner for cobbler), and yum will find that first. --Michael From maxwell.bottiger at itt.com Mon Jan 28 18:19:07 2008 From: maxwell.bottiger at itt.com (Maxwell Bottiger) Date: Mon, 28 Jan 2008 13:19:07 -0500 Subject: [et-mgmt-tools] Create network bridged to eth0 Message-ID: <1201544347.2974.7.camel@glaucus.modsim.mil> Hi, I'm not sure if this is a mailing list for development specific questions or not, but it's the only one listed on RedHat's Virt-manager page. So, if I've posted to the wrong list, I apologize in advance. I'm using virt-manager 0.5-3 with KVM and Qemu, and want to create virtual machines which have a static IP on my lab's network. I haven't figured out how to do this within the manager application, and I haven't seen any libvirt network xml definitions which look very much like the ones on my system. Is there a howto document out there which covers this? Maybe I've been coddled for too many years by VMWare's installer which set up the bridging for me, but I'm pretty lost at the moment. Thanks for your help. -Max This e-mail and any files transmitted with it may be proprietary and are intended solely for the use of the individual or entity to whom they are addressed. If you have received this e-mail in error please notify the sender. Please note that any views or opinions presented in this e-mail are solely those of the author and do not necessarily represent those of ITT Corporation. The recipient should check this e-mail and any attachments for the presence of viruses. ITT accepts no liability for any damage caused by any virus transmitted by this e-mail. From swissslinky at gmail.com Mon Jan 28 22:45:24 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Mon, 28 Jan 2008 23:45:24 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <479DFB3B.6050809@redhat.com> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> Message-ID: I'm personally ok. I've brought this up with the team though, so we shall see. We seem to mirror the rest of the world. Short of mailing Dag, how can you bump it in dag? On 28/01/2008, Michael DeHaan wrote: > > Dan Dengate wrote: > > Not us. Blame dag. > > > You might want to consider adding EPEL to your yum configuration -- it > will always have the newest version (I'm the EPEL owner for cobbler), > and yum will find that first. > > > --Michael > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- MT Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Mon Jan 28 22:49:19 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 28 Jan 2008 17:49:19 -0500 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> Message-ID: <479E5BEF.6000101@redhat.com> Dan Dengate wrote: > I'm personally ok. I've brought this up with the team though, so we > shall see. We seem to mirror the rest of the world. > Including EPEL in the list of repos you use will fix your problem there. > Short of mailing Dag, how can you bump it in dag? Probably email DAG and ask him to not mirror EPEL packages, or to mirror EPEL packages directly. Regardless -- putting EPEL in your list of yum repos will ensure you get the updated software. --Michael > > > On 28/01/2008, *Michael DeHaan* > wrote: > > Dan Dengate wrote: > > Not us. Blame dag. > > > You might want to consider adding EPEL to your yum configuration -- it > will always have the newest version (I'm the EPEL owner for cobbler), > and yum will find that first. > > > --Michael > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > > -- > MT > Dan > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From fekete at melkor.dnp.fmph.uniba.sk Tue Jan 29 11:31:30 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Tue, 29 Jan 2008 12:31:30 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <479E5BEF.6000101@redhat.com> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> Message-ID: <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> Hi, I tryied to add epel to yum repos using: rpm -Uvh http://download.fedora.redhat.com/pub/epel/5/i386/epel-release-5-2.noarch.rpm and after update yum said that there is missing dependency for python: Resolving Dependencies --> Populating transaction set with selected packages. Please wait. ---> Downloading header for cobbler to pack into transaction set. cobbler-0.6.4-2.el5.noarc 100% |=========================| 37 kB 00:00 ---> Package cobbler.noarch 0:0.6.4-2.el5 set to be updated ---> Downloading header for koan to pack into transaction set. koan-0.6.3-3.el5.noarch.r 100% |=========================| 7.8 kB 00:00 ---> Package koan.noarch 0:0.6.3-3.el5 set to be updated ---> Downloading header for python-cheetah to pack into transaction set. python-cheetah-2.0-0.6.rc 100% |=========================| 23 kB 00:00 ---> Package python-cheetah.x86_64 0:2.0-0.6.rc8.el5 set to be updated --> Running transaction check --> Processing Dependency: rtld(GNU_HASH) for package: python-cheetah --> Processing Dependency: python(abi) = 2.4 for package: python-cheetah --> Processing Dependency: python(abi) = 2.4 for package: koan --> Processing Dependency: python(abi) = 2.4 for package: cobbler --> Finished Dependency Resolution Error: Missing Dependency: rtld(GNU_HASH) is needed by package python-cheetah Error: Missing Dependency: python(abi) = 2.4 is needed by package python-cheetah Error: Missing Dependency: python(abi) = 2.4 is needed by package koan Error: Missing Dependency: python(abi) = 2.4 is needed by package cobbler which is true, because after last upgrade python was in version 2.3.4 but this is problem of ScientificLinux. Anyway, there is tftp server available as well as som pxe utils, so I should be able to make what I need without using cobbler. V. On Mon, Jan 28, 2008 at 05:49:19PM -0500, Michael DeHaan wrote: > Dan Dengate wrote: > >I'm personally ok. I've brought this up with the team though, so we > >shall see. We seem to mirror the rest of the world. > > > Including EPEL in the list of repos you use will fix your problem there. > > >Short of mailing Dag, how can you bump it in dag? > > Probably email DAG and ask him to not mirror EPEL packages, or to mirror > EPEL packages directly. > Regardless -- putting EPEL in your list of yum repos will ensure you get > the updated software. > > --Michael > > > > > > > > >On 28/01/2008, *Michael DeHaan* >> wrote: > > > > Dan Dengate wrote: > > > Not us. Blame dag. > > > > > You might want to consider adding EPEL to your yum configuration -- it > > will always have the newest version (I'm the EPEL owner for cobbler), > > and yum will find that first. > > > > > > --Michael > > > > > > > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > > > > > > >-- > >MT > >Dan > >------------------------------------------------------------------------ > > > >_______________________________________________ > >et-mgmt-tools mailing list > >et-mgmt-tools at redhat.com > >https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mdehaan at redhat.com Tue Jan 29 14:37:52 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 29 Jan 2008 09:37:52 -0500 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> References: <20080128094034.GA14289@melkor.dnp.fmph.uniba.sk> <20080128121221.GG29571@math.ohiou.edu> <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> Message-ID: <479F3A40.8000402@redhat.com> > > which is true, because after last upgrade python was in version 2.3.4 but > this is problem of ScientificLinux. You could of course just do rpmbuild --rebuild on the cobbler src RPM. Cobbler does not require that particular Python ABI. From fekete at melkor.dnp.fmph.uniba.sk Tue Jan 29 15:27:07 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Tue, 29 Jan 2008 16:27:07 +0100 Subject: [et-mgmt-tools] Crash of cobbler on x86_64 machine In-Reply-To: <479F3A40.8000402@redhat.com> References: <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> <479F3A40.8000402@redhat.com> Message-ID: <20080129152707.GA12515@melkor.dnp.fmph.uniba.sk> On Tue, Jan 29, 2008 at 09:37:52AM -0500, Michael DeHaan wrote: > > > > >which is true, because after last upgrade python was in version 2.3.4 but > >this is problem of ScientificLinux. > > > You could of course just do rpmbuild --rebuild on the cobbler src RPM. > > Cobbler does not require that particular Python ABI. Well... Rebuilding of package or any software is always a nasty thing (hell of dependencies & compilers & environment). I don't want to spend 95% of my life with searching for correct dependency version and correct dependency version for previous dependency, e.t.c.... That's the reason why people invented thing such as yum or apt. Next problem is upgrade of the system (=upgrade of dependencies => new build which does not have to work, because of new features which may need new stuff you don't know about, so you spend 6 nights just to check what you really need). Also I'd like to add, that it should be responsibility of maintainer to keep things running for given platform. Besides, if the highest version of application in specific distribution is so old that it cannot work with rest of the distro then it's time to think about another distribution. (As I said in previous mail, problem is already solved: Debian+tftp server+pxe... without problem) V. > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mdehaan at redhat.com Tue Jan 29 15:51:25 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 29 Jan 2008 10:51:25 -0500 Subject: [et-mgmt-tools] EPEL/Scientific Linux [WAS: "Crash" of cobbler on x86_64 machine] In-Reply-To: <20080129152707.GA12515@melkor.dnp.fmph.uniba.sk> References: <20080128145318.GB15091@melkor.dnp.fmph.uniba.sk> <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> <479F3A40.8000402@redhat.com> <20080129152707.GA12515@melkor.dnp.fmph.uniba.sk> Message-ID: <479F4B7D.9060808@redhat.com> Vladimir Fekete wrote: > On Tue, Jan 29, 2008 at 09:37:52AM -0500, Michael DeHaan wrote: > >>> which is true, because after last upgrade python was in version 2.3.4 but >>> this is problem of ScientificLinux. >>> >> You could of course just do rpmbuild --rebuild on the cobbler src RPM. >> >> Cobbler does not require that particular Python ABI. >> > > I don't want to spend > 95% of my life with searching for correct dependency version and correct > dependency version for previous dependency, e.t.c.... > > You seem to be saying that the packages in the EPEL 4/5 tree do not work for Scientific Linux? I'm finding this a bit strange, as EPEL is intended to work for that distro and this is the first I've heard of that. http://fedoraproject.org/wiki/EPEL/About#head-7f5c56add93b7a549e24150cc4e0183aefc9120b > Next problem is upgrade of the system (=upgrade of dependencies => new build > which does not have to work, because of new features which may need new > stuff you don't know about, so you spend 6 nights just to check what you > really need). Also I'd like to add, that it should be responsibility of > maintainer to keep things running for given platform. > If your SL can't use the EPEL packages, it would be best to bring it up with those lists. > Besides, if the highest version of application in specific distribution > is so old that it cannot work with rest of the distro then it's time to think > about another distribution. (As I said in previous mail, problem is already solved: > Debian+tftp server+pxe... without problem) > Cobbler is not just TFTP. If all you need is a simple /tftpboot tree, of course, you can use whatever you like. Of course your current distro already contains a tftp server as well :) From fekete at melkor.dnp.fmph.uniba.sk Tue Jan 29 16:20:45 2008 From: fekete at melkor.dnp.fmph.uniba.sk (Vladimir Fekete) Date: Tue, 29 Jan 2008 17:20:45 +0100 Subject: [et-mgmt-tools] EPEL/Scientific Linux [WAS: "Crash" of cobbler on x86_64 machine] In-Reply-To: <479F4B7D.9060808@redhat.com> References: <20080128152324.GA23629@melkor.dnp.fmph.uniba.sk> <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> <479F3A40.8000402@redhat.com> <20080129152707.GA12515@melkor.dnp.fmph.uniba.sk> <479F4B7D.9060808@redhat.com> Message-ID: <20080129162045.GA14640@melkor.dnp.fmph.uniba.sk> On Tue, Jan 29, 2008 at 10:51:25AM -0500, Michael DeHaan wrote: > Vladimir Fekete wrote: > >On Tue, Jan 29, 2008 at 09:37:52AM -0500, Michael DeHaan wrote: > > > >>>which is true, because after last upgrade python was in version 2.3.4 but > >>>this is problem of ScientificLinux. > >>> > >>You could of course just do rpmbuild --rebuild on the cobbler src RPM. > >> > >>Cobbler does not require that particular Python ABI. > >> > > > >I don't want to spend > >95% of my life with searching for correct dependency version and correct > >dependency version for previous dependency, e.t.c.... > > > > > You seem to be saying that the packages in the EPEL 4/5 tree do not work > for Scientific Linux? > Yes, so it seems. I followed instructions at page: http://fedoraproject.org/wiki/EPEL/FAQ#howtouse then I run yum update yum upgrade and after that cobbler dep. problem showed up I guess that page is general way how to use epel. Obviously I modified/created repo files for yum correctly (tree has been found and yum wanted to start with upgrade). I have no doubts that EPEL is intended to work for SLC, however, when I followed steps on mentioned site, it does not work. > I'm finding this a bit strange, as EPEL is intended to work for that > distro and this is the first I've heard of > that. > > http://fedoraproject.org/wiki/EPEL/About#head-7f5c56add93b7a549e24150cc4e0183aefc9120b > > > >Next problem is upgrade of the system (=upgrade of dependencies => new > >build > >which does not have to work, because of new features which may need new > >stuff you don't know about, so you spend 6 nights just to check what you > >really need). Also I'd like to add, that it should be responsibility of > >maintainer to keep things running for given platform. > > > > If your SL can't use the EPEL packages, it would be best to bring it up > with those lists. > > >Besides, if the highest version of application in specific distribution > >is so old that it cannot work with rest of the distro then it's time to > >think > >about another distribution. (As I said in previous mail, problem is > >already solved: > >Debian+tftp server+pxe... without problem) > > > > Cobbler is not just TFTP. If all you need is a simple /tftpboot tree, of > course, you can use whatever you like. > Of course your current distro already contains a tftp server as well :) I wanted to create an installation server. TFTP+PXE were only part of the services that server should provide. However, when this problem with cobbler arrised, I started to think whether I really need it or not. Result is that now it's not beauty, it's not easy to maintain, it is slow, but it works, at least. (I did not wanted to install cobbler only because of TFTP) V. > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From swissslinky at gmail.com Tue Jan 29 18:47:49 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Tue, 29 Jan 2008 19:47:49 +0100 Subject: [et-mgmt-tools] EPEL/Scientific Linux [WAS: "Crash" of cobbler on x86_64 machine] In-Reply-To: <20080129162045.GA14640@melkor.dnp.fmph.uniba.sk> References: <479DFB3B.6050809@redhat.com> <479E5BEF.6000101@redhat.com> <20080129113130.GA7244@melkor.dnp.fmph.uniba.sk> <479F3A40.8000402@redhat.com> <20080129152707.GA12515@melkor.dnp.fmph.uniba.sk> <479F4B7D.9060808@redhat.com> <20080129162045.GA14640@melkor.dnp.fmph.uniba.sk> Message-ID: All I know is that I just installed Cobbler and Koan on a clean up-to-date EL(slc)4 from SRPM and it works like a dream. All I needed was python-cheetah & and something or other so insignificant that I care not to remember.. it was CERN branded though, It was a real breeze* However, when I tried this last night with yum, all hell broke out. dag was useless but hopefully it can be fixed quickly. It's a problem with CERN Scientific Linux, not necessarily SL or Cobbler. It's a fragile thing, SLC4, so when you give it a bit more room to play with, it can all go wrong. I just threw EPEL in and went off to grab Cobbler but it all went horribly wrong... I tried newer versions of python (2.3 & 2.4 & 2.5), but this kills yum completely... nothing works because of this dependency and that dependency. being wrong with t'other dependency. So I removed myself from that situation as I wasn't in control. I'm going to try and do it that way again shortly, to find out why. point is, it does work on EL4*. I'm going to compare to the devel branch shortly. Vladimir, I recommend you give Cobbler a quick go again, as it's so damn cool I think you'll like it**. A recipe like: [No warranty offered, if it breaks, keep both pieces] check your /etc/*.repo's are all enabled in the right places. yum install python-cheetah wget -vvv http://cobbler.et.redhat.com/download/cobbler-0.6.5-1.fc8.src.rpm wget -vvv http://cobbler.et.redhat.com/download/koan-0.6.4.tar.gz rpmbuild --rebuild cobbler-0.6.5-1.fc8.src.rpm rpmbuild --rebuild koan-0.6.4.tar.gz rpm -Uvh /usr/src/redhat/RPMS/noarch/cobbler-0.6.5-1.noarch.rpm rpm -Uvh /usr/src/redhat/RPMS/noarch/koan-0.6.4-1.noarch.rpm (you might be missing the odd thing here and there, but nothing which should cause a problem.) Run "cobbler check" just to familiarise yourself with what you're about to go through. Things like [1] /sbin/services httpd start [2] /sbin/services cobblerd start [3] /sbin/services iptables stop [4] Amend NEXT_SERVER and SERVER fields in /var/lib/cobbler/settings You might want to formulate a plan for [3]. Then follow the instructions in the man page: WEB USER INTERFACE Most of the day-to-day actions in cobbler's command line can be performed in Cobbler's Web UI. To enable and access the WebUI, perform the following steps. 1) Set xmlrpc_rw_enabled to 1 in /var/lib/cobbler/settings to enable network control. 2) Change the admin xmlrpc secret in /etc/cobbler/auth.conf. You won't have to remember it. 3) The default Web UI password is "cobbler/ILoveCobbler", to change this, run: htdigest /var/www/cgi-bin/cobbler/.htpasswd "Cobbler WebUI Authentication" cobbler 4) SELinux users may also have to run: setsebool httpd_can_network_connect true chcon httpd_sys_content_t /etc/cobbler/auth.conf 5) Run /sbin/service cobblerd restart. 6) Visit http://yourserver.example.org/cgi-bin/cobbler/webui.cgi and log in with whatever you chose in step 3. I didn't have to worry about 4) really. By now, things should be green. Seriously, read "man cobbler" and "man koan". They are actually two man pages which are nice to read. Hope some of that helps :) * /var/www/cgi-bin/cobbler/.htaccess not containing AuthDigestFile stumped me for a coffee again. ** why are you creating this setup from stratch? Does using AIMS not interest you? On 29/01/2008, Vladimir Fekete wrote: > > On Tue, Jan 29, 2008 at 10:51:25AM -0500, Michael DeHaan wrote: > > Vladimir Fekete wrote: > > >On Tue, Jan 29, 2008 at 09:37:52AM -0500, Michael DeHaan wrote: > > > > > >>>which is true, because after last upgrade python was in version 2.3.4but > > >>>this is problem of ScientificLinux. > > >>> > > >>You could of course just do rpmbuild --rebuild on the cobbler src RPM. > > >> > > >>Cobbler does not require that particular Python ABI. > > >> > > > > > >I don't want to spend > > >95% of my life with searching for correct dependency version and > correct > > >dependency version for previous dependency, e.t.c.... > > > > > > > > You seem to be saying that the packages in the EPEL 4/5 tree do not work > > for Scientific Linux? > > > Yes, so it seems. > I followed instructions at page: > http://fedoraproject.org/wiki/EPEL/FAQ#howtouse > > then I run > > yum update > yum upgrade > > and after that cobbler dep. problem showed up > > I guess that page is general way how to use epel. Obviously I > modified/created repo files for yum correctly (tree has been found and yum > wanted to start with upgrade). I have no doubts that EPEL is intended to > work for SLC, however, when I followed steps on mentioned site, it does > not > work. > > > I'm finding this a bit strange, as EPEL is intended to work for that > > distro and this is the first I've heard of > > that. > > > > > http://fedoraproject.org/wiki/EPEL/About#head-7f5c56add93b7a549e24150cc4e0183aefc9120b > > > > > > >Next problem is upgrade of the system (=upgrade of dependencies => new > > >build > > >which does not have to work, because of new features which may need new > > >stuff you don't know about, so you spend 6 nights just to check what > you > > >really need). Also I'd like to add, that it should be responsibility of > > >maintainer to keep things running for given platform. > > > > > > > If your SL can't use the EPEL packages, it would be best to bring it up > > with those lists. > > > > >Besides, if the highest version of application in specific distribution > > >is so old that it cannot work with rest of the distro then it's time to > > >think > > >about another distribution. (As I said in previous mail, problem is > > >already solved: > > >Debian+tftp server+pxe... without problem) > > > > > > > Cobbler is not just TFTP. If all you need is a simple /tftpboot tree, of > > course, you can use whatever you like. > > Of course your current distro already contains a tftp server as well :) > > I wanted to create an installation server. TFTP+PXE were only part of > the services that server should provide. However, when this problem with > cobbler arrised, I started to think whether I really need it or not. > Result > is that now it's not beauty, it's not easy to maintain, it is slow, but it > works, at least. (I did not wanted to install cobbler only because of > TFTP) > > V. > > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- MT Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From gavin at redhat.com Tue Jan 29 20:44:59 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Tue, 29 Jan 2008 15:44:59 -0500 Subject: [et-mgmt-tools] koan patch: reinstall from kickstart file without a cobbler server Message-ID: <479F904B.7040303@redhat.com> koan patch: reinstall from kickstart file without a cobbler server This is my first crack at a koan patch to do this, please review it and consider it for inclusion in koan, I'm happy to make whatever changes are needed, and fix whatever mistakes I've made. This patch alters koan to allow it reinstall a machine from a kickstart file, without the need for a cobbler server. It's a rather large patch given what it does. Most of the changes are updating koan not to assume that it is connected to a cobbler server. If you specify the --kickstart=URL option instead of a --server option, it will load the kickstart file from the given URL. From the kickstart file it determines the URL of the install tree, and gets the install kernel and initrd from within the install tree. -gavin... -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: patch URL: From mdehaan at redhat.com Tue Jan 29 20:47:35 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 29 Jan 2008 15:47:35 -0500 Subject: [et-mgmt-tools] koan patch: reinstall from kickstart file without a cobbler server In-Reply-To: <479F904B.7040303@redhat.com> References: <479F904B.7040303@redhat.com> Message-ID: <479F90E7.2040203@redhat.com> Gavin Romig-Koch wrote: > koan patch: reinstall from kickstart file without a cobbler server > > This is my first crack at a koan patch to do this, please review it > and consider it for inclusion in koan, I'm happy to make whatever > changes are needed, and fix whatever mistakes I've made. > > This patch alters koan to allow it reinstall a machine from a > kickstart file, without the need for a cobbler server. It's a rather > large patch given what it does. Most of the changes are updating koan > not to assume that it is connected to a cobbler server. > > If you specify the --kickstart=URL option instead of a --server > option, it will load the kickstart file from the given URL. From the > kickstart file it determines the URL of the install tree, and gets the > install kernel and initrd from within the install tree. Great, I'll test this out... Possibly to reduce confusion, I might rename this to "--no-cobbler=" or equivalent. --Michael > > > -gavin... > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From viswanath.t at atc.tcs.com Wed Jan 30 10:32:32 2008 From: viswanath.t at atc.tcs.com (Viswanath T K) Date: Wed, 30 Jan 2008 16:02:32 +0530 Subject: [et-mgmt-tools] problem with using virt-manager In-Reply-To: <4794EFAB.6010800@redhat.com> References: <4794EFAB.6010800@redhat.com> Message-ID: <1201689152.2926.16.camel@localhost.localdomain> Hi All, I am using a x86 machine ( 64 bit ) and fedora 7. When I try to install a guest using virt-manager I am getting the error : TCP/IP Error : VNC connection to hypervisor host got refused or disconnected. virt-manager version I am using is 0.5.3 Help needed Thanks Viswanath T K From agx at sigxcpu.org Wed Jan 30 14:20:14 2008 From: agx at sigxcpu.org (Guido Guenther) Date: Wed, 30 Jan 2008 15:20:14 +0100 Subject: [et-mgmt-tools] [PATCH] virt-installer: update Debian OS type documentation Message-ID: <20080130142014.GA10942@bogon.ms20.nix> Hi, this updates the Debian os type documentation to what's actually implemented. Patch is against current hg. Please apply if appropriate. -- Guido # HG changeset patch # User Guido Guenther # Date 1201702187 -3600 # Node ID 224c30552ef4fa092a69f27a8ce837b9a15427a3 # Parent 96a103cd78dc7616a2e1e54f2e100fe156bebafd Docuement the Debian ostypes (patch by Trent W. Buck) diff -r 96a103cd78dc -r 224c30552ef4 man/en/virt-install.pod --- a/man/en/virt-install.pod Tue Jan 29 15:51:28 2008 +0100 +++ b/man/en/virt-install.pod Wed Jan 30 15:09:47 2008 +0100 @@ -330,6 +330,14 @@ Fedora 7 Suse Linux Enterprise Server 10.x +=item debianEtch + +Debian 4.0 (Etch) + +=item debianLenny + +Debian Lenny + =item generic26 Generic Linux 2.6.x kernel From crobinso at redhat.com Wed Jan 30 22:35:48 2008 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 30 Jan 2008 17:35:48 -0500 Subject: [et-mgmt-tools] [PATCH] virt-installer: update Debian OS type documentation In-Reply-To: <20080130142014.GA10942@bogon.ms20.nix> References: <20080130142014.GA10942@bogon.ms20.nix> Message-ID: <47A0FBC4.1030602@redhat.com> Guido Guenther wrote: > Hi, > this updates the Debian os type documentation to what's actually > implemented. Patch is against current hg. Please apply if appropriate. > -- Guido > > # HG changeset patch > # User Guido Guenther > # Date 1201702187 -3600 > # Node ID 224c30552ef4fa092a69f27a8ce837b9a15427a3 > # Parent 96a103cd78dc7616a2e1e54f2e100fe156bebafd > Docuement the Debian ostypes (patch by Trent W. Buck) > > diff -r 96a103cd78dc -r 224c30552ef4 man/en/virt-install.pod > --- a/man/en/virt-install.pod Tue Jan 29 15:51:28 2008 +0100 > +++ b/man/en/virt-install.pod Wed Jan 30 15:09:47 2008 +0100 > @@ -330,6 +330,14 @@ Fedora 7 > > Suse Linux Enterprise Server 10.x > > +=item debianEtch > + > +Debian 4.0 (Etch) > + > +=item debianLenny > + > +Debian Lenny > + > =item generic26 > > Generic Linux 2.6.x kernel > I used pod2man to generate the appropriate man page and applied this: http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=9eccd3abccb6 Thanks! Cole From mdehaan at redhat.com Thu Jan 31 16:30:42 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 31 Jan 2008 11:30:42 -0500 Subject: [et-mgmt-tools] koan patch: reinstall from kickstart file without a cobbler server In-Reply-To: <479F904B.7040303@redhat.com> References: <479F904B.7040303@redhat.com> Message-ID: <47A1F7B2.2060903@redhat.com> Gavin Romig-Koch wrote: > koan patch: reinstall from kickstart file without a cobbler server > > This is my first crack at a koan patch to do this, please review it > and consider it for inclusion in koan, I'm happy to make whatever > changes are needed, and fix whatever mistakes I've made. > > This patch alters koan to allow it reinstall a machine from a > kickstart file, without the need for a cobbler server. It's a rather > large patch given what it does. Most of the changes are updating koan > not to assume that it is connected to a cobbler server. > > If you specify the --kickstart=URL option instead of a --server > option, it will load the kickstart file from the given URL. From the > kickstart file it determines the URL of the install tree, and gets the > install kernel and initrd from within the install tree. > I've applied this with just some very minor modifications. This is in git now and will be part of 0.6.5, which will likely be released at the same time we do cobbler 0.8 (the culmination of the development branch work on 0.7.x) which should be coming very shortly, after there's a final round of 0.7.X testing*. (Just to avoid confusion I'll probably just bump koan to 0.8 then as well.) I've basically changed --kickstart to --no-cobbler to reduce some confusion, changed some of the output text a bit, and added a section describing the usage to the manpage. This is a very neat patch because there are a lot of times people would want the equivalent of "reinstall this system" and for whatever reason don't have a Cobbler server, so I expect this will get a lot of use as people find out about it. Thanks, --Michael * = the two final things I want to explore for Cobbler 0.7.X are VMware and support for recursive syncs (effectively eliminating most needs of running "cobbler sync") and deletes. From berrange at redhat.com Thu Jan 31 16:44:39 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 31 Jan 2008 16:44:39 +0000 Subject: [et-mgmt-tools] FYI: virt-manager authentication support Message-ID: <20080131164439.GA19721@redhat.com> Now that libvirt 0.4.0 is available, it is possible to do authentication against PolicyKit and SASL. The latter gives Kerberos, PAM and Username & passord based auth, and whatever else SASL supports. Of course some of these methods require prompting for passwords, so this needs support in client apps using libvirt. The changes I've just pushed to virt-manager provide UI to let us prompt for usernames & passwords if the remote libvirt server requests them. So we should now be able to handle any auth type that libvirt offers. Currently it'll always prompt each time you connect if the server needs authentication, so we may wish to do further work to integrate with the gnome-keyring to (optionally) save usernames/passwords. Personally I use Kerberos authentication - Free IPA (www.freeipa.org) makes it very easy to get a kerberos server up & running on your LAN. If you make your desktop authenticate against kerberos, and turn on kerberos in your libvirt servers, then you get automatic single-sign-on between virt-manager & libvirt servers, no passwords required post-login. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From gavin at redhat.com Thu Jan 31 19:54:12 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Thu, 31 Jan 2008 14:54:12 -0500 Subject: [et-mgmt-tools] koan patch: reinstall from kickstart file without a cobbler server In-Reply-To: <47A1F7B2.2060903@redhat.com> References: <479F904B.7040303@redhat.com> <47A1F7B2.2060903@redhat.com> Message-ID: <47A22764.4000607@redhat.com> thank you. -gavin...