April 2010 - Libguestfs - libguestfs List Archives

[PATCH] fish: Add -N option for making prepared disk images.
by Richard W.M. Jones 22 Apr '10

22 Apr '10

-- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones New in Fedora 11: Fedora Windows cross-compiler. Compile Windows programs, test, and build Windows installers. Over 70 libraries supprt'd http://fedoraproject.org/wiki/MinGW http://www.annexia.org/fedora_mingw

1 0

virtio is enabled by default in libguestfs?
by Yufang Zhang 22 Apr '10

22 Apr '10

Hi all, Could I ask if and from which version that virtio is enabled by default in libguestfs? It seems that it is enabled by default in fedora-12, on which the latest version of libguestfs is 1.0.85. But for fedora-11(the latest version of libguestfs is 1.0.72), I could enable virtio only when I set it explicitly via config command. Thus it is not enabled by default on fedora-11. So I'd like to know on which situation virtio will be enabled automatically by libguestfs? Thanks. --Yufang

2 2

[PATCH 1/2] Try to load the loop module before running mkinitrd
by Matthew Booth 22 Apr '10

22 Apr '10

mkinitrd needs to mount files using loop. loop might be a module, so try to load it first. --- lib/Sys/VirtV2V/GuestOS/RedHat.pm | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/lib/Sys/VirtV2V/GuestOS/RedHat.pm b/lib/Sys/VirtV2V/GuestOS/RedHat.pm index 0e469f5..08027b6 100644 --- a/lib/Sys/VirtV2V/GuestOS/RedHat.pm +++ b/lib/Sys/VirtV2V/GuestOS/RedHat.pm @@ -1114,6 +1114,13 @@ sub prepare_bootable $g->modprobe("ext2"); }; + # loop is a module in RHEL 5. Try to load it. Doesn't matter for other + # OSs if it doesn't exist, but RHEL 5 will complain: + # All of your loopback devices are in use. + eval { + $g->modprobe("loop"); + }; + $g->command(["/sbin/mkinitrd", @preload_args, $initrd, $version]); } -- 1.6.6.1

2 3

[PATCH] Whitelist the loop kernel module
by Matthew Booth 22 Apr '10

22 Apr '10

loop is sometimes a module, for example on RHEL 5. --- appliance/kmod.whitelist.in | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/appliance/kmod.whitelist.in b/appliance/kmod.whitelist.in index 814b9d0..96fc4c3 100644 --- a/appliance/kmod.whitelist.in +++ b/appliance/kmod.whitelist.in @@ -62,3 +62,5 @@ reiserfs.ko crc*.ko libcrc*.ko + +loop.ko -- 1.6.6.1

2 1

RFE: reuse of the libguestfs daemon + host API in a externally managed VM
by Daniel P. Berrange 22 Apr '10

22 Apr '10

The libvirt-TCK is a integration test suite we are using to validate the correct operation of libvirt drivers. Currently it is just checking the basic operation & functionality of drivers from the host side. we just boot a random Fedora kernel + initrd since we don't actually care that the guest OS boots into any particular state, we only care that its running in some form. We would like to extend our tests, to include validation of changes inside the guest. For example, in the TCK test case for CDROM media change, we want to be able to validate that the guest OS can actually see the new media. Or in disk hotplug we want to validate that the new PCI device appeared in the guest. etc etc. This requires that we have a real guest OS image running + some form of channel to get data out of the guest OS. To me it looks like the libguestfs appliance image / guest daemon would fit our needs just fine. On the host side, the libguestfs Perl bindings also fit the bill nicely. The only issue is that the guest VM can't be under libguestfs' control. The libvirt TCK test cases need to spawn / control the guest in whatever config it needs. My thought is that anytime the libvirt TCK needs a guest OS it can talk to, it can simply configure its VM to point at the libguestfs appliance image. Then tell the libguestfs API to attach to this pre-running VM via whatever chardev channel libvirt configured. IIUC, this could be a change such that instead of: guestfs_launch(h); the app could do: guestfs_attach(h, "/var/run/libvirt-tck/virtio-serial-guestfs.sock"); But perhaps also allow query of the expected kernel/initrd, so apps don't have to hardcode those location vmlinuz = guestfs_appliance_kernel(h); initrd = guestfs_appliance_initrd(h); Regards, Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://deltacloud.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[PATCH 1/2] Don't die during inspection if initrd doesn't exist
by Matthew Booth 21 Apr '10

21 Apr '10

This fixes a problem where inspection would die if grub.conf referenced a non-existent initrd. Just return an empty initrd instead. --- perl/lib/Sys/Guestfs/Lib.pm | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/perl/lib/Sys/Guestfs/Lib.pm b/perl/lib/Sys/Guestfs/Lib.pm index 5428b10..b5b3906 100644 --- a/perl/lib/Sys/Guestfs/Lib.pm +++ b/perl/lib/Sys/Guestfs/Lib.pm @@ -1825,7 +1825,7 @@ sub _inspect_initrd # Disregard old-style compressed ext2 files and only work with real # compressed cpio files, since cpio takes ages to (fail to) process anything # else. - if ($g->file ($path) =~ /cpio/) { + if ($g->exists($path) && $g->file($path) =~ /cpio/) { eval { @modules = $g->initrd_list ($path); }; -- 1.6.6.1

2 4

libguestfs mounting solaris 10 ZFS guest
by John Larson 20 Apr '10

20 Apr '10

Not sure if this possible, but I have a KVM guest running Solaris 10 with the OS on ZFS and I am trying to use libguestfs/guestfish/guestmount to get to the VM. I am running Red Hat EL 5.4 with EPEL rpms as required. The VM is on a LV and it boots fine, but I can't seem to get the syntax correct to get libguestfs to deal with it. Guestmount seemed like the best option because it supports FUSE modules. I have fuse installed and zfs-fuse and I can create ZFS filesystems on the physical host machine. Just really not sure how to make libguestfs deal with it. I have tried mounting the pool and the underlying virtual disk device with no success. # guestmount -o modules=zfs -a /dev/kvmVG00/sol10u6 -m newpool/ROOT/s10up-08 --ro /mnt libguestfs: error: mount_options: mount_options_stub: newpool/ROOT/s10up-08: expecting a device name # guestmount -o modules=zfs -a /dev/kvmVG00/sol10u6 -m /dev/dsk/c0d0s0 --ro /mnt libguestfs: error: mount_options: mount_options_stub: /dev/dsk/c0d0s0: No such file or directory # guestmount -o modules=zfs -a /dev/kvmVG00/sol10u6 -m /dev/dsk/c0t0d0s0 --ro /mnt libguestfs: error: mount_options: mount_options_stub: /dev/dsk/c0t0d0s0: No such file or directory # rpm -qa | grep fuse zfs-fuse-0.6.0-6.el5 fuse-2.7.4-8.el5 fuse-libs-2.7.4-8.el5 # rpm -qa | grep guest libguestfs-1.0.85-1.el5.9 guestfish-1.0.85-1.el5.9 perl-libguestfs-1.0.85-1.el5.9 libguestfs-tools-1.0.85-1.el5.9 libguestfs-mount-1.0.85-1.el5.9 libguestfs-devel-1.0.85-1.el5.9 Any help MUCH appreciated. John L

2 3

Error when compile libguestfs from source in Fedora-12
by Yufang Zhang 20 Apr '10

20 Apr '10

Hi all, I am trying to compile libguestfs from source in Fedora12, but failed with such error: cp: cannot stat `initramfs/fakeroot.log': No such file or directory make[2]: *** [../initramfs/fakeroot.log] Error 1 make[2]: Leaving directory `/home/yufang/Code/libguestfs/appliance' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/yufang/Code/libguestfs' make: *** [all] Error 2 I have installed all required packages listed in http://cvs.fedoraproject.org/viewvc/devel/libguestfs/libguestfs.spec?view=m… but still cannot find the reason. Could anybody do me a favour to have a look at this issue? config.log and output of make(make.log) are uploaded in the attachment to track problems. Thanks. --Yufang

2 6

Re: [Libguestfs] guestfish ... [SEC=UNCLASSIFIED]
by Richard W.M. Jones 20 Apr '10

20 Apr '10

On Mon, Apr 19, 2010 at 10:17:07PM +0800, Wilkinson, Alex wrote: > The aforementioned article says that libguestfs supports VMware's > VMDKs. This is partly true. We rely on qemu's support for VMDK which is rather sketchy. In particular, qemu cannot handle some VMDKs from ESX 4.0. However ... > So if these VMDKs live on a vmfs on an ESX server, what do > you recomend as the best approach to access them via libguestfs ? > Would it be best to mount the vmfs over nfs (i.e. to a linux box > that supports libguestfs) ? Best thing is to access the flat disk image, which ESX [all versions] makes available to you as a pseudo-file called <vm>-flat.vmdk. Despite the ".vmdk" extension, it's really just a raw disk image which libguestfs or any other program will have no problems with. I can't quite remember now if this is supported from NFS, but it's definitely supported if you ssh into the ESX storage (or use sshfs). Don't try to edit files if the virtual machine is running. Since there is no way for libguestfs to know if a remote VMWare VM is running, this is doubly important when modifying VMWare images. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones New in Fedora 11: Fedora Windows cross-compiler. Compile Windows programs, test, and build Windows installers. Over 70 libraries supprt'd http://fedoraproject.org/wiki/MinGW http://www.annexia.org/fedora_mingw

2 1

Should umask takes effect when we create device file via mknod?
by Yufang Zhang 19 Apr '10

19 Apr '10

Hi all, Currently, umask takes effect when we create device file via mknod, as bellow commands show: ><fs> mknod-b 0760 8 1 /dev/sdf ><fs> ll /dev/sdf brwxr----- 1 root root 8, 1 Apr 15 11:10 /sysroot/dev/sdf But I wonder whether it is reasonable? For mknod(1), when we use option -m mode, we set file permission bits to MODE, not a=rw - umask. Should this also be applicable to mknod in guestfish, thus when we create device file with mode set, umask doesn't take effect?

2 3