From dennisml at conversis.de Fri May 1 00:42:27 2009 From: dennisml at conversis.de (Dennis J.) Date: Fri, 01 May 2009 02:42:27 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090430223014.GC8394@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> Message-ID: <49FA4573.8030408@conversis.de> On 05/01/2009 12:30 AM, Richard W.M. Jones wrote: > On Fri, May 01, 2009 at 12:25:44AM +0200, Dennis J. wrote: >> On 05/01/2009 12:11 AM, Richard W.M. Jones wrote: >>> On Thu, Apr 30, 2009 at 10:39:37PM +0100, Richard W.M. Jones wrote: >>>> I haven't tried it, but it's possibly something you can do from >>>> libguestfs, or if not, it's something that we could add to libguestfs. >>> >>> OK, I just added grub-install to libguestfs: >>> >>> http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=b55bf8158f0b7f6b1760b7b3b5f7c1274a149127 >>> >>> (Make sure you have a back up of your guest image before experimenting >>> with this ...) >> >> Looking at that patch wouldn't specifying /dev/sda as device actually >> overwrite the physical boot sector on the host system rather than the one >> in the image? >> Or does libguestfs manipulate the device table so that access to /dev/sda >> get automatically gets re-routed to access the image instead? > > Neither. libguestfs boots a mini virtual machine, so /dev/sda really > is the right device all along. Ah, I didn't know that libguestfs actually uses its own vm. I just tried compiling libguestfs from git on f10 but it doesn't seem to like that at all: ... mv initramfs.fedora-10.i686.img initramfs.fedora-10.i686.img.bak mv: cannot stat `initramfs.fedora-10.i686.img': No such file or directory make[2]: [initramfs/fakeroot.log] Error 1 (ignored) mv vmlinuz.fedora-10.i686 vmlinuz.fedora-10.i686.bak mv: cannot stat `vmlinuz.fedora-10.i686': No such file or directory make[2]: [initramfs/fakeroot.log] Error 1 (ignored) if ! bash ./make-initramfs.sh; then rm -f initramfs/fakeroot.log; exit 1; fi run_yum: line 1: 19874 Segmentation fault yum -y -c "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap --noplugins --nogpgcheck --installroot="$target" install "$@" chroot: cannot run command `chmod': No such file or directory make[2]: *** [initramfs/fakeroot.log] Error 1 make[2]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' make: *** [all] Error 2 Regards, Dennis From rjones at redhat.com Fri May 1 08:34:23 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 09:34:23 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <49FA4573.8030408@conversis.de> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> Message-ID: <20090501083423.GA9933@amd.home.annexia.org> On Fri, May 01, 2009 at 02:42:27AM +0200, Dennis J. wrote: > Ah, I didn't know that libguestfs actually uses its own vm. > I just tried compiling libguestfs from git on f10 but it doesn't seem to > like that at all: > > ... > mv initramfs.fedora-10.i686.img initramfs.fedora-10.i686.img.bak > mv: cannot stat `initramfs.fedora-10.i686.img': No such file or directory > make[2]: [initramfs/fakeroot.log] Error 1 (ignored) > mv vmlinuz.fedora-10.i686 vmlinuz.fedora-10.i686.bak > mv: cannot stat `vmlinuz.fedora-10.i686': No such file or directory > make[2]: [initramfs/fakeroot.log] Error 1 (ignored) > if ! bash ./make-initramfs.sh; then rm -f initramfs/fakeroot.log; exit 1; fi > run_yum: line 1: 19874 Segmentation fault yum -y -c > "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap > --noplugins --nogpgcheck --installroot="$target" install "$@" > chroot: cannot run command `chmod': No such file or directory > make[2]: *** [initramfs/fakeroot.log] Error 1 > make[2]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' > make[1]: *** [all-recursive] Error 1 > make[1]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' > make: *** [all] Error 2 While it's possible to get libguestfs working on F10, it's not our minimum target, which is F11. In particular on F10 you will need to get the F11/F12 versions of the following packages: febootstrap >= 1.7 fakechroot >= 2.9-22 fakeroot >= 1.12.2 qemu >= 0.10-7 The following command should work _in theory_, but last time I tried, it didn't because of the F11 freeze: # yum --enablerepo=rawhide install febootstrap fakechroot fakeroot qemu so instead you need to get the packages from http://koji.fedoraproject.org (type the package name in the box at the top right). You could also try: rpmbuild -ta libguestfs-1.0.17.tar.gz Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From erenoglu at gmail.com Fri May 1 11:40:13 2009 From: erenoglu at gmail.com (Emre Erenoglu) Date: Fri, 1 May 2009 13:40:13 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501083423.GA9933@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> Message-ID: On Fri, May 1, 2009 at 10:34 AM, Richard W.M. Jones wrote: > > While it's possible to get libguestfs working on F10, it's not > our minimum target, which is F11. > > In particular on F10 you will need to get the F11/F12 versions of the > following packages: > > febootstrap >= 1.7 > fakechroot >= 2.9-22 > fakeroot >= 1.12.2 > qemu >= 0.10-7 > I would like to use libguestfs in my distro, but is the dependency to debootstrap mandatory? (as my distro is not Fedora?) -- Emre -------------- next part -------------- An HTML attachment was scrubbed... URL: From rjones at redhat.com Fri May 1 12:18:28 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 13:18:28 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> Message-ID: <20090501121828.GA10510@amd.home.annexia.org> On Fri, May 01, 2009 at 01:40:13PM +0200, Emre Erenoglu wrote: > On Fri, May 1, 2009 at 10:34 AM, Richard W.M. Jones wrote: > > While it's possible to get libguestfs working on F10, it's not > > our minimum target, which is F11. > > > > In particular on F10 you will need to get the F11/F12 versions of the > > following packages: > > > > febootstrap >= 1.7 > > fakechroot >= 2.9-22 > > fakeroot >= 1.12.2 > > qemu >= 0.10-7 > > I would like to use libguestfs in my distro, but is the dependency to > debootstrap mandatory? (as my distro is not Fedora?) Not sure if you misspelled 'febootstrap' there or if you really mean debootstrap. Anyhow it does need something like febootstrap/debootstrap, ie. something which can build appliances as non-root. So depending on your level of skill and free-software purity you have several choices: (1) Port febootstrap plus its deps to your distro. Approximately those deps would be: yum, python-rpm, fakeroot, fakechroot >= 2.9. or: (2) Port libguestfs to use debootstrap - if you're on Debian or Ubuntu or another Debian derivative this would make most sense. This is not as much effort as it sounds, since febootstrap is loosely based on debootstrap anyway. Also because we have debootstrap in Fedora-land, I can collaborate and test your patches. or: (3) Copy the binary blobs (the appliance) from /usr/lib64/guestfs in the Fedora RPMs: $ ll /usr/lib64/guestfs/ total 23968 -rw-r--r--. 1 root root 21869259 2009-04-30 20:35 initramfs.fedora-10.x86_64.img -rw-r--r--. 1 root root 2637056 2009-04-30 20:35 vmlinuz.fedora-10.x86_64 Choice (3) limits your freedom to modify the code and may have GPL implications, so check the licenses carefully. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From dennisml at conversis.de Fri May 1 12:34:22 2009 From: dennisml at conversis.de (Dennis J.) Date: Fri, 01 May 2009 14:34:22 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501083423.GA9933@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> Message-ID: <49FAEC4E.7020005@conversis.de> On 05/01/2009 10:34 AM, Richard W.M. Jones wrote: > On Fri, May 01, 2009 at 02:42:27AM +0200, Dennis J. wrote: >> Ah, I didn't know that libguestfs actually uses its own vm. >> I just tried compiling libguestfs from git on f10 but it doesn't seem to >> like that at all: >> >> ... >> mv initramfs.fedora-10.i686.img initramfs.fedora-10.i686.img.bak >> mv: cannot stat `initramfs.fedora-10.i686.img': No such file or directory >> make[2]: [initramfs/fakeroot.log] Error 1 (ignored) >> mv vmlinuz.fedora-10.i686 vmlinuz.fedora-10.i686.bak >> mv: cannot stat `vmlinuz.fedora-10.i686': No such file or directory >> make[2]: [initramfs/fakeroot.log] Error 1 (ignored) >> if ! bash ./make-initramfs.sh; then rm -f initramfs/fakeroot.log; exit 1; fi >> run_yum: line 1: 19874 Segmentation fault yum -y -c >> "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap >> --noplugins --nogpgcheck --installroot="$target" install "$@" >> chroot: cannot run command `chmod': No such file or directory >> make[2]: *** [initramfs/fakeroot.log] Error 1 >> make[2]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' >> make[1]: *** [all-recursive] Error 1 >> make[1]: Leaving directory `/home/dennis/Downloads/source/libguestfs-1.0.16' >> make: *** [all] Error 2 > > While it's possible to get libguestfs working on F10, it's not > our minimum target, which is F11. > > In particular on F10 you will need to get the F11/F12 versions of the > following packages: > > febootstrap>= 1.7 > fakechroot>= 2.9-22 > fakeroot>= 1.12.2 > qemu>= 0.10-7 I've rebuilt all of those and installed them: febootstrap-1.7-1.fc10.1.noarch fakechroot-2.9-22.fc10.i386 fakeroot-1.12.2-21.fc10.i386 qemu-0.10-15.fc10.i386 Also I'm using the brand-new libguestfs-1.0.18-1.fc11.src.rpm to try to rebuild it for f10 but I still get fatal errors: Lots of these: /usr/bin/pod2html: inspector/virt-inspector.pl: unknown pod directive 'encoding' in paragraph 5. ignoring. /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve L in paragraph 44. /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve L in paragraph 69. /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve L in paragraph 139. ... Then I see this: malloc: 4103338240: Cannot allocate memory DB_ENV->set_lk_detect: unknown deadlock detection mode specified DB_ENV->txn_stat_print interface requires an environment configured for the transaction subsystem run_yum: line 1: 5326 Segmentation fault yum -y -c "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap --noplugins --nogpgcheck --installroot="$target" install "$@" and finally this: chroot: cannot run command `chmod': No such file or directory Regards, Dennis From rjones at redhat.com Fri May 1 12:55:48 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 13:55:48 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <49FAEC4E.7020005@conversis.de> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> Message-ID: <20090501125548.GA10512@amd.home.annexia.org> On Fri, May 01, 2009 at 02:34:22PM +0200, Dennis J. wrote: > I've rebuilt all of those and installed them: > > febootstrap-1.7-1.fc10.1.noarch > fakechroot-2.9-22.fc10.i386 > fakeroot-1.12.2-21.fc10.i386 > qemu-0.10-15.fc10.i386 > > Also I'm using the brand-new libguestfs-1.0.18-1.fc11.src.rpm to try to > rebuild it for f10 but I still get fatal errors: > > Lots of these: > /usr/bin/pod2html: inspector/virt-inspector.pl: unknown pod directive > 'encoding' in paragraph 5. ignoring. > /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve > L in paragraph 44. > /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve > L in paragraph 69. > /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve > L in paragraph 139. Ignore these, they're just documentation warnings. > Then I see this: > malloc: 4103338240: Cannot allocate memory > DB_ENV->set_lk_detect: unknown deadlock detection mode specified > DB_ENV->txn_stat_print interface requires an environment configured for > the transaction subsystem These seem to be from the Python BerkleyDB library. Trying to allocate 4 GB of memory, nice one Python. I don't have a Fedora 10 i386 box to test against, but my suggestion is to upgrade to the latest Yum (especially) and Python (if possible) from F11/F12. This is what I'm using on F11 i386: $ rpm -q python yum python-2.6-7.fc11.i586 yum-3.2.22-4.fc11.noarch [...] > and finally this: > chroot: cannot run command `chmod': No such file or directory This is a consequence of the earlier failure of yum to do anything useful. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From ondrejj at salstar.sk Fri May 1 13:17:53 2009 From: ondrejj at salstar.sk (=?utf-8?B?SsOhbiBPTkRSRUogKFNBTCk=?=) Date: Fri, 1 May 2009 15:17:53 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501125548.GA10512@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> Message-ID: <20090501131753.GB19880@salstar.sk> On Fri, May 01, 2009 at 01:55:48PM +0100, Richard W.M. Jones wrote: > On Fri, May 01, 2009 at 02:34:22PM +0200, Dennis J. wrote: > > I've rebuilt all of those and installed them: > > > > febootstrap-1.7-1.fc10.1.noarch This one still has CDPATH problems: https://admin.fedoraproject.org/updates/febootstrap-1.7-1.fc10.1 Can you please update this? > > fakechroot-2.9-22.fc10.i386 > > fakeroot-1.12.2-21.fc10.i386 > > qemu-0.10-15.fc10.i386 I have qemu-kvm-new copyed into ~/bin from qemu-system-x86-0.10-15 package, because after update to qemu-0.10 on my Fedora 10, my virtual machines corrupts images. Then: ./autogen.sh --with-java-home=no --with-qemu=qemu-kvm-new And now it's working for me. You don't need whole qemu>0.10, just qemu-kvm or qemu binary + another qemu installation. > > Then I see this: > > malloc: 4103338240: Cannot allocate memory > > DB_ENV->set_lk_detect: unknown deadlock detection mode specified > > DB_ENV->txn_stat_print interface requires an environment configured for > > the transaction subsystem > > These seem to be from the Python BerkleyDB library. Trying to > allocate 4 GB of memory, nice one Python. I have no similar problems on Fedora 10 i386. > I don't have a Fedora 10 i386 box to test against, but my suggestion > is to upgrade to the latest Yum (especially) and Python (if possible) > from F11/F12. May be you can install one as virtual machine. :) SAL From rjones at redhat.com Fri May 1 13:42:58 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 14:42:58 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501131753.GB19880@salstar.sk> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501131753.GB19880@salstar.sk> Message-ID: <20090501134258.GC10510@amd.home.annexia.org> On Fri, May 01, 2009 at 03:17:53PM +0200, J?n ONDREJ (SAL) wrote: > This one still has CDPATH problems: > https://admin.fedoraproject.org/updates/febootstrap-1.7-1.fc10.1 > Can you please update this? I pushed a new version which should fix any problems. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Fri May 1 13:49:25 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 14:49:25 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501125548.GA10512@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> Message-ID: <20090501134925.GB10512@amd.home.annexia.org> I just uploaded some Fedora 11 / i586 binary RPMs to the site: http://et.redhat.com/~rjones/libguestfs/files/ These _may_ work on Fedora 10, I haven't tried. They will certainly require qemu >= 0.10 from Fedora 11 though. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From dennisml at conversis.de Fri May 1 13:57:39 2009 From: dennisml at conversis.de (Dennis J.) Date: Fri, 01 May 2009 15:57:39 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501125548.GA10512@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> Message-ID: <49FAFFD3.9040104@conversis.de> On 05/01/2009 02:55 PM, Richard W.M. Jones wrote: > On Fri, May 01, 2009 at 02:34:22PM +0200, Dennis J. wrote: >> I've rebuilt all of those and installed them: >> >> febootstrap-1.7-1.fc10.1.noarch >> fakechroot-2.9-22.fc10.i386 >> fakeroot-1.12.2-21.fc10.i386 >> qemu-0.10-15.fc10.i386 >> >> Also I'm using the brand-new libguestfs-1.0.18-1.fc11.src.rpm to try to >> rebuild it for f10 but I still get fatal errors: >> >> Lots of these: >> /usr/bin/pod2html: inspector/virt-inspector.pl: unknown pod directive >> 'encoding' in paragraph 5. ignoring. >> /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve >> L in paragraph 44. >> /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve >> L in paragraph 69. >> /usr/bin/pod2html: inspector/virt-inspector.pl: cannot resolve >> L in paragraph 139. > > Ignore these, they're just documentation warnings. > >> Then I see this: >> malloc: 4103338240: Cannot allocate memory >> DB_ENV->set_lk_detect: unknown deadlock detection mode specified >> DB_ENV->txn_stat_print interface requires an environment configured for >> the transaction subsystem > > These seem to be from the Python BerkleyDB library. Trying to > allocate 4 GB of memory, nice one Python. > > I don't have a Fedora 10 i386 box to test against, but my suggestion > is to upgrade to the latest Yum (especially) and Python (if possible) > from F11/F12. > > This is what I'm using on F11 i386: > > $ rpm -q python yum > python-2.6-7.fc11.i586 > yum-3.2.22-4.fc11.noarch I updated yum but still get the segfault but the malloc error didn't reappear. Updating Python really isn't an option as that basically forces a distribution upgrade anyway due to its massive dependency tail. In the line: run_yum: line 1: 13054 Segmentation fault yum -y -c "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap --noplugins --nogpgcheck --installroot="$target" install "$@" what are the contents of "$tmpdir/febootstrap.repo"? I tried finding it in the build directory after the failure but (also run_yum) but I couldn't find them anywhere. I want to execute that command manually in order to track down the cause of the segfault. Regards, Dennis From rjones at redhat.com Fri May 1 14:09:25 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 15:09:25 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <49FAFFD3.9040104@conversis.de> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <49FAFFD3.9040104@conversis.de> Message-ID: <20090501140925.GA11304@amd.home.annexia.org> On Fri, May 01, 2009 at 03:57:39PM +0200, Dennis J. wrote: > In the line: > run_yum: line 1: 13054 Segmentation fault yum -y -c > "$tmpdir"/febootstrap.repo --disablerepo=* --enablerepo=febootstrap > --noplugins --nogpgcheck --installroot="$target" install "$@" > > what are the contents of "$tmpdir/febootstrap.repo"? I tried finding it > in the build directory after the failure but (also run_yum) but I > couldn't find them anywhere. I want to execute that command manually in > order to track down the cause of the segfault. Have a look at febootstrap (it's just a big shell script). febootstrap.repo will contain: [febootstrap] name=febootstrap $repo $arch failovermethod=priority enabled=1 gpgcheck=0 $proxy where $repo = fedora-10, $arch = i386 and $proxy will be empty in this case. 'run_yum' is a function in the febootstrap shell script. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From dennisml at conversis.de Fri May 1 14:16:23 2009 From: dennisml at conversis.de (Dennis J.) Date: Fri, 01 May 2009 16:16:23 +0200 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501134925.GB10512@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501134925.GB10512@amd.home.annexia.org> Message-ID: <49FB0437.9040103@conversis.de> On 05/01/2009 03:49 PM, Richard W.M. Jones wrote: > > I just uploaded some Fedora 11 / i586 binary RPMs to the site: > > http://et.redhat.com/~rjones/libguestfs/files/ > > These _may_ work on Fedora 10, I haven't tried. They will certainly > require qemu>= 0.10 from Fedora 11 though. This seems to work nicely for me on f10 so far, thanks! The grub-install command still doesn't work though: [root at nexus ~]# guestfish -a /var/lib/libvirt/images/test.img -m /dev/vg_mobile/lv_root qemu: loading initrd (0x15b0c16 bytes) at 0x000000000b53f000 Welcome to guestfish, the libguestfs filesystem interactive shell for editing virtual machine filesystems. Type: 'help' for help with commands 'quit' to quit the shell > cat /etc/redhat-release Fedora release 10.92 (Rawhide) > grub-install / /dev/sda libguestfs: error: grub-install: /dev/mapper/vg_mobile-lv_root does not have any corresponding BIOS drive. Regards, Dennis From rjones at redhat.com Fri May 1 14:21:38 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 15:21:38 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <49FB0437.9040103@conversis.de> References: <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501134925.GB10512@amd.home.annexia.org> <49FB0437.9040103@conversis.de> Message-ID: <20090501142138.GB11304@amd.home.annexia.org> On Fri, May 01, 2009 at 04:16:23PM +0200, Dennis J. wrote: > [root at nexus ~]# guestfish -a /var/lib/libvirt/images/test.img -m > /dev/vg_mobile/lv_root [...] > > grub-install / /dev/sda > libguestfs: error: grub-install: /dev/mapper/vg_mobile-lv_root does not > have any corresponding BIOS drive. Grub just doesn't work on LVs. You have to install it on a 'real' partition. This is why Fedora uses a separate /boot partition. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From rjones at redhat.com Fri May 1 15:36:10 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 16:36:10 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <20090501142138.GB11304@amd.home.annexia.org> References: <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501134925.GB10512@amd.home.annexia.org> <49FB0437.9040103@conversis.de> <20090501142138.GB11304@amd.home.annexia.org> Message-ID: <20090501153610.GC11304@amd.home.annexia.org> Here's an example where I've used a separate /boot partition and root is on an LV: -------------------------------------------------- /tmp/test.sh #!/bin/sh - guestfish < References: <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501134925.GB10512@amd.home.annexia.org> <49FB0437.9040103@conversis.de> <20090501142138.GB11304@amd.home.annexia.org> <20090501153610.GC11304@amd.home.annexia.org> Message-ID: <49FB3B87.20506@conversis.de> On 05/01/2009 05:36 PM, Richard W.M. Jones wrote: > Here's an example where I've used a separate /boot partition and root > is on an LV: > > -------------------------------------------------- /tmp/test.sh > #!/bin/sh - > > guestfish< alloc /tmp/test.img 500MB > run > sfdisk /dev/sda 0 0 0 ",10 ," > > echo Size of /dev/sda1: > blockdev-getsize64 /dev/sda1 > echo Size of /dev/sda2: > blockdev-getsize64 /dev/sda2 > > mkfs ext2 /dev/sda1 > pvcreate /dev/sda2 > vgcreate VG /dev/sda2 > lvcreate LV VG 400M > mkfs ext2 /dev/VG/LV > > mount /dev/VG/LV / > mkdir /boot > mount /dev/sda1 /boot > > grub-install / /dev/sda > I'm actually using the same setup but I forgot to mount the /boot partition. After doing that grub-install worked fine although I'm not sure why I got the BIOS error message simply because /boot wasn't mounted. Isn't the MBR written to the first sector of the disk and as a result independent of any particular partition? Why would the mounting of a partition have an influence on that? (Of course without the mount grub's boot files get written to the wrong place but that doesn't explain grub complaining about the device layout) Regards, Dennis From rjones at redhat.com Fri May 1 19:54:17 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 1 May 2009 20:54:17 +0100 Subject: [fedora-virt] P2V and grub bootloader In-Reply-To: <49FB3B87.20506@conversis.de> References: <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <49FAEC4E.7020005@conversis.de> <20090501125548.GA10512@amd.home.annexia.org> <20090501134925.GB10512@amd.home.annexia.org> <49FB0437.9040103@conversis.de> <20090501142138.GB11304@amd.home.annexia.org> <20090501153610.GC11304@amd.home.annexia.org> <49FB3B87.20506@conversis.de> Message-ID: <20090501195417.GA12986@amd.home.annexia.org> On Fri, May 01, 2009 at 08:12:23PM +0200, Dennis J. wrote: > I'm actually using the same setup but I forgot to mount the /boot > partition. After doing that grub-install worked fine although I'm not > sure why I got the BIOS error message simply because /boot wasn't > mounted. Isn't the MBR written to the first sector of the disk and as a > result independent of any particular partition? Why would the mounting of > a partition have an influence on that? (Of course without the mount > grub's boot files get written to the wrong place but that doesn't explain > grub complaining about the device layout) Well this is a question for a grub mailing list, but the basic reason is because the boot sector needs to know on which partition the first stage is located. The boot sector is very limited and has access to just BIOS disks and a simple MBR-based partition decoder. Thus the first stage must be on an MBR partition (not an LV) on a BIOS disk -- /boot or /dev/sda1 in this example. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From m.a.young at durham.ac.uk Sun May 3 14:15:31 2009 From: m.a.young at durham.ac.uk (M A Young) Date: Sun, 3 May 2009 15:15:31 +0100 (BST) Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels In-Reply-To: References: <20090328163545.GM31725@salstar.sk> <20090414143618.GH351@redhat.com> Message-ID: Another kernel is available at http://koji.fedoraproject.org/koji/taskinfo?taskID=1333417 and at the repository http://fedorapeople.org/~myoung/dom0/ This version (2.6.30-0.1.2.23.rc4.xendom0.fc12) is based on the xen-tip/master branch, and also contains a bit of experimentation as CONFIG_XEN_GNTDEV and CONFIG_XEN_PCI_PASSTHROUGH are now enabled, and I have done a bit of hacking to get CONFIG_CC_STACKPROTECTOR working. Michael Young From charles at dyfis.net Mon May 4 00:05:37 2009 From: charles at dyfis.net (Charles Duffy) Date: Sun, 03 May 2009 19:05:37 -0500 Subject: [fedora-virt] libguestfs on non-Fedora targets Message-ID: <49FE3151.5040600@dyfis.net> Howdy. Is there any intention of supporting libguestfs on non-Fedora targets (given a sufficiently new qemu or kvm)? As background: I have a bunch of in-house QA infrastructure, which happens to include my own, internally developed tools intended to address the same problems as libguestfs. (One tool, for FUSE-mounting from a guest, used serial ports rather than the vmchannel mechanism, a buildroot/uClibc embedded userland rather than a stripped Fedora, qemu-nbd+nbd-client proxied over serial for getting the block device into the guest, UML rather than qemu or kvm, and a patched ccgfs+FUSE for getting the guest's filesystem back out onto the host). Libguestfs is clearly more mature and flexible than my personal hackery, so I'm looking at migrating to it -- but right now, I'm having trouble building it for my VM servers' platform, which is effectively RHEL5 with kvm-85 and a modern host kernel. Specific issues follow: ---- futimens() isn't available on glibc 2.5, and do_touch() doesn't presently degrade gracefully to a less-capable call without it: guestfsd-file.o: In function `do_touch': /home/cduffy/public_git/libguestfs/daemon/file.c:49: undefined reference to `futimens' ---- It was necessary to explicitly set OCAMLC=no OCAMLFIND=no (bypassing build of the ocaml bits) to prevent the build from breaking early, apparently due to C interface changes between the expected OCaml and the ocaml-3.09.3-1.el5 which is present. [TODO: more detail] ---- make-initramfs had numerous complaints: + it expected /var/lib/yum to exist within the initramfs root, which it did not + library versions mismatched between host and guest: libtinfo.so.5 (from ncurses, needed by /bin/sh) isn't available on RHEL5 natively (easily worked around), and the glibc versions don't match (not easily worked around). + unless my PATH included /usr/sbin, builds failed with "/usr/bin/fakechroot: line 107: exec: chroot: not found" + rpm %pre scripts all failed due to the previously mentioned version mismatches. ---- I haven't gotten beyond the last of these issues on account of the glibc mismatch. Before I proceed -- is my goal reasonable, or am I better off sticking with my in-house solution for the time being? Thanks! From markmc at redhat.com Mon May 4 14:11:57 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Mon, 04 May 2009 15:11:57 +0100 Subject: [fedora-virt] does f11 preview fix the network install issue? In-Reply-To: References: Message-ID: <1241446317.24899.1.camel@blaa> On Tue, 2009-04-28 at 11:51 -0400, Robert P. J. Day wrote: > can anyone verify that the newly-unleashed f11 preview will no > longer try to install a virtual image over the network when you select > the DVD? thanks. https://bugzilla.redhat.com/496298 Nope, not fixed in the preview release - it was fixed with udev-141-3, but the preview release included udev-141-1 We need to make sure to test this as soon as the release candidate is composed, just to be sure it really, really is fixed. Cheers, Mark. From rpjday at crashcourse.ca Mon May 4 14:17:52 2009 From: rpjday at crashcourse.ca (Robert P. J. Day) Date: Mon, 4 May 2009 10:17:52 -0400 (EDT) Subject: [fedora-virt] does f11 preview fix the network install issue? In-Reply-To: <1241446317.24899.1.camel@blaa> References: <1241446317.24899.1.camel@blaa> Message-ID: On Mon, 4 May 2009, Mark McLoughlin wrote: > On Tue, 2009-04-28 at 11:51 -0400, Robert P. J. Day wrote: > > can anyone verify that the newly-unleashed f11 preview will no > > longer try to install a virtual image over the network when you > > select the DVD? thanks. > > https://bugzilla.redhat.com/496298 > > Nope, not fixed in the preview release - it was fixed with > udev-141-3, but the preview release included udev-141-1 > > We need to make sure to test this as soon as the release candidate > is composed, just to be sure it really, really is fixed. so there's no way to test this out here, is there? we need to wait for a release that contains udev-141-3 already. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA Linux Consulting, Training and Annoying Kernel Pedantry. Web page: http://crashcourse.ca Linked In: http://www.linkedin.com/in/rpjday Twitter: http://twitter.com/rpjday ======================================================================== From markmc at redhat.com Mon May 4 16:44:11 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Mon, 04 May 2009 17:44:11 +0100 Subject: [fedora-virt] Fedora virt status Message-ID: <1241455451.9593.1.camel@blaa> 2009-05-12 Compose & Stage Release Candidate (7 days) Truly we are in the final countdown to the Fedora 11 release. This coming Thursday (May 7th) we are holding a Fedora Test Day in an attempt to fully test all aspects of Fedora's virtualization support. More below details on how to join in the test day. Virt Test Day ============= On May 7th, the Fedora 11 Virtualization Test Day will take place on #fedora-qa (irc.freenode.net). http://www.redhat.com/archives/fedora-virt/2009-April/msg00287.html Please join us for the day on #fedora-qa (irc.freenode.net) to help with testing Fedora's virt tools, writing test cases, filing bugs and debugging issues. On the day, you'll have the opportunity to work closely with other Fedora developers and contributors to make a real impact on the quality of Fedora's virtualization features. In advance of the test day, a wiki page is being prepared listing the various areas which will be tested on the day and individual test cases for each of the test areas: https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization Currently, the Xen DomU test area has the most comprehensive set of test cases, but hopefully folks won't be long rectifying that! Fedora 12 ========= F12 tracker bugs have been created. See the wiki page for more details: https://fedoraproject.org/w/index.php?title=Virtualization_bugs The F12 blocker and target bugs depend on the F11 counterparts. The idea is that if we don't fix a bug in F11, it's automatically on the F12 tracker but we still have a list of bugs worth fixing in a post-GA F11 update. Also, dist-f12 now has builds of kvm-85 and libvirt-0.6.3. As soon as the F11 release has been composed, rawhide will switch to being composed from dist-f12 and these packages will appear in rawhide. FWN === Dale Bewley has contributed two more excellent virtualization sections to Fedora Weekly News: https://fedoraproject.org/wiki/FWN/Issue174#Virtualization https://fedoraproject.org/wiki/FWN/Issue173#Virtualization Schroedinger's VM ================= Jerry James experienced the same confusion many other people must have had when he went away leaving his kernel compile running in a VM and came back to find it wouldn't progress unless we was actually watching the VM window: http://www.redhat.com/archives/fedora-virt/2009-April/msg00207.html Is this expected? Do KVM-based VMs have to be looked at to do anything? If so, is there an option somewhere along the lines of RunEvenWhenIAmNotLookingAtYou=true? A much more rational explanation was forthcoming :-) http://www.redhat.com/archives/fedora-virt/2009-April/msg00212.html When using the SDL backend (i.e. instead of qemu-kvm -vnc), if you switch workspaces using Ctrl-Alt-Arrow, then SDL tries to grab the pointer and fails and keeps retrying until you switch back to the workspace. Fully gory details, here: https://bugzilla.redhat.com/480065 libguestfs ========== Rich Jones continued his progress with libguestfs and released version 1.0.0 followed by several more release. He also posted a number of interesting guestfish recipes, demonstrating the power of libguestfs: http://et.redhat.com/~rjones/libguestfs/recipes.html Dom0 Kernels ============ Michael Young posted some updated Dom0 kernel builds: http://www.redhat.com/archives/fedora-virt/2009-April/msg00243.html http://www.redhat.com/archives/fedora-virt/2009-May/msg00016.html Dan Berrange reported some success testing these kernels under Xen in a KVM guest. Bugs ==== DOOM-O-METER: 178 bugs open two weeks ago, 182 this week. Blockers: https://bugzilla.redhat.com/496298 F11 Beta :: under KVM anaconda detect cdrom media because udev doesn't probe for filesystem type Further progress was made on figuring out why anaconda fails to install packages from DVD media on KVM. The difference from baremetal turns out to be that qemu doesn't implement GPCMD_READ_DISC_INFO. https://bugzilla.redhat.com/493692 Creating new VM; SELinux prevents opening iso image of install media When starting a VM, sVirt re-labels a disk image with the svirt_image_t:MCS label so that only that VM may access the image. However, we weren't doing this for readonly images like ISOs. Dan Berrange and Dan Walsh worked together to come up with a patch to fix this - readonly images will now be re-labelled virt_content_t:s0 when the VM is started. https://bugzilla.redhat.com/497602 Virtualization group doesn't pull in qemu or kvm When the qemu package was split, we decided to have the Virtualization package group install qemu-kvm via a Provides on the qemu-system-x86 package. It turns out that yum groupinstall will only install packages, not rpm provides. To fix this, it was decided to make qemu-kvm a meta-package requiring the qemu-system-x86 package. https://bugzilla.redhat.com/497429 pwritev writes ramdom junk Very serious data corruptor which thankfully only affects latest development qemu upstream. Jakub already has a fix. https://bugzilla.redhat.com/492838 kvm: PTHREAD_PRIO_INHERIT mutexes fail to unlock sometimes (canberra-gtk-play crash locks desktop metacity bugbuddy) An odd issue whereby PTHREAD_PRIO_INHERIT mutexs fail to unlock on i386 KVM guests. Manifests itself as pulseaudio related desktop app segfaults. qemu: https://bugzilla.redhat.com/496642 qemu qcow2 image corruption Gleb pointed out another qcow2 corruption fix upstream and this was also included for F11. https://bugzilla.redhat.com/498405 qemu disk image corruption with qcow2/virtio QCOW2 image corruption seen with latest Fedora 11 and virtio-blk. Also reproducible with the latest upstream qemu-kvm stable branch. https://bugzilla.redhat.com/497170 qemu segfaults with "raw_aio_remove: aio request not found!" Rich Jones discovered a qemu-kvm segv when using read-only images. Patches have been identified upstream and ported to the stable branch. We'll pull them into an F12 update in due course. https://bugzilla.redhat.com/498242 QEMU should set CLONE_IO flag when creating any threads Dan Berrange suggests that qemu should use CLONE_IO for each of its threads so that all qemu threads share the same I/O priority so that ionice can be used with qemu. https://bugzilla.redhat.com/497341 Make the /dev/kvm device world accessible to all users by default After some discussion with the KVM maintainers, we've concluded that we should make /dev/kvm world accessible by default in F12. https://bugzilla.redhat.com/496627 qemu-kvm es1370 sound card emulation segfault https://bugzilla.redhat.com/495964 qemu locks up at shutdown with sdl audio driver Debugging issues with qemu's various sound backends continues. It looks like the pulseaudio backend might work better for people. libvirt: https://bugzilla.redhat.com/460649 libvirtd requires restart in order to detect new capabilities like KVM support Cole has come up with a patch to have libvirtd automatically detect if qemu-kvm is installed. Hopefully we'll have this patch in before the F12 release. https://bugzilla.redhat.com/474182 LVs created in an existing VG have wrong SELinux label https://bugzilla.redhat.com/497131 svirt fails to relabel qcow2 backing files sVirt re-labels images before it starts a guest, but it doesn't re-label qcow2 backing files. https://bugzilla.redhat.com/496945 libvirt should own /var/cache/libvirt in spec file libvirt not owning this directory causes it to be mislabelled which, in turn, causes virDomainPeek() to fail. virtinst: https://bugzilla.redhat.com/496340 Creating guests where the .iso image is on a NFS share doesn't work anymore Much discussion on the what svirt needs to do to support images on NFS shares. Cole resolved this by only warning if selinux labeling appears to be wrong. grub: https://bugzilla.redhat.com/479760 [PATCH] add support for virtio_blk devices to grub-install Jan ONDREJ came across this problem when testing the preview release and prompted Peter Jones to commit this patch. https://bugzilla.redhat.com/498416 Unable to boot Fedora11-Preview Jan also reports that on his F10 system, with F11 virt packages, he could boot a newly installed F11 Beta guest but not an F11 Preview guest. It looks most likely to be a grub issue, but this has not been reported on an F10 host yet. etherboot: https://bugzilla.redhat.com/498230 virtio PXE Roms "No IP address" Another report of etherboot PXE ROMs failing, but this time it seems that virtio is failing as well as e1000. xen: https://bugzilla.redhat.com/493103 Network periodically hangs during install of xen guest This problem has magically been resolved. From markmc at redhat.com Mon May 4 16:59:45 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Mon, 04 May 2009 17:59:45 +0100 Subject: [fedora-virt] does f11 preview fix the network install issue? In-Reply-To: References: <1241446317.24899.1.camel@blaa> Message-ID: <1241456385.9593.4.camel@blaa> On Mon, 2009-05-04 at 10:17 -0400, Robert P. J. Day wrote: > On Mon, 4 May 2009, Mark McLoughlin wrote: > > > On Tue, 2009-04-28 at 11:51 -0400, Robert P. J. Day wrote: > > > can anyone verify that the newly-unleashed f11 preview will no > > > longer try to install a virtual image over the network when you > > > select the DVD? thanks. > > > > https://bugzilla.redhat.com/496298 > > > > Nope, not fixed in the preview release - it was fixed with > > udev-141-3, but the preview release included udev-141-1 > > > > We need to make sure to test this as soon as the release candidate > > is composed, just to be sure it really, really is fixed. > > so there's no way to test this out here, is there? we need to wait > for a release that contains udev-141-3 already. You could try building an ISO using pungi. Cheers, Mark. From rpjday at crashcourse.ca Mon May 4 17:24:18 2009 From: rpjday at crashcourse.ca (Robert P. J. Day) Date: Mon, 4 May 2009 13:24:18 -0400 (EDT) Subject: [fedora-virt] does f11 preview fix the network install issue? In-Reply-To: <1241456385.9593.4.camel@blaa> References: <1241446317.24899.1.camel@blaa> <1241456385.9593.4.camel@blaa> Message-ID: On Mon, 4 May 2009, Mark McLoughlin wrote: > On Mon, 2009-05-04 at 10:17 -0400, Robert P. J. Day wrote: > > On Mon, 4 May 2009, Mark McLoughlin wrote: > > > > > On Tue, 2009-04-28 at 11:51 -0400, Robert P. J. Day wrote: > > > > can anyone verify that the newly-unleashed f11 preview will no > > > > longer try to install a virtual image over the network when you > > > > select the DVD? thanks. > > > > > > https://bugzilla.redhat.com/496298 > > > > > > Nope, not fixed in the preview release - it was fixed with > > > udev-141-3, but the preview release included udev-141-1 > > > > > > We need to make sure to test this as soon as the release candidate > > > is composed, just to be sure it really, really is fixed. > > > > so there's no way to test this out here, is there? we need to wait > > for a release that contains udev-141-3 already. > > You could try building an ISO using pungi. i hate to sound lazy but if someone else wanted to build that image and make it available somewhere, i'd be happy to test it, but i really don't have the time to figure out pungi and build it myself. it *was* useful to know that the preview still has the network install issue, so i know there's little point for me to test the preview with respect to virtualization. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA Linux Consulting, Training and Annoying Kernel Pedantry. Web page: http://crashcourse.ca Linked In: http://www.linkedin.com/in/rpjday Twitter: http://twitter.com/rpjday ======================================================================== From jlaska at redhat.com Tue May 5 20:13:22 2009 From: jlaska at redhat.com (James Laska) Date: Tue, 05 May 2009 16:13:22 -0400 Subject: [fedora-virt] Virt Test Day - Graphics console test case review Message-ID: <1241554402.2466.96.camel@flatline.devel.redhat.com> Greetings folks, I've stubbed out a few test cases for the Fedora11 ImprovedVirtConsole feature [1]. I can't help but think more is needed, but I'm experiencing writers block. What does the group think? https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization_Graphical_Console * Should there be a test to confirm that older os-types continue to have the absolute pointer? * Should the tests explicitly try different virt types (e.g. KVM, qemu and xen)? * What else needs to be confirmed with this feature? Thanks, James -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part URL: From markmc at redhat.com Wed May 6 07:33:15 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 06 May 2009 08:33:15 +0100 Subject: [fedora-virt] Virt Test Day - Graphics console test case review In-Reply-To: <1241554402.2466.96.camel@flatline.devel.redhat.com> References: <1241554402.2466.96.camel@flatline.devel.redhat.com> Message-ID: <1241595195.12673.2.camel@blaa> On Tue, 2009-05-05 at 16:13 -0400, James Laska wrote: > Greetings folks, > > I've stubbed out a few test cases for the Fedora11 ImprovedVirtConsole > feature [1]. I can't help but think more is needed, but I'm > experiencing writers block. > > What does the group think? > > https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization_Graphical_Console I think you've got it covered. There's not much testing to do for it and it's all about checking that it "just works". Cheers, Mark. From markmc at redhat.com Wed May 6 07:39:32 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 06 May 2009 08:39:32 +0100 Subject: [fedora-virt] Virt Test Day - PCI device assignment test cases Message-ID: <1241595572.12673.9.camel@blaa> Hi, I've posted some test cases for testing KVM PCI device assignment: https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization_KVM_PCI_Device_Assignment The test cases are: 1. libvirt nodedev operations 2. assigning a device using libvirt 3. assigning a device using virt-manager 4. virt-install --host-device Note the "known bugs" section. It seems this feature has acquired some blocker bugs. Cheers, Mark From jlaska at redhat.com Wed May 6 11:33:57 2009 From: jlaska at redhat.com (James Laska) Date: Wed, 06 May 2009 07:33:57 -0400 Subject: [fedora-virt] Virt Test Day - PCI device assignment test cases In-Reply-To: <1241595572.12673.9.camel@blaa> References: <1241595572.12673.9.camel@blaa> Message-ID: <1241609637.2370.16.camel@flatline.devel.redhat.com> On Wed, 2009-05-06 at 08:39 +0100, Mark McLoughlin wrote: > Hi, > I've posted some test cases for testing KVM PCI device assignment: > > https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization_KVM_PCI_Device_Assignment > > The test cases are: Well written test cases. Thank you! > 1. libvirt nodedev operations The first step lists "verify the correct nodedev name" ... what should the correct name be? Is this supposed to match the lspci XXXX:YYYY string for the device? > 2. assigning a device using libvirt > 3. assigning a device using virt-manager > 4. virt-install --host-device I've gone ahead and made some minor wiki formatting tweaks to several cases. > Note the "known bugs" section. It seems this feature has acquired some > blocker bugs. Great to have the 'known bugs' section for users/testers before they dive in. Thanks, James -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part URL: From rjones at redhat.com Wed May 6 12:02:16 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 13:02:16 +0100 Subject: [fedora-virt] libguestfs on non-Fedora targets In-Reply-To: <49FE3151.5040600@dyfis.net> References: <49FE3151.5040600@dyfis.net> Message-ID: <20090506120216.GA28900@amd.home.annexia.org> On Sun, May 03, 2009 at 07:05:37PM -0500, Charles Duffy wrote: > Howdy. Is there any intention of supporting libguestfs on non-Fedora > targets (given a sufficiently new qemu or kvm)? First of all, sorry for the delay in answering (I'm currently on holiday). Secondly, thank you for your detailed and specific bug reports. Many of these things are simply bugs in libguestfs and/or febootstrap. We don't have a bug tracker for libguestfs yet, and so I will add these to the top level 'BUGS' file in the git repo (shortly). > As background: I have a bunch of in-house QA infrastructure, which > happens to include my own, internally developed tools intended to > address the same problems as libguestfs. (One tool, for FUSE-mounting > from a guest, used serial ports rather than the vmchannel mechanism, a > buildroot/uClibc embedded userland rather than a stripped Fedora, > qemu-nbd+nbd-client proxied over serial for getting the block device > into the guest, UML rather than qemu or kvm, and a patched ccgfs+FUSE > for getting the guest's filesystem back out onto the host). Yes, this is certainly an approach that we looked at. The reasons why we chose to not do this are summarised on my blog here: http://rwmj.wordpress.com/2009/03/20/why-not-use-a-minimal-distribution/ > Libguestfs is clearly more mature and flexible than my personal hackery, > so I'm looking at migrating to it -- but right now, I'm having trouble > building it for my VM servers' platform, which is effectively RHEL5 with > kvm-85 and a modern host kernel. Specific issues follow: > > ---- > > futimens() isn't available on glibc 2.5, and do_touch() doesn't > presently degrade gracefully to a less-capable call without it: Yes, this is obviously a bug in libguestfs. Noted. > guestfsd-file.o: In function `do_touch': > /home/cduffy/public_git/libguestfs/daemon/file.c:49: undefined reference > to `futimens' > > ---- > > It was necessary to explicitly set OCAMLC=no OCAMLFIND=no (bypassing > build of the ocaml bits) to prevent the build from breaking early, > apparently due to C interface changes between the expected OCaml and the > ocaml-3.09.3-1.el5 which is present. [TODO: more detail] Also a bug in libguestfs. Noted. We shouldn't depend on anything in OCaml 3.10, but since I haven't tested it with any earlier versions, there is some hidden assumption somewhere. > ---- > > make-initramfs had numerous complaints: > + it expected /var/lib/yum to exist within the initramfs root, which it > did not A bug in febootstrap. Noted as http://bugzilla.redhat.com/499369 > + library versions mismatched between host and guest: libtinfo.so.5 > (from ncurses, needed by /bin/sh) isn't available on RHEL5 natively > (easily worked around), and the glibc versions don't match (not easily > worked around). The background to this one is that we build an appliance with one Fedora distro (eg. Fedora 10), and we also build on a possibly different platform (eg. CentOS 5). Copying programs built on the host (eg. guestfsd) to the appliance is problematic. This appears to be a similar, but different problem, with febootstrap / yum. I'm a little confused actually because there shouldn't be a mismatch possible here. Anyhow, when building libguestfs in Fedora, we avoid any problems by making sure the appliance is built from precisely the same RPMs as the host distribution. So copying guestfsd from the host to the appliance should not present any problem. When building on CentOS 5.3, I intend that the appliance should also be CentOS 5.3. > + unless my PATH included /usr/sbin, builds failed with > "/usr/bin/fakechroot: line 107: exec: chroot: not found" I suspect a bug in fakechroot, but I will look into this one in more detail later today. > + rpm %pre scripts all failed due to the previously mentioned version > mismatches. > > ---- > > I haven't gotten beyond the last of these issues on account of the glibc > mismatch. Before I proceed -- is my goal reasonable, or am I better off > sticking with my in-house solution for the time being? Well I hope you can use libguestfs of course! Anyhow, I am currently building, or trying to build, libguestfs and the dependencies for CentOS 5.3 (i386). I will follow up with notes as things proceed. Thanks again for the bug reports. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From jlaska at redhat.com Wed May 6 12:08:29 2009 From: jlaska at redhat.com (James Laska) Date: Wed, 06 May 2009 08:08:29 -0400 Subject: [fedora-virt] Virt Test Day - Added doc links Message-ID: <1241611709.2370.33.camel@flatline.devel.redhat.com> Greetings, Just a heads up, I've added a documentation section to the Virtualization test day wiki. https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization#Documentation Thanks to Rahul Sundaram for the "getting started" link. Please feel free to add any introductory/getting_started docs to this section. Thanks, James -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part URL: From rjones at redhat.com Wed May 6 12:40:12 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 13:40:12 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <49FE3151.5040600@dyfis.net> References: <49FE3151.5040600@dyfis.net> Message-ID: <20090506124012.GB28900@amd.home.annexia.org> Just following up here with my notes on building libguestfs and dependencies on RHEL 5 and derived distributions. In fact this build is being done on a CentOS 5.3 / i386 machine. ---------------------------------------------------------------------- (1) I note that fakeroot and fakechroot in EPEL are ancient history. Therefore pull fakeroot and fakechroot source RPMs from Rawhide, and use those instead. fakeroot 1.12.2-21 fakechroot 2.9-20 (2) Neither SRPM can be rebuilt directly because of the MD5/SHA change that happened in RPM. Therefore I have to unpack them by hand using `rpm2cpio'. (3) fakeroot builds with no problem. (4) fakechroot requires a small patch to autogen.sh in order to build on CentOS [attached]. After building, all binary RPMs are installed. (5) Build febootstrap from git repo. (6) Test febootstrap: ./febootstrap fedora-10 f10 Many scriptlets fail because of a glibc version problem. I don't understand why it's using the host glibc and not the appliance / fakechrooted glibc, so this may be a bug in fakechroot or a problem with fakechroot on this older glibc. ./febootstrap centos-5 c5 http://mirror.centos.org/centos-5/5.3/os/i386/ [More to come in a followup posting] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top -------------- next part -------------- --- fakechroot-2.9.orig/autogen.sh 2009-03-19 14:43:19.000000000 +0000 +++ fakechroot-2.9/autogen.sh 2009-05-06 12:33:20.000000000 +0100 @@ -13,10 +13,10 @@ rm -f depcomp install-sh libtool ltmain.sh missing stamp-h1 rm -rf autom4te.cache - aclocal-${automake_version} "$@" + aclocal "$@" autoheader libtoolize --force --copy - automake-${automake_version} --add-missing --copy + automake --add-missing --copy autoconf rm -rf autom4te.cache From rjones at redhat.com Wed May 6 13:32:33 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 14:32:33 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506124012.GB28900@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> Message-ID: <20090506133233.GA2311@amd.home.annexia.org> On Wed, May 06, 2009 at 01:40:12PM +0100, Richard W.M. Jones wrote: > ./febootstrap centos-5 c5 http://mirror.centos.org/centos-5/5.3/os/i386/ This works perfectly. Barring the known problems with the version of fakechroot being used, such as error "/etc/ld.so.cache~: Permission denied" which are all non-fatal. (7) Size of centos-5 @Core febootstrapped installation: $ du -sh c5 414M c5 (8) Proceed to building libguestfs. I had to install a few dependencies such as OCaml, java-devel, ruby-devel, perl-devel, python-devel, etc. (9) Build libguestfs from git repo. Since I haven't installed febootstrap, I need to adjust the PATH to use my compiled-but-not- installed version. Also disable the qemu vmchannel test since we'll be using a custom-built qemu later. We also have to set the appliance parameters so that we end up with a CentOS 5.3-derived appliance. PATH=../febootstrap:$PATH \ vmchannel_test=no \ ./autogen.sh \ --with-repo=centos-5 \ --with-mirror=http://mirror.centos.org/centos-5/5.3/os/i386/ \ --with-updates=none [More to follow ...] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Wed May 6 13:51:12 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 14:51:12 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506133233.GA2311@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> <20090506133233.GA2311@amd.home.annexia.org> Message-ID: <20090506135112.GB2311@amd.home.annexia.org> On Wed, May 06, 2009 at 02:32:33PM +0100, Richard W.M. Jones wrote: > PATH=../febootstrap:$PATH \ > vmchannel_test=no \ > ./autogen.sh \ > --with-repo=centos-5 \ > --with-mirror=http://mirror.centos.org/centos-5/5.3/os/i386/ \ > --with-updates=none (10) Forgot about the Perl modules, more OCaml stuff, rake and Augeas, so I install those from EPEL (perl-Test-Pod, perl-Test-Pod-Coverage, augeas-devel, rubygem-rake, ocaml-ocamldoc, ocaml-findlib-devel) perl-Sys-Virt isn't available in CentOS or EPEL. Yet. It will be in RHEL 5.4. For now we won't be able to build virt-inspector. (11) Try the above configure command again ... it works. (12) Notice that there's a bug in the configure script which means that even though there are Perl modules missing which are needed for virt-inspector, it still configures that component. Bug noted. [More to follow] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Wed May 6 14:28:36 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 15:28:36 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506135112.GB2311@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> <20090506133233.GA2311@amd.home.annexia.org> <20090506135112.GB2311@amd.home.annexia.org> Message-ID: <20090506142836.GC2311@amd.home.annexia.org> (13) make (14) Warning about -fstrict-aliasing in guestfs_get_default_main_loop. I'm certain this can be ignored. (15) Lots of warnings about -fstrict-aliasing in rpcgen-generated code. Ignore these for now. We had the same errors in libvirt for a long time with no apparent effect on code generation. (16) Missing futimens function in do_touch (previously reported bug). Fixed. (17) fish.c: Implicit declaration of `isspace'. Fixed. (18) In OCaml bindings, implicit declaration of `CAMLreturnT'. Fixed. (19) In Ruby bindings, implicit declaration of `RARRAY_LEN'. Fixed. [More to come ...] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Wed May 6 15:21:44 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 16:21:44 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506142836.GC2311@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> <20090506133233.GA2311@amd.home.annexia.org> <20090506135112.GB2311@amd.home.annexia.org> <20090506142836.GC2311@amd.home.annexia.org> Message-ID: <20090506152144.GD2311@amd.home.annexia.org> On Wed, May 06, 2009 at 03:28:36PM +0100, Richard W.M. Jones wrote: > (13) make (20) Get the PATH right so we pick up febootstrap: PATH=../febootstrap:$PATH make (21) "No package augeas-libs available". /me grumbles about package name changes. You have to change `-i augeas-libs' to `-i augeas' in make-initramfs.sh and make-initramfs.sh.in. Unfortunately this change is only applicable to RHEL 5, and cannot be made in libguestfs upstream. [See attached patch] (22) "exec: chroot not found" To workaround this temporarily I do: PATH=../febootstrap:/usr/sbin:$PATH make [That is building away now, more to follow] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top -------------- next part -------------- diff --git a/make-initramfs.sh.in b/make-initramfs.sh.in index 2e3befb..7812e4e 100755 --- a/make-initramfs.sh.in +++ b/make-initramfs.sh.in @@ -23,7 +23,7 @@ set -e unset CDPATH modules=" --i augeas-libs +-i augeas -i bash -i coreutils -i dosfstools From rjones at redhat.com Wed May 6 15:33:56 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 16:33:56 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506152144.GD2311@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> <20090506133233.GA2311@amd.home.annexia.org> <20090506135112.GB2311@amd.home.annexia.org> <20090506142836.GC2311@amd.home.annexia.org> <20090506152144.GD2311@amd.home.annexia.org> Message-ID: <20090506153355.GE2311@amd.home.annexia.org> (23) Unfortunately there's no augeas at all in CentOS 5.3 base distribution. We should really make support for Augeas completely optional, since a lot of the time those aug_* commands won't be needed by end users. I don't have the stomach to make this change right now, since I'm doing this on a green-screen terminal without even copy and paste never mind such luxuries as the "WorldWideWeb" ... I hacked around this for now, but noted the problem in the TODO file. [More to follow] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Wed May 6 16:17:31 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 17:17:31 +0100 Subject: [fedora-virt] libguestfs on non-Fedora targets In-Reply-To: <20090506120216.GA28900@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506120216.GA28900@amd.home.annexia.org> Message-ID: <20090506161731.GC28900@amd.home.annexia.org> On Wed, May 06, 2009 at 01:02:16PM +0100, Richard W.M. Jones wrote: > > make-initramfs had numerous complaints: > > + it expected /var/lib/yum to exist within the initramfs root, which it > > did not > > A bug in febootstrap. Noted as http://bugzilla.redhat.com/499369 FYI I couldn't reproduce this one. Yum from CentOS 5.3 works fine for me, modulo the other bugs which are now fixed also. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Wed May 6 16:38:51 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 6 May 2009 17:38:51 +0100 Subject: Notes on building libguestfs on RHEL 5 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <20090506153355.GE2311@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090506124012.GB28900@amd.home.annexia.org> <20090506133233.GA2311@amd.home.annexia.org> <20090506135112.GB2311@amd.home.annexia.org> <20090506142836.GC2311@amd.home.annexia.org> <20090506152144.GD2311@amd.home.annexia.org> <20090506153355.GE2311@amd.home.annexia.org> Message-ID: <20090506163851.GF2311@amd.home.annexia.org> [Continuing to build ...] (24) febootstrap-minimize program is not removing locale files, which made the appliance image about 75MB larger than it needed to be. Small patch added to git repo to fix this. (25) Meanwhile, building a custom qemu from upstream qemu git. I am using the attached wrapper script, and I will invoke libguestfs programs first setting this environment variable: LIBGUESTFS_QEMU=/path/to/qemu.wrapper including running the testsuite like this: LIBGUESTFS_QEMU=/path/to/qemu.wrapper make check and guestfish w/o installing like this: LIBGUESTFS_QEMU=/path/to/qemu.wrapper ./fish/guestfish [More to follow ...] -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ -------------- next part -------------- #!/bin/sh - qemudir=/home/rjones/d/qemu exec $qemudir/i386-softmmu/qemu -L $qemudir/pc-bios "$@" From pasik at iki.fi Wed May 6 19:00:21 2009 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Wed, 6 May 2009 22:00:21 +0300 Subject: [Fedora-xen] Re: [fedora-virt] Re: Dom0 kernels / VGA text console works now In-Reply-To: <49C746F0.3050704@redhat.com> References: <49C0B8BC.1050301@redhat.com> <20090322115432.GX15052@edu.joroinen.fi> <49C746F0.3050704@redhat.com> Message-ID: <20090506190021.GZ24960@edu.joroinen.fi> On Mon, Mar 23, 2009 at 09:23:12AM +0100, Gerd Hoffmann wrote: > Pasi K?rkk?inen wrote: > > On Wed, Mar 18, 2009 at 10:02:52AM +0100, Gerd Hoffmann wrote: > >> M A Young wrote: > >>> (with nomodeset I get no text characters during > >>> boot - only a cursor that moves as invisible characters are presumably > >>> printed to the screen > >> known issue with vga text mode (32bit only, 64bit works ok). vga window > >> at 0xc0000 isn't mapped correctly. Workaround: drive the card in gfx > >> mode. Either kernel mode setting or vesafb (use something like > >> vga=gfx-1024x768x16 on the *xen* kernel command line). > > > > So the reason for this issue is known.. any patches available? > > According to jeremy someone works on unifying 32bit and 64bit page table > initalization, which should nicely fix this issue as a side effect. > VGA text console works OK for me now, on 32bit PAE pv_ops dom0, using the latest xen-tip/next tree! -- Pasi From pasik at iki.fi Wed May 6 19:31:50 2009 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Wed, 6 May 2009 22:31:50 +0300 Subject: [fedora-virt] Fedora 11 (rawhide) success with Xen pv_ops dom0 kernel (2.6.30-rc3) Message-ID: <20090506193150.GA24960@edu.joroinen.fi> Hello! Earlier I was testing Xen pv_ops dom0 kernel with Fedora 10, and after I got it working I decided to repeat the tests with Fedora 11 (rawhide). There was problems with the latest pv_ops dom0 xen-tip/next kernel tree for a while, but those problems are now resolved, and I got my Fedora 11 testbox working, so I decided to write here about my success. My setup: - Fedora 11 (rawhide as of 2009-05-05) - Xen included in F11, no external patches (xen-3.3.1-11.fc11) - xen-tip/next pv_ops dom0 kernel as of 2009-05-06, Linux 2.6.30-rc3. - All the rest was standard stuff included in Fedora 11 aswell With this setup I'm able to run Xen paravirtual (PV) domUs, install new domUs using virt-install and also install and manage domains with virt-manager. Current limitations: - Xen HVM support is not yet working with pv_ops dom0 kernel Earlier problems with VGA text console have also been fixed, so grub menu.lst entry like this now works: title Fedora Xen pv_ops dom0-test (2.6.30-rc3-tip) root (hd0,0) kernel /xen-3.3.gz dom0_mem=1024M loglvl=all guest_loglvl=all module /vmlinuz-2.6.30-rc3-tip ro root=/dev/vg00/lv01 module /initrd-2.6.30-rc3-tip.img Or if you want to have serial console, use this: title Fedora Xen pv_ops dom0-test (2.6.30-rc3-tip) / serial console root (hd0,0) kernel /xen-3.3.gz dom0_mem=1024M loglvl=all guest_loglvl=all com1=19200,8n1 console=com1 module /vmlinuz-2.6.30-rc3-tip ro root=/dev/vg00/lv01 console=hvc0 earlyprintk=xen module /initrd-2.6.30-rc3-tip.img Hopefully someone finds this information interesting! :) -- Pasi From markmc at redhat.com Thu May 7 06:30:31 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Thu, 07 May 2009 07:30:31 +0100 Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <1241125048.1335.24.camel@blaa> References: <1241125048.1335.24.camel@blaa> Message-ID: <1241677831.7718.7.camel@blaa> On Thu, 2009-04-30 at 21:57 +0100, Mark McLoughlin wrote: > Hey, > The F11 Test Day for virtualization is next week (Thurs May 7). > > https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization > > Please join us for the day on #fedora-qa (irc.freenode.net) to help > with testing Fedora's virt tools, writing test cases, filing bugs and > debugging issues. Just a reminder to folks - today is test day! :-) Cheers, Mark. From markmc at redhat.com Thu May 7 06:52:15 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Thu, 07 May 2009 07:52:15 +0100 Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <1241677831.7718.7.camel@blaa> References: <1241125048.1335.24.camel@blaa> <1241677831.7718.7.camel@blaa> Message-ID: <1241679135.7718.9.camel@blaa> On Thu, 2009-05-07 at 07:30 +0100, Mark McLoughlin wrote: > On Thu, 2009-04-30 at 21:57 +0100, Mark McLoughlin wrote: > > Hey, > > The F11 Test Day for virtualization is next week (Thurs May 7). > > > > https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization > > > > Please join us for the day on #fedora-qa (irc.freenode.net) to help > > with testing Fedora's virt tools, writing test cases, filing bugs and > > debugging issues. > > Just a reminder to folks - today is test day! :-) If folks would like to follow bugzilla activity today, add fedora-virt-maint at redhat.com to your bugzilla watchlist here: https://bugzilla.redhat.com/userprefs.cgi?tab=email You can just remove it again when you're bored :-) Cheers, Mark. From dlbewley at lib.ucdavis.edu Thu May 7 07:05:10 2009 From: dlbewley at lib.ucdavis.edu (Dale Bewley) Date: Thu, 7 May 2009 00:05:10 -0700 (PDT) Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <14451226.571241679794980.JavaMail.dlbewley@seitan.home.bewley.net> Message-ID: <8655327.591241679909641.JavaMail.dlbewley@seitan.home.bewley.net> Already been poking![1] But now it's bed time. Hopefully I can do some more testing tomorrow. [1] http://fedoraproject.org/wiki/User:Dale/Test_Day:2009-05-07_Virtualization:XenDomU -- Dale Bewley - Unix Administrator - Shields Library - UC Davis GPG: 0xB098A0F3 0D5A 9AEB 43F4 F84C 7EFD 1753 064D 2583 B098 A0F3 ----- "Mark McLoughlin" wrote: > On Thu, 2009-04-30 at 21:57 +0100, Mark McLoughlin wrote: > > Hey, > > The F11 Test Day for virtualization is next week (Thurs May 7). > > > > https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization > > > > Please join us for the day on #fedora-qa (irc.freenode.net) to > help > > with testing Fedora's virt tools, writing test cases, filing bugs > and > > debugging issues. > > Just a reminder to folks - today is test day! :-) > > Cheers, > Mark. > > _______________________________________________ > Fedora-virt mailing list > Fedora-virt at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-virt From clalance at redhat.com Thu May 7 10:10:34 2009 From: clalance at redhat.com (Chris Lalancette) Date: Thu, 07 May 2009 12:10:34 +0200 Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <8655327.591241679909641.JavaMail.dlbewley@seitan.home.bewley.net> References: <8655327.591241679909641.JavaMail.dlbewley@seitan.home.bewley.net> Message-ID: <4A02B39A.6060307@redhat.com> Dale Bewley wrote: > Already been poking![1] But now it's bed time. Hopefully I can do some more testing tomorrow. > > [1] http://fedoraproject.org/wiki/User:Dale/Test_Day:2009-05-07_Virtualization:XenDomU Thanks a lot for starting in on this! Quick question about your ballooning results; did you remember to balloon down the dom0 before trying to balloon up your guest? It doesn't do it automatically on when you run the "virsh setmem" command. If you did balloon your dom0 down first, then this might be a bug we have to file. I'm going to try this out shortly on my own, and I'll see what results I get. -- Chris Lalancette From dlbewley at lib.ucdavis.edu Thu May 7 14:52:54 2009 From: dlbewley at lib.ucdavis.edu (Dale Bewley) Date: Thu, 7 May 2009 07:52:54 -0700 (PDT) Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <22323652.651241707152302.JavaMail.dlbewley@seitan.home.bewley.net> Message-ID: <8409713.671241707971921.JavaMail.dlbewley@seitan.home.bewley.net> ----- "Chris Lalancette" wrote: > Dale Bewley wrote: > > Already been poking![1] But now it's bed time. Hopefully I can do > some more testing tomorrow. > > > > [1] > http://fedoraproject.org/wiki/User:Dale/Test_Day:2009-05-07_Virtualization:XenDomU > > Thanks a lot for starting in on this! Quick question about your > ballooning > results; did you remember to balloon down the dom0 before trying to > balloon up > your guest? It doesn't do it automatically on when you run the "virsh > setmem" > command. Actually I did not modify the dom0. Can you take a peek at the discussion page[1]? Perhaps the test case could be clarified a little bit. Is step 1 run in the dom0? Is step 2 modifying xend to affect dom0 or a guest config? I assumed the latter but guest configs aren't there anymore. Just in case, I did try shrinking Domain-0 and testing again. [root at felix tmp]# virsh dominfo Domain-0 ... Max memory: no limit Used memory: 5181440 kB [root at felix tmp]# virst setmem Domain-0 $(( 1024 * 3000 )) [root at felix tmp]# virsh dominfo Domain-0 ... Max memory: no limit Used memory: 3072000 kB But I could not setmem to a value higher than of the domU config. [root at felix tmp]# virsh setmem f11-32 $(( 1024 * 2000 )) error: Invalid value of 2048000 for memory size I could balloon to values between and , and 'xm list f11-32' would show the expected value, but the guest would not show more than . So apparently I misunderstood the meaning of that element. I just tested another domU and I'll update my notes later, but: * If it's assumed that should not be used for this test (contrary to what I put on the discusson page), and that you can't balloon a guest beyond then everything works fine. I can shrink and grow below with no issues. * If it's expected you can balloon to values between and , then something may be wrong. [1] http://fedoraproject.org/wiki/QA_talk:Testcase_Virtualization_XenDomU_Memory_Ballooning > -- > Chris Lalancette From clalance at redhat.com Thu May 7 15:04:20 2009 From: clalance at redhat.com (Chris Lalancette) Date: Thu, 07 May 2009 17:04:20 +0200 Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <8409713.671241707971921.JavaMail.dlbewley@seitan.home.bewley.net> References: <8409713.671241707971921.JavaMail.dlbewley@seitan.home.bewley.net> Message-ID: <4A02F874.3000803@redhat.com> Dale Bewley wrote: > ----- "Chris Lalancette" wrote: >> Dale Bewley wrote: >>> Already been poking![1] But now it's bed time. Hopefully I can do >> some more testing tomorrow. >>> [1] >> http://fedoraproject.org/wiki/User:Dale/Test_Day:2009-05-07_Virtualization:XenDomU >> >> >> Thanks a lot for starting in on this! Quick question about your ballooning >> results; did you remember to balloon down the dom0 before trying to >> balloon up your guest? It doesn't do it automatically on when you run the >> "virsh setmem" command. > > Actually I did not modify the dom0. Can you take a peek at the discussion > page[1]? Perhaps the test case could be clarified a little bit. > > Is step 1 run in the dom0? No, that's in the domU. > > Is step 2 modifying xend to affect dom0 or a guest config? I assumed the > latter but guest configs aren't there anymore. Yes, that's right, it's for changing the guest configuration. > > Just in case, I did try shrinking Domain-0 and testing again. > > [root at felix tmp]# virsh dominfo Domain-0 ... Max memory: no limit Used > memory: 5181440 kB > > [root at felix tmp]# virst setmem Domain-0 $(( 1024 * 3000 )) > > [root at felix tmp]# virsh dominfo Domain-0 ... Max memory: no limit Used > memory: 3072000 kB > > But I could not setmem to a value higher than of the domU config. > [root at felix tmp]# virsh setmem f11-32 $(( 1024 * 2000 )) error: Invalid value > of 2048000 for memory size > > I could balloon to values between and , and 'xm list > f11-32' would show the expected value, but the guest would not show more than > . So apparently I misunderstood the meaning of that element. > > I just tested another domU and I'll update my notes later, but: > > * If it's assumed that should not be used for this test > (contrary to what I put on the discusson page), and that you can't balloon a > guest beyond then everything works fine. I can shrink and grow below > with no issues. > > * If it's expected you can balloon to values between and > , then something may be wrong. Yeah, I also tested it, and it's definitely a bug. You should be able to balloon anywhere between the value of 0 (OK, that would be stupid, but should exist in theory) and "", inclusive. I've filed http://bugzilla.redhat.com/499587 to track it. Thanks for testing it out again. -- Chris Lalancette From rjones at redhat.com Thu May 7 19:40:06 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 7 May 2009 20:40:06 +0100 Subject: libguestfs RPMs for RHEL 5.3 (was: Re: [fedora-virt] libguestfs on non-Fedora targets) In-Reply-To: <49FE3151.5040600@dyfis.net> References: <49FE3151.5040600@dyfis.net> Message-ID: <20090507194006.GA6062@amd.home.annexia.org> I'm just uploading RPMs of libguestfs that I built for RHEL 5.3 i386 (actually CentOS 5.3 + EPEL). You will be able to get them here: http://et.redhat.com/~rjones/libguestfs/files/RHEL-5 in around 60 minutes time. You absolutely must read the README file and follow the instructions there. There are four issues which I know about at the moment, and probably a few others. The package does however pass all of the C API tests, which is generally a good sign. The issues are: (1) On RHEL 5, IDE devices are called '/dev/hd*', whereas on Fedora 11, IDE devices are called '/dev/sd*'. This means that any call which takes or returns a device name, and there are many such calls, must use the /dev/hd* form, eg. guestfs_list_partitions (g); ==> [ "/dev/hda1", "/dev/hda2" ] guestfs_pvcreate (g, "/dev/hda1"); This is annoying if you are writing scripts that must work on both versions of RHEL (5 & 6). You have to first call guestfs_- list_devices, and from that you can tell the format of devices names that the current version uses. If you look in the main() function of 'tests.c' in the source, you can see some code which does this, so that the tests can run using both formats. (2) RHEL 5 didn't support FUSE or NTFS-3g. So unfortunately you may not be able to read Windows (NTFS) partitions. This is a matter of backporting the relevant code to EPEL. I haven't yet worked out how hard that is going to be. (Note there may be other ways around this, such as using the in-kernel NTFS device driver). (3) You need to compile your own recent qemu and provide a qemu wrapper for it. See the README for more details. (4) There is no Augeas support in this version. (Bug in febootstrap). Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From markmc at redhat.com Thu May 7 22:52:29 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Thu, 07 May 2009 23:52:29 +0100 Subject: [fedora-virt] Virtualization Test Day In-Reply-To: <1241679135.7718.9.camel@blaa> References: <1241125048.1335.24.camel@blaa> <1241677831.7718.7.camel@blaa> <1241679135.7718.9.camel@blaa> Message-ID: <1241736749.3314.1.camel@blaa> Hey, Okay, I think we made heaps of progress today. To give you an idea of what went on today, see below for the list of bugs that were filed during the test day. You can also take a peak at the wiki: https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization for results from people's testing. I especially like the Guest Installs page: https://fedoraproject.org/wiki/Test_Day:2009-05-07_Virtualization_Guest_Installs Thanks to everyone who got stuck in, but I thought it would be nice to credit the people who filed bugs ... so, from a quick query: 7 clalance 6 markmc 5 berrange 4 mtosatti 4 jlaska 3 paniraja_km 3 ondrejj 3 charles_rose 2 Rainer.Koenig 2 lili 2 apevec 2 mike.hinz 1 wwoods 1 tjb 1 ravishankar.srinivasan 1 paul.moore 1 esandeen 1 daire.byrne 1 aron No doubt the testing will continue and there'll be a bunch more bugs in my inbox in the morning :-) Thanks again, Mark. Device assignment: https://bugzilla.redhat.com/499614 System locks up when booting with "intel_iommu=on" A report from Fujitsu that intel_iommu=on locks up one of their machines at boot. Likely a BIOS bug. https://bugzilla.redhat.com/499637 IOMMU: no free domain ids Looks like a DMAR domain ID leak in the kernel. https://bugzilla.redhat.com/499261 KVM PCI device assignment fails with "assigned_dev_register_regions: Error: Couldn't mmap 0xffa40000!" This turned out to be libvirt interpreting managed=yes incorrectly. https://bugzilla.redhat.com/499678 libvirt should be able to reset a PCI function even if it causes other unused devices/functions to be reset libvirt's PCI reset logic needs work. Not trivial. https://bugzilla.redhat.com/499561 libvirt does not automatically re-attach an assigned device in the host after guest shutdown Should be fixable. https://bugzilla.redhat.com/499386 libvirt's qemu driver interprets the "managed" attribute incorrectly libvirt was treating managed=yes as managed=no. https://bugzilla.redhat.com/499708 SELinux blocks libvirtd sysfs writes for virNodeDeviceDettach()/Reattach()/Reset() Fixed by latest selinux-policy. https://bugzilla.redhat.com/499259 svirt denials breaks KVM PCI device assignment Not trival to fix correctly. https://bugzilla.redhat.com/499267 Assigning a PCI device using virt-manager/virt-install fails virtinst's code to detach/reset the device before starting the guest was broken. Cole cooked up a patch to fix it. https://bugzilla.redhat.com/499352 Re-enable CONFIG_DMAR_DEFAULT_ON Somewhere along the way, DMAR got disabled by default. Even though the original issue was fixed, it didn't get re-enabled. VNC Authentication: https://bugzilla.redhat.com/499589 virt-manager console widget doesn't know how to collect a 'username' for authentication with VNC https://bugzilla.redhat.com/499594 virt-viewer doesn't know how to provide a username/password to libvirt connections https://bugzilla.redhat.com/499595 virt-viewer mixes up password and username credentials for VNC qemu: https://bugzilla.redhat.com/499697 failure to install new kvm packages on FC10 -> FC11 DVD upgrade https://bugzilla.redhat.com/499542 Graphics crashes when vmware video card is selected https://bugzilla.redhat.com/499547 After OS install, reboot causes the guest to shutdown https://bugzilla.redhat.com/499596 32 bit KVM guest hangs enabling NX protection; booting with -cpu qemu32 works https://bugzilla.redhat.com/499601 KVM guest fails to boot with if=virtio; works with if=ide https://bugzilla.redhat.com/499586 Installation of guest using Network boot(PXE) is failing https://bugzilla.redhat.com/499666 Reboot fails when using qemu -kernel https://bugzilla.redhat.com/499712 [PATCH] qemu-kvm: avoid harmless unhandled wrmsr 0xc0010117 messages libvirt: https://bugzilla.redhat.com/499584 info balloon is repeated in qemu log once per second https://bugzilla.redhat.com/499633 libnuma: Warning: /sys not mounted or invalid https://bugzilla.redhat.com/499698 'virsh destroy' destroys multiple VMs https://bugzilla.redhat.com/499704 VM migration fails with "TCP migration is not supported with this QEMU binary" https://bugzilla.redhat.com/499750 libvirt VM migration fails with "error: Unknown failure" SELinux/sVirt related: https://bugzilla.redhat.com/499536 libvirtd AVC denials - uml backend writing to /root/ https://bugzilla.redhat.com/499551 AVC denial from virsh dominfo - libvirtd (virtd_t) "getattr" svirt_t. https://bugzilla.redhat.com/499569 Guest with source-less cdrom fails to start :: Failed to set security label https://bugzilla.redhat.com/499638 SELinux is preventing libvirtd (virtd_t) "getattr" unconfined_notrans_t. https://bugzilla.redhat.com/499711 svirt error when starting guest - SELinuxSetSecurityLabel: unable to set security context 'system_u:system_r:svirt_t:s0:c152,c660': Invalid argument. Xen DomU: https://bugzilla.redhat.com/499611 Hotplugging vcpus to an F-11 64-bit Xen domU doesn't work https://bugzilla.redhat.com/499587 Cannot balloon a F-11 Xen domU guest above the initial starting memory https://bugzilla.redhat.com/499592 F-11 Xen 64-bit domU cannot be started with > 2047MB of memory https://bugzilla.redhat.com/499621 F-11 Xen 64-bit domU stack traces when unmounting a disk https://bugzilla.redhat.com/499630 F-11 Xen 64-bit domU hangs when trying to format a fake 5TB disk https://bugzilla.redhat.com/499644 F-11 Xen 64-bit domU does not give memory back to the dom0 https://bugzilla.redhat.com/499648 F-11 Xen 64-bit domU cannot crashdump virtinst: https://bugzilla.redhat.com/499555 virtinst: Debian Lenny cannot handle virtio NIC https://bugzilla.redhat.com/499654 failure to add more than 16 virtio-blk devices in virt-manager https://bugzilla.redhat.com/499718 virtinst incorrectly guesses os variant from F11Preview .treeinfo virt-manager: https://bugzilla.redhat.com/499656 virt-manager Scaling->Always - scales the guest screen up but not down. https://bugzilla.redhat.com/499664 inconsistency between reported device names / guest device names in virt-manager https://bugzilla.redhat.com/499665 When window focus set to click - grabs pointer even when virt-manager isn't focused https://bugzilla.redhat.com/499669 hot-removing in-use virtio-blk device in virt-manager should fail https://bugzilla.redhat.com/499685 RFE - include virt type in virt-manager Guest->Details tab https://bugzilla.redhat.com/499703 virt-manager should run stats refresh operation in a background thread per connection anaconda: https://bugzilla.redhat.com/499563 anaconda in a KVM guest only occupies 800x600 of a 1024x768 screen https://bugzilla.redhat.com/499580 F11Preview: anaconda writes ifcfg-eth0 with unquoted NAME parameter From charles at dyfis.net Fri May 8 00:58:56 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 07 May 2009 19:58:56 -0500 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <20090507194006.GA6062@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> Message-ID: <4A0383D0.5060800@dyfis.net> Richard, Thank you kindly for going above and beyond on this one! I've tried rebuilding these packages on an x86_64 host; the process goes smoothly until febootstrap attempts to install the RPMs within the initramfs, at which point things start to go awry; it smells like we're trying to LD_PRELOAD the x86_64-native fakechroot libraries into 32-bit processes running in the guest environment being created. Tweaking the spec file to pull packages for %{_arch}, rather than i386 unconditionally, works around this, but eventually fails during the test phase -- as I had augeas installed on my host (from the EPEL packages), and the guest didn't have those libraries pulled in: guestfsd: error while loading shared libraries: libaugeas.so.0: cannot open shared object file: No such file or directory Kernel panic - not syncing: Attempted to kill init! This was trivial to work around by uninstalling augeas from the host before rebuilding; it did, however, expose that the qemu instance never exits when a panic occurs when running under the test suite. Perhaps "panic=1" should be passed as a kernel parameter to the guest to trigger a reboot on panic, coupled with -no-reboot to ensure that qemu exits? After this point, the only remaining issue is that the ocaml bindings aren't being built, resulting in a packaging failure: RPM build errors: File not found: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmxa File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmx File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.mli File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/stublibs/*.so File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/stublibs/*.so.owner File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmxa File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmx File not found by glob: /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.mli I'll look into this one tomorrow. Thanks again! From charles at dyfis.net Fri May 8 01:18:26 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 07 May 2009 20:18:26 -0500 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <4A0383D0.5060800@dyfis.net> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> <4A0383D0.5060800@dyfis.net> Message-ID: <4A038862.6000805@dyfis.net> Charles Duffy wrote: > After this point, the only remaining issue is that the ocaml bindings > aren't being built, resulting in a packaging failure: > > RPM build errors: > File not found: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a > [...] As it turns out, the issue was not that the ocaml bindings aren't being built, but that they were being installed under /usr/lib/ocaml as opposed to /usr/lib64/ocaml. From rjones at redhat.com Fri May 8 07:34:00 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 8 May 2009 08:34:00 +0100 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <4A0383D0.5060800@dyfis.net> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> <4A0383D0.5060800@dyfis.net> Message-ID: <20090508073400.GA12649@amd.home.annexia.org> On Thu, May 07, 2009 at 07:58:56PM -0500, Charles Duffy wrote: > Richard, > > Thank you kindly for going above and beyond on this one! > > I've tried rebuilding these packages on an x86_64 host; the process goes > smoothly until febootstrap attempts to install the RPMs within the > initramfs, at which point things start to go awry; it smells like we're > trying to LD_PRELOAD the x86_64-native fakechroot libraries into 32-bit > processes running in the guest environment being created. > > Tweaking the spec file to pull packages for %{_arch}, rather than i386 > unconditionally, works around this, but eventually fails during the test > phase -- as I had augeas installed on my host (from the EPEL packages), > and the guest didn't have those libraries pulled in: Right - the appliance has to have the same architecture as the host, so that is indeed a sensible change to make. > guestfsd: error while loading shared libraries: libaugeas.so.0: cannot > open shared object file: No such file or directory > Kernel panic - not syncing: Attempted to kill init! Yes - this is really a bug in febootstrap. It should be possible to pass several repositories to febootstrap, so that we can pass CentOS + EPEL + updates. This isn't possible with febootstrap at the moment, and because CentOS doesn't contain Augeas, we end up with the situation that you described of Augeas being found on the host but not on the appliance. The workaround you used is the same as the one I used. > This was trivial to work around by uninstalling augeas from the host > before rebuilding; it did, however, expose that the qemu instance never > exits when a panic occurs when running under the test suite. Perhaps > "panic=1" should be passed as a kernel parameter to the guest to trigger > a reboot on panic, coupled with -no-reboot to ensure that qemu exits? An excellent suggestion, thanks. I have added it to the git repo. > After this point, the only remaining issue is that the ocaml bindings > aren't being built, resulting in a packaging failure: > > RPM build errors: > File not found: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmxa > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmx > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.mli > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/stublibs/*.so > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/stublibs/*.so.owner > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmxa > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.cmx > File not found by glob: > /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.mli > > I'll look into this one tomorrow. This might be a missing dep. The ones needed for OCaml are, I think, ocaml ocaml-findlib-devel ocaml-ocamldoc I have checked and it does work with OCaml 3.09.3 from EPEL now (after I fixed the bug that you raised earlier). Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Fri May 8 07:37:41 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 8 May 2009 08:37:41 +0100 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <4A038862.6000805@dyfis.net> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> <4A0383D0.5060800@dyfis.net> <4A038862.6000805@dyfis.net> Message-ID: <20090508073741.GB12649@amd.home.annexia.org> On Thu, May 07, 2009 at 08:18:26PM -0500, Charles Duffy wrote: > Charles Duffy wrote: >> After this point, the only remaining issue is that the ocaml bindings >> aren't being built, resulting in a packaging failure: >> >> RPM build errors: >> File not found: >> /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs >> File not found by glob: >> /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a >> [...] > As it turns out, the issue was not that the ocaml bindings aren't being > built, but that they were being installed under /usr/lib/ocaml as > opposed to /usr/lib64/ocaml. Strange one - this works for me on Fedora x86_64 .. I'll look into this a bit more when I come to add libguestfs to EPEL. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From m.a.young at durham.ac.uk Fri May 8 19:12:26 2009 From: m.a.young at durham.ac.uk (M A Young) Date: Fri, 8 May 2009 20:12:26 +0100 (BST) Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels In-Reply-To: References: <20090328163545.GM31725@salstar.sk> <20090414143618.GH351@redhat.com> Message-ID: My latest build is at http://koji.fedoraproject.org/koji/taskinfo?taskID=1340915 and at the repository http://fedorapeople.org/~myoung/dom0/ i686 boots again, due to an upstream problem fix, and I turned off CC_STACKPROTECTOR for i686 again because my test code wasn't enough to get it working in this case. Michael Young From rjones at redhat.com Tue May 12 21:19:32 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 12 May 2009 22:19:32 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' Message-ID: <20090512211932.GA16102@amd.home.annexia.org> I've got the guest kernel hanging randomly at: ACPI: Core revision 20090320 ftrace: converting mcount calls to 0f 1f 44 00 00 ftrace: allocating 18780 entries in 37 pages ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 ..MP-BIOS bug: 8254 timer not connected to IO-APIC ...trying to set up timer (IRQ0) through the 8259A ... ..... (found apic 0 pin 0) ... ....... failed. ...trying to set up timer as Virtual Wire IRQ... This is the very latest on Rawhide. Full packages, versions, etc in the build log here: http://koji.fedoraproject.org/koji/getfile?taskID=1351416&name=root.log http://koji.fedoraproject.org/koji/getfile?taskID=1351416&name=build.log Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From berrange at redhat.com Tue May 12 21:48:41 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 12 May 2009 22:48:41 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090512211932.GA16102@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> Message-ID: <20090512214841.GB12171@redhat.com> On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > I've got the guest kernel hanging randomly at: > > ACPI: Core revision 20090320 > ftrace: converting mcount calls to 0f 1f 44 00 00 > ftrace: allocating 18780 entries in 37 pages > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > ...trying to set up timer (IRQ0) through the 8259A ... > ..... (found apic 0 pin 0) ... > ....... failed. > ...trying to set up timer as Virtual Wire IRQ... > > This is the very latest on Rawhide. Full packages, versions, etc in > the build log here: Last time I saw this error was due a KVM build problem causing it to use the wrong BIOS. ie it was using the plain QEMU bios, instead of the KVM modified bios. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From rjones at redhat.com Tue May 12 22:16:15 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 12 May 2009 23:16:15 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090512214841.GB12171@redhat.com> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> Message-ID: <20090512221615.GA16342@amd.home.annexia.org> On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > I've got the guest kernel hanging randomly at: > > > > ACPI: Core revision 20090320 > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > ftrace: allocating 18780 entries in 37 pages > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > ...trying to set up timer (IRQ0) through the 8259A ... > > ..... (found apic 0 pin 0) ... > > ....... failed. > > ...trying to set up timer as Virtual Wire IRQ... > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > the build log here: > > Last time I saw this error was due a KVM build problem causing it to > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > KVM modified bios. I've never had this error happening on my local machines. However this situation is a bit confusing - this is running in Koji, where the kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and the userspace is QEMU from Rawhide. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From markmc at redhat.com Wed May 13 08:48:58 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 13 May 2009 09:48:58 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090512221615.GA16342@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> Message-ID: <1242204538.31003.9.camel@blaa> On Tue, 2009-05-12 at 23:16 +0100, Richard W.M. Jones wrote: > On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > > > I've got the guest kernel hanging randomly at: > > > > > > ACPI: Core revision 20090320 > > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > > ftrace: allocating 18780 entries in 37 pages > > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > > ...trying to set up timer (IRQ0) through the 8259A ... > > > ..... (found apic 0 pin 0) ... > > > ....... failed. > > > ...trying to set up timer as Virtual Wire IRQ... > > > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > > the build log here: > > > > Last time I saw this error was due a KVM build problem causing it to > > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > > KVM modified bios. > > I've never had this error happening on my local machines. However > this situation is a bit confusing - this is running in Koji, where the > kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and > the userspace is QEMU from Rawhide. As Daniel says, I've seen this happen when /usr/share/qemu/bios.bin was the stock bochs BIOS and not the KVM specific BIOS. Sounds like the kvm-85 build recently pushed to devel/ is busted. Cheers, Mark. From rjones at redhat.com Wed May 13 09:00:55 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 13 May 2009 10:00:55 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <1242204538.31003.9.camel@blaa> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> Message-ID: <20090513090055.GA21623@amd.home.annexia.org> On Wed, May 13, 2009 at 09:48:58AM +0100, Mark McLoughlin wrote: > On Tue, 2009-05-12 at 23:16 +0100, Richard W.M. Jones wrote: > > On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > > > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > > > > > I've got the guest kernel hanging randomly at: > > > > > > > > ACPI: Core revision 20090320 > > > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > > > ftrace: allocating 18780 entries in 37 pages > > > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > > > ...trying to set up timer (IRQ0) through the 8259A ... > > > > ..... (found apic 0 pin 0) ... > > > > ....... failed. > > > > ...trying to set up timer as Virtual Wire IRQ... > > > > > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > > > the build log here: > > > > > > Last time I saw this error was due a KVM build problem causing it to > > > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > > > KVM modified bios. > > > > I've never had this error happening on my local machines. However > > this situation is a bit confusing - this is running in Koji, where the > > kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and > > the userspace is QEMU from Rawhide. > > As Daniel says, I've seen this happen when /usr/share/qemu/bios.bin was > the stock bochs BIOS and not the KVM specific BIOS. > > Sounds like the kvm-85 build recently pushed to devel/ is busted. I checked into this, and it is using /usr/bin/qemu-kvm from the qemu* 2:0.10.50-3.kvm85.fc12 packages. See: http://koji.fedoraproject.org/koji/getfile?taskID=1351416&name=root.log At the moment I can't even find where '0.10.50-3' comes from. It doesn't seem to be in Koji. There's no upstream 'qemu 0.10.50' of course. Any ideas ...? Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From rjones at redhat.com Wed May 13 09:02:02 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 13 May 2009 10:02:02 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090513090055.GA21623@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> <20090513090055.GA21623@amd.home.annexia.org> Message-ID: <20090513090202.GA22229@amd.home.annexia.org> On Wed, May 13, 2009 at 10:00:55AM +0100, Richard W.M. Jones wrote: > At the moment I can't even find where '0.10.50-3' comes from. It > doesn't seem to be in Koji. There's no upstream 'qemu 0.10.50' of > course. OK, I looked back at the Koji page, and there it is, right in the middle ... Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Wed May 13 09:20:15 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 13 May 2009 10:20:15 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <1242204538.31003.9.camel@blaa> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> Message-ID: <20090513092015.GB21623@amd.home.annexia.org> On Wed, May 13, 2009 at 09:48:58AM +0100, Mark McLoughlin wrote: > On Tue, 2009-05-12 at 23:16 +0100, Richard W.M. Jones wrote: > > On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > > > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > > > > > I've got the guest kernel hanging randomly at: > > > > > > > > ACPI: Core revision 20090320 > > > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > > > ftrace: allocating 18780 entries in 37 pages > > > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > > > ...trying to set up timer (IRQ0) through the 8259A ... > > > > ..... (found apic 0 pin 0) ... > > > > ....... failed. > > > > ...trying to set up timer as Virtual Wire IRQ... > > > > > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > > > the build log here: > > > > > > Last time I saw this error was due a KVM build problem causing it to > > > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > > > KVM modified bios. > > > > I've never had this error happening on my local machines. However > > this situation is a bit confusing - this is running in Koji, where the > > kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and > > the userspace is QEMU from Rawhide. > > As Daniel says, I've seen this happen when /usr/share/qemu/bios.bin was > the stock bochs BIOS and not the KVM specific BIOS. > > Sounds like the kvm-85 build recently pushed to devel/ is busted. I installed that version of qemu-kvm and tested it, and it seems to be using: /usr/share/qemu/bios.bin /usr/share/qemu/vgabios-cirrus.bin which I'm guessing is the wrong one. How does it choose the right BIOS to use? (assuming the -L flag is not set) Also does it make a difference if qemu-kvm falls back to software emulation mode, which is likely to be the case for my Koji build. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Wed May 13 10:18:07 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 13 May 2009 11:18:07 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090512211932.GA16102@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> Message-ID: <20090513101807.GA22656@amd.home.annexia.org> I think this bug is distinctly different from previous bugs discussed so I've opened a new BZ for it: https://bugzilla.redhat.com/show_bug.cgi?id=500564 It's quite reproducible on my test machine. Use the 'test-bootbootboot' script from libguestfs git repo. http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=09c42a97daeae4fdccf40374620e533469a281a6 Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From markmc at redhat.com Wed May 13 11:28:22 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 13 May 2009 12:28:22 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090513090055.GA21623@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> <20090513090055.GA21623@amd.home.annexia.org> Message-ID: <1242214102.27745.1.camel@blaa> On Wed, 2009-05-13 at 10:00 +0100, Richard W.M. Jones wrote: > There's no upstream 'qemu 0.10.50' of course. It's the version the qemu folks are using on their development branch. kvm-85 is a snapshot of that branch, basically. Cheers, Mark. From markmc at redhat.com Wed May 13 11:34:46 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 13 May 2009 12:34:46 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <20090513092015.GB21623@amd.home.annexia.org> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> <20090513092015.GB21623@amd.home.annexia.org> Message-ID: <1242214486.27745.6.camel@blaa> On Wed, 2009-05-13 at 10:20 +0100, Richard W.M. Jones wrote: > On Wed, May 13, 2009 at 09:48:58AM +0100, Mark McLoughlin wrote: > > On Tue, 2009-05-12 at 23:16 +0100, Richard W.M. Jones wrote: > > > On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > > > > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > > > > > > > I've got the guest kernel hanging randomly at: > > > > > > > > > > ACPI: Core revision 20090320 > > > > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > > > > ftrace: allocating 18780 entries in 37 pages > > > > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > > > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > > > > ...trying to set up timer (IRQ0) through the 8259A ... > > > > > ..... (found apic 0 pin 0) ... > > > > > ....... failed. > > > > > ...trying to set up timer as Virtual Wire IRQ... > > > > > > > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > > > > the build log here: > > > > > > > > Last time I saw this error was due a KVM build problem causing it to > > > > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > > > > KVM modified bios. > > > > > > I've never had this error happening on my local machines. However > > > this situation is a bit confusing - this is running in Koji, where the > > > kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and > > > the userspace is QEMU from Rawhide. > > > > As Daniel says, I've seen this happen when /usr/share/qemu/bios.bin was > > the stock bochs BIOS and not the KVM specific BIOS. > > > > Sounds like the kvm-85 build recently pushed to devel/ is busted. > > I installed that version of qemu-kvm and tested it, and it seems > to be using: > > /usr/share/qemu/bios.bin > /usr/share/qemu/vgabios-cirrus.bin > > which I'm guessing is the wrong one. How does it choose the right > BIOS to use? (assuming the -L flag is not set) 'readlink /usr/share/qemu/bios.bin' is more instructive; it points to BIOS-bochs-kvm. Glauber, does kvm-85 require some updates to BIOS-bochs-kvm? > Also does it make a difference if qemu-kvm falls back to software > emulation mode, which is likely to be the case for my Koji build. That shouldn't be a problem. Cheers, Mark. From rjones at redhat.com Wed May 13 11:51:32 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 13 May 2009 12:51:32 +0100 Subject: [fedora-virt] Guest kernel hanging at '...trying to set up timer as Virtual Wire IRQ...' In-Reply-To: <1242214486.27745.6.camel@blaa> References: <20090512211932.GA16102@amd.home.annexia.org> <20090512214841.GB12171@redhat.com> <20090512221615.GA16342@amd.home.annexia.org> <1242204538.31003.9.camel@blaa> <20090513092015.GB21623@amd.home.annexia.org> <1242214486.27745.6.camel@blaa> Message-ID: <20090513115132.GC21623@amd.home.annexia.org> On Wed, May 13, 2009 at 12:34:46PM +0100, Mark McLoughlin wrote: > On Wed, 2009-05-13 at 10:20 +0100, Richard W.M. Jones wrote: > > On Wed, May 13, 2009 at 09:48:58AM +0100, Mark McLoughlin wrote: > > > On Tue, 2009-05-12 at 23:16 +0100, Richard W.M. Jones wrote: > > > > On Tue, May 12, 2009 at 10:48:41PM +0100, Daniel P. Berrange wrote: > > > > > On Tue, May 12, 2009 at 10:19:32PM +0100, Richard W.M. Jones wrote: > > > > > > > > > > > > I've got the guest kernel hanging randomly at: > > > > > > > > > > > > ACPI: Core revision 20090320 > > > > > > ftrace: converting mcount calls to 0f 1f 44 00 00 > > > > > > ftrace: allocating 18780 entries in 37 pages > > > > > > ..TIMER: vector=0x30 apic1=0 pin1=0 apic2=-1 pin2=-1 > > > > > > ..MP-BIOS bug: 8254 timer not connected to IO-APIC > > > > > > ...trying to set up timer (IRQ0) through the 8259A ... > > > > > > ..... (found apic 0 pin 0) ... > > > > > > ....... failed. > > > > > > ...trying to set up timer as Virtual Wire IRQ... > > > > > > > > > > > > This is the very latest on Rawhide. Full packages, versions, etc in > > > > > > the build log here: > > > > > > > > > > Last time I saw this error was due a KVM build problem causing it to > > > > > use the wrong BIOS. ie it was using the plain QEMU bios, instead of the > > > > > KVM modified bios. > > > > > > > > I've never had this error happening on my local machines. However > > > > this situation is a bit confusing - this is running in Koji, where the > > > > kernel is RHEL 5.(?), I believe the whole thing is running in Xen, and > > > > the userspace is QEMU from Rawhide. > > > > > > As Daniel says, I've seen this happen when /usr/share/qemu/bios.bin was > > > the stock bochs BIOS and not the KVM specific BIOS. > > > > > > Sounds like the kvm-85 build recently pushed to devel/ is busted. > > > > I installed that version of qemu-kvm and tested it, and it seems > > to be using: > > > > /usr/share/qemu/bios.bin > > /usr/share/qemu/vgabios-cirrus.bin > > > > which I'm guessing is the wrong one. How does it choose the right > > BIOS to use? (assuming the -L flag is not set) > > 'readlink /usr/share/qemu/bios.bin' is more instructive; it points to > BIOS-bochs-kvm. Yes indeed you are correct: bios.bin -> ../bochs/BIOS-bochs-kvm > Glauber, does kvm-85 require some updates to BIOS-bochs-kvm? > > > Also does it make a difference if qemu-kvm falls back to software > > emulation mode, which is likely to be the case for my Koji build. > > That shouldn't be a problem. > > Cheers, > Mark. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From pasik at iki.fi Mon May 18 12:51:58 2009 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Mon, 18 May 2009 15:51:58 +0300 Subject: [fedora-virt] [Xen-devel] [ANNOUNCE] Xen 3.4.0 released! Message-ID: <20090518125158.GQ24960@edu.joroinen.fi> ----- Forwarded message from Keir Fraser ----- From: Keir Fraser To: "xen-devel at lists.xensource.com" , xen-users at lists.xensource.com Cc: Date: Mon, 18 May 2009 05:26:28 -0700 Subject: [Xen-devel] [ANNOUNCE] Xen 3.4.0 released! Folks, We're pleased to announce the official release of Xen 3.4.0! This release contains a number of important new features and updates including: - Device passthrough improvements, with particular emphasis on support for client devices (further support is available as part of the XCI project at http://xenbits.xensource.com/xenclient/) - RAS features: cpu and memory offlining - Power management - improved frequency/voltage controls and deep-sleep support. Scheduler and timers optimised for peak power savings. - Support for the Viridian (Hyper-V) enlightenment interface - Many other x86 and ia64 enhancements and fixes You can get the source using mercurial from: http://xenbits.xensource.com/xen-3.4-testing.hg A pre-prepared source tarball is available from: http://www.xen.org/download/ Cheers, Keir (on behalf of the whole Xen dev team) _______________________________________________ Xen-devel mailing list Xen-devel at lists.xensource.com http://lists.xensource.com/xen-devel ----- End forwarded message ----- From ondrejj at salstar.sk Mon May 18 17:19:28 2009 From: ondrejj at salstar.sk (=?utf-8?B?SsOhbiBPTkRSRUogKFNBTCk=?=) Date: Mon, 18 May 2009 19:19:28 +0200 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 Message-ID: <20090518171928.GD24363@salstar.sk> Hello, during reading FWN I see, that not only I don't like broken dependencies in Fedora repositories. I have an suggestion, how to fix this and make it functional for Fedora 10. Richard, can you add qemu-0.10 (at least binary) into libguestfs package for Fedora 10? Please change you spec file, add qemu sources, compile qemu-kvm as alternate name, like qemu-kvm-libguestfs and configure your libguestfs to use this version of qemu/kvm. I don't use libguestfs rpm package yet, but using libguestfs from git repository with succes in similar way. I have only qemu-kvm copyed from rawhide as alternate name and compiled libguestfs as described above. Everything works fine. Installing qemu-0.10 from rawhide into my Fedora 10 resulted always with qcow2 image corruptios and this package does not worked for me. May be these problems are already fixed, but I still think, this is not a good idea. SAL From rjones at redhat.com Mon May 18 20:15:41 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Mon, 18 May 2009 21:15:41 +0100 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 In-Reply-To: <20090518171928.GD24363@salstar.sk> References: <20090518171928.GD24363@salstar.sk> Message-ID: <20090518201541.GA10256@amd.home.annexia.org> On Mon, May 18, 2009 at 07:19:28PM +0200, J?n ONDREJ (SAL) wrote: > during reading FWN I see, that not only I don't like broken dependencies > in Fedora repositories. I have an suggestion, how to fix this and make it > functional for Fedora 10. > > Richard, can you add qemu-0.10 (at least binary) into libguestfs package > for Fedora 10? Please change you spec file, add qemu sources, compile > qemu-kvm as alternate name, like qemu-kvm-libguestfs and configure your > libguestfs to use this version of qemu/kvm. This is certainly one way to do it. Bundling a whole qemu binary is arguably as bad as breaking dependencies isn't it? I also had a go at backporting the changes to qemu that we ship in F-10. qemu in F-10 is based on qemu 0.9, and doesn't include the vmchannel patch. The vmchannel patch looks deceptively simple: http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html However when I started to backport it, I discovered that the patch depends on many extra functions and features added to slirp since qemu 0.9. I ended up with a qemu which compiled, but kept segfaulting, and it was tricky to diagnose exactly why. > I don't use libguestfs rpm package yet, but using libguestfs from git > repository with succes in similar way. I have only qemu-kvm copyed from > rawhide as alternate name and compiled libguestfs as described above. > Everything works fine. > > Installing qemu-0.10 from rawhide into my Fedora 10 resulted always with > qcow2 image corruptios and this package does not worked for me. May be these > problems are already fixed, but I still think, this is not a good idea. Is it really a problem to use the libguestfs and/or qemu packages from Fedora 11 builds? You can grab the latest builds out of Koji: http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=libguestfs http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=qemu This worked OK for me, although I have now moved to using Fedora 11-Preview, and have mostly abandoned Fedora 10. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From markmc at redhat.com Tue May 19 06:37:10 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Tue, 19 May 2009 07:37:10 +0100 Subject: [fedora-virt] Re: rpms/qemu/F-11 qemu-fix-x86-feature-modifications-for-features-that-set.patch, NONE, 1.1 qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch, NONE, 1.1 qemu-trim-cpu-features-not-supported-by-kvm.patch, NONE, 1.1 qemu.spec, 1.94, 1.95 In-Reply-To: <20090519030127.037F370114@cvs1.fedora.phx.redhat.com> References: <20090519030127.037F370114@cvs1.fedora.phx.redhat.com> Message-ID: <1242715030.6038.30.camel@blaa> Hi Glauber, I've pushed qemu-0.10.4-4.fc11 to stable, since it's had some good testing: https://admin.fedoraproject.org/updates/F11/FEDORA-2009-5050 So, qemu-0.10.4-5.fc11 in testing just has the cpuid trimming backport now: https://admin.fedoraproject.org/updates/qemu-0.10.4-5.fc11 Also, now that I look more closely, Jan's cpuinfo in bug #499596 does actually contain nx. Maybe cpuid trimming won't help him after all? On Tue, 2009-05-19 at 03:01 +0000, Glauber Costa wrote: > Author: glommer > > Update of /cvs/pkgs/rpms/qemu/F-11 > In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv8104 > > Modified Files: > qemu.spec > Added Files: > qemu-fix-x86-feature-modifications-for-features-that-set.patch > qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch > qemu-trim-cpu-features-not-supported-by-kvm.patch > Log Message: > - Backport cpuid trimming from upstream (#499596) ... > qemu-trim-cpu-features-not-supported-by-kvm.patch: > > --- NEW FILE qemu-trim-cpu-features-not-supported-by-kvm.patch --- > >From 3b944bee95c6a5ee561acfc4c4d75d8cc971a567 Mon Sep 17 00:00:00 2001 > From: Avi Kivity > Date: Sun, 3 May 2009 17:04:04 +0300 > Subject: [PATCH STABLE 3/3] kvm: Trim cpu features not supported by kvm > > Remove cpu features that are not supported by kvm from the cpuid features > reported to the guest. > > Signed-off-by: Avi Kivity > Signed-off-by: Anthony Liguori > Signed-off-by: Glauber Costa > --- > target-i386/helper.c | 31 ++++++++++++++++++++++++++++++- > 1 files changed, 30 insertions(+), 1 deletions(-) > > diff --git a/target-i386/helper.c b/target-i386/helper.c > index 1433857..6af5d23 100644 > --- a/target-i386/helper.c > +++ b/target-i386/helper.c > @@ -93,6 +93,21 @@ static void add_flagname_to_bitmaps(char *flagname, uint32_t *features, > } > } > > +static void kvm_trim_features(uint32_t *features, uint32_t supported, > + const char *names[]) > +{ > + int i; > + uint32_t mask; > + > + for (i = 0; i < 32; ++i) { > + mask = 1U << i; > + if ((*features & mask) && !(supported & mask)) { > + printf("Processor feature %s not supported by kvm\n", names[i]); Anthony later killed off this message, it's too verbose. > + *features &= ~mask; > + } > + } > +} > + > typedef struct x86_def_t { > const char *name; > uint32_t level; > @@ -1672,7 +1687,21 @@ CPUX86State *cpu_x86_init(const char *cpu_model) > #ifdef USE_KQEMU > kqemu_init(env); > #endif > - if (kvm_enabled()) > + if (kvm_enabled()) { > kvm_init_vcpu(env); > + kvm_trim_features(&env->cpuid_features, > + kvm_arch_get_supported_cpuid(env, 1, R_EDX), > + feature_name); > + kvm_trim_features(&env->cpuid_ext_features, > + kvm_arch_get_supported_cpuid(env, 1, R_ECX), > + ext_feature_name); > + kvm_trim_features(&env->cpuid_ext2_features, > + kvm_arch_get_supported_cpuid(env, 0x80000001, R_EDX), > + ext2_feature_name); > + kvm_trim_features(&env->cpuid_ext3_features, > + kvm_arch_get_supported_cpuid(env, 0x80000001, R_ECX), > + ext3_feature_name); This doesn't work on my machine without: http://patchwork.kernel.org/patch/23233/ > Index: qemu.spec > =================================================================== > RCS file: /cvs/pkgs/rpms/qemu/F-11/qemu.spec,v > retrieving revision 1.94 > retrieving revision 1.95 > diff -u -p -r1.94 -r1.95 > --- qemu.spec 14 May 2009 11:00:13 -0000 1.94 > +++ qemu.spec 19 May 2009 03:00:56 -0000 1.95 > @@ -1,7 +1,7 @@ > Summary: QEMU is a FAST! processor emulator > Name: qemu > Version: 0.10.4 > -Release: 4%{?dist} > +Release: 5%{?dist} > # Epoch because we pushed a qemu-1.0 package > Epoch: 2 > License: GPLv2+ and LGPLv2+ and BSD > @@ -33,6 +33,10 @@ Patch17: qemu-dma-aio-cancellation1.patc > Patch18: qemu-dma-aio-cancellation2.patch > Patch19: qemu-dma-aio-cancellation3.patch > Patch20: qemu-dma-aio-cancellation4.patch > +Patch21: qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch > +Patch22: qemu-fix-x86-feature-modifications-for-features-that-set.patch > +Patch23: qemu-trim-cpu-features-not-supported-by-kvm.patch > + > > BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n) > BuildRequires: SDL-devel zlib-devel which texi2html gnutls-devel cyrus-sasl-devel > @@ -233,6 +237,9 @@ such as kvmtrace and kvm_stat. > %patch18 -p1 > %patch19 -p1 > %patch20 -p1 > +%patch21 -p1 > +#%patch22 -p1 > +#%patch23 -p1 Why aren't you applying the patches? Cheers, Mark. From markmc at redhat.com Tue May 19 06:59:36 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Tue, 19 May 2009 07:59:36 +0100 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 In-Reply-To: <20090518171928.GD24363@salstar.sk> References: <20090518171928.GD24363@salstar.sk> Message-ID: <1242716376.6038.32.camel@blaa> On Mon, 2009-05-18 at 19:19 +0200, J?n ONDREJ (SAL) wrote: > Installing qemu-0.10 from rawhide into my Fedora 10 resulted always with > qcow2 image corruptios and this package does not worked for me. qemu-0.10.4-4.fc11 should fix these issues, AFAIK. But more generally, I think if people want libguestfs, they update to F-11. Cheers, Mark. From berrange at redhat.com Tue May 19 08:59:02 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 19 May 2009 09:59:02 +0100 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 In-Reply-To: <20090518171928.GD24363@salstar.sk> References: <20090518171928.GD24363@salstar.sk> Message-ID: <20090519085902.GB7732@redhat.com> On Mon, May 18, 2009 at 07:19:28PM +0200, J?n ONDREJ (SAL) wrote: > Richard, can you add qemu-0.10 (at least binary) into libguestfs package > for Fedora 10? Please change you spec file, add qemu sources, compile > qemu-kvm as alternate name, like qemu-kvm-libguestfs and configure your > libguestfs to use this version of qemu/kvm. Please please no. We already have 3 different copies of the QEMU source code in Fedora 10. We don't need another. Doing security errata is painful enough work with 3 copies of the code to audit. Just upgrade to F11 which is practically at GA now. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From ondrejj at salstar.sk Tue May 19 10:21:11 2009 From: ondrejj at salstar.sk (=?utf-8?B?SsOhbiBPTkRSRUogKFNBTCk=?=) Date: Tue, 19 May 2009 12:21:11 +0200 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 In-Reply-To: <20090518201541.GA10256@amd.home.annexia.org> References: <20090518171928.GD24363@salstar.sk> <20090518201541.GA10256@amd.home.annexia.org> Message-ID: <20090519102111.GG24363@salstar.sk> On Mon, May 18, 2009 at 09:15:41PM +0100, Richard W.M. Jones wrote: > On Mon, May 18, 2009 at 07:19:28PM +0200, J?n ONDREJ (SAL) wrote: > > during reading FWN I see, that not only I don't like broken dependencies > > in Fedora repositories. I have an suggestion, how to fix this and make it > > functional for Fedora 10. > > > > Richard, can you add qemu-0.10 (at least binary) into libguestfs package > > for Fedora 10? Please change you spec file, add qemu sources, compile > > qemu-kvm as alternate name, like qemu-kvm-libguestfs and configure your > > libguestfs to use this version of qemu/kvm. > > This is certainly one way to do it. Bundling a whole qemu binary is > arguably as bad as breaking dependencies isn't it? I think it's less disturbing for user, but may be harder for packager. Another way can be to add only binary to libguestfs sources, but it's not an opensource way. > I also had a go at backporting the changes to qemu that we ship in > F-10. qemu in F-10 is based on qemu 0.9, and doesn't include the > vmchannel patch. The vmchannel patch looks deceptively simple: > > http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html > > However when I started to backport it, I discovered that the patch > depends on many extra functions and features added to slirp since > qemu 0.9. I ended up with a qemu which compiled, but kept > segfaulting, and it was tricky to diagnose exactly why. Hmm, even if I don't use KVM in Fedora 10 for production, may be somebody uses, so playing with this package in a stable system is not a good idea. Backporting changes to qemu-0.9 may be the right way, just we can pray to not end with totally non-functional qemu (like your segfaulting). > Is it really a problem to use the libguestfs and/or qemu packages from > Fedora 11 builds? You can grab the latest builds out of Koji: > > http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=libguestfs > http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=qemu > > This worked OK for me, although I have now moved to using Fedora > 11-Preview, and have mostly abandoned Fedora 10. There are still problems with qemu-0.10 on some systems. Look at least for my bugs 499601 or 499596. I also can't install 11-Preview on qemu-0.10 with in graphics mode. Also set proper automatic updates for these packages (each user should update at least their qemu packages) can be harder. SAL From glommer at redhat.com Tue May 19 10:33:53 2009 From: glommer at redhat.com (Glauber Costa) Date: Tue, 19 May 2009 07:33:53 -0300 Subject: [fedora-virt] Re: rpms/qemu/F-11 qemu-fix-x86-feature-modifications-for-features-that-set.patch, NONE, 1.1 qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch, NONE, 1.1 qemu-trim-cpu-features-not-supported-by-kvm.patch, NONE, 1.1 qemu.spec, 1.94, 1.95 In-Reply-To: <1242715030.6038.30.camel@blaa> References: <20090519030127.037F370114@cvs1.fedora.phx.redhat.com> <1242715030.6038.30.camel@blaa> Message-ID: <20090519103353.GW27295@poweredge.glommer> On Tue, May 19, 2009 at 07:37:10AM +0100, Mark McLoughlin wrote: > Hi Glauber, > > I've pushed qemu-0.10.4-4.fc11 to stable, since it's had some good > testing: > > https://admin.fedoraproject.org/updates/F11/FEDORA-2009-5050 > > So, qemu-0.10.4-5.fc11 in testing just has the cpuid trimming backport > now: > > https://admin.fedoraproject.org/updates/qemu-0.10.4-5.fc11 > > Also, now that I look more closely, Jan's cpuinfo in bug #499596 does > actually contain nx. Maybe cpuid trimming won't help him after all? Let him say it. If it doesn't, we may have a bigger problem > > On Tue, 2009-05-19 at 03:01 +0000, Glauber Costa wrote: > > > Author: glommer > > > > Update of /cvs/pkgs/rpms/qemu/F-11 > > In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv8104 > > > > Modified Files: > > qemu.spec > > Added Files: > > qemu-fix-x86-feature-modifications-for-features-that-set.patch > > qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch > > qemu-trim-cpu-features-not-supported-by-kvm.patch > > Log Message: > > - Backport cpuid trimming from upstream (#499596) > ... > > qemu-trim-cpu-features-not-supported-by-kvm.patch: > > > > --- NEW FILE qemu-trim-cpu-features-not-supported-by-kvm.patch --- > > >From 3b944bee95c6a5ee561acfc4c4d75d8cc971a567 Mon Sep 17 00:00:00 2001 > > From: Avi Kivity > > Date: Sun, 3 May 2009 17:04:04 +0300 > > Subject: [PATCH STABLE 3/3] kvm: Trim cpu features not supported by kvm > > > > Remove cpu features that are not supported by kvm from the cpuid features > > reported to the guest. > > > > Signed-off-by: Avi Kivity > > Signed-off-by: Anthony Liguori > > Signed-off-by: Glauber Costa > > --- > > target-i386/helper.c | 31 ++++++++++++++++++++++++++++++- > > 1 files changed, 30 insertions(+), 1 deletions(-) > > > > diff --git a/target-i386/helper.c b/target-i386/helper.c > > index 1433857..6af5d23 100644 > > --- a/target-i386/helper.c > > +++ b/target-i386/helper.c > > @@ -93,6 +93,21 @@ static void add_flagname_to_bitmaps(char *flagname, uint32_t *features, > > } > > } > > > > +static void kvm_trim_features(uint32_t *features, uint32_t supported, > > + const char *names[]) > > +{ > > + int i; > > + uint32_t mask; > > + > > + for (i = 0; i < 32; ++i) { > > + mask = 1U << i; > > + if ((*features & mask) && !(supported & mask)) { > > + printf("Processor feature %s not supported by kvm\n", names[i]); > > Anthony later killed off this message, it's too verbose. > > > + *features &= ~mask; > > + } > > + } > > +} > > + > > typedef struct x86_def_t { > > const char *name; > > uint32_t level; > > @@ -1672,7 +1687,21 @@ CPUX86State *cpu_x86_init(const char *cpu_model) > > #ifdef USE_KQEMU > > kqemu_init(env); > > #endif > > - if (kvm_enabled()) > > + if (kvm_enabled()) { > > kvm_init_vcpu(env); > > + kvm_trim_features(&env->cpuid_features, > > + kvm_arch_get_supported_cpuid(env, 1, R_EDX), > > + feature_name); > > + kvm_trim_features(&env->cpuid_ext_features, > > + kvm_arch_get_supported_cpuid(env, 1, R_ECX), > > + ext_feature_name); > > + kvm_trim_features(&env->cpuid_ext2_features, > > + kvm_arch_get_supported_cpuid(env, 0x80000001, R_EDX), > > + ext2_feature_name); > > + kvm_trim_features(&env->cpuid_ext3_features, > > + kvm_arch_get_supported_cpuid(env, 0x80000001, R_ECX), > > + ext3_feature_name); > > This doesn't work on my machine without: > > http://patchwork.kernel.org/patch/23233/ > > > Index: qemu.spec > > =================================================================== > > RCS file: /cvs/pkgs/rpms/qemu/F-11/qemu.spec,v > > retrieving revision 1.94 > > retrieving revision 1.95 > > diff -u -p -r1.94 -r1.95 > > --- qemu.spec 14 May 2009 11:00:13 -0000 1.94 > > +++ qemu.spec 19 May 2009 03:00:56 -0000 1.95 > > @@ -1,7 +1,7 @@ > > Summary: QEMU is a FAST! processor emulator > > Name: qemu > > Version: 0.10.4 > > -Release: 4%{?dist} > > +Release: 5%{?dist} > > # Epoch because we pushed a qemu-1.0 package > > Epoch: 2 > > License: GPLv2+ and LGPLv2+ and BSD > > @@ -33,6 +33,10 @@ Patch17: qemu-dma-aio-cancellation1.patc > > Patch18: qemu-dma-aio-cancellation2.patch > > Patch19: qemu-dma-aio-cancellation3.patch > > Patch20: qemu-dma-aio-cancellation4.patch > > +Patch21: qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch > > +Patch22: qemu-fix-x86-feature-modifications-for-features-that-set.patch > > +Patch23: qemu-trim-cpu-features-not-supported-by-kvm.patch > > + > > > > BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n) > > BuildRequires: SDL-devel zlib-devel which texi2html gnutls-devel cyrus-sasl-devel > > @@ -233,6 +237,9 @@ such as kvmtrace and kvm_stat. > > %patch18 -p1 > > %patch19 -p1 > > %patch20 -p1 > > +%patch21 -p1 > > +#%patch22 -p1 > > +#%patch23 -p1 > > Why aren't you applying the patches? > > Cheers, > Mark. > From ondrejj at salstar.sk Tue May 19 10:30:18 2009 From: ondrejj at salstar.sk (=?utf-8?B?SsOhbiBPTkRSRUogKFNBTCk=?=) Date: Tue, 19 May 2009 12:30:18 +0200 Subject: [fedora-virt] Re: rpms/qemu/F-11 qemu-fix-x86-feature-modifications-for-features-that-set.patch, NONE, 1.1 qemu-make-x86-cpuid-feature-names-available-in-file-scope.patch, NONE, 1.1 qemu-trim-cpu-features-not-supported-by-kvm.patch, NONE, 1.1 qemu.spec, 1.94, 1.95 In-Reply-To: <20090519103353.GW27295@poweredge.glommer> References: <20090519030127.037F370114@cvs1.fedora.phx.redhat.com> <1242715030.6038.30.camel@blaa> <20090519103353.GW27295@poweredge.glommer> Message-ID: <20090519103018.GH24363@salstar.sk> > > So, qemu-0.10.4-5.fc11 in testing just has the cpuid trimming backport > > now: > > > > https://admin.fedoraproject.org/updates/qemu-0.10.4-5.fc11 > > > > Also, now that I look more closely, Jan's cpuinfo in bug #499596 does > > actually contain nx. Maybe cpuid trimming won't help him after all? > Let him say it. If it doesn't, we may have a bigger problem My reply is already in bugzilla. Still same problem for me. :-( SAL From rjones at redhat.com Tue May 19 10:42:02 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 19 May 2009 11:42:02 +0100 Subject: [fedora-virt] libguestfs broken dependencies on Fedora 10 In-Reply-To: <20090519102111.GG24363@salstar.sk> References: <20090518171928.GD24363@salstar.sk> <20090518201541.GA10256@amd.home.annexia.org> <20090519102111.GG24363@salstar.sk> Message-ID: <20090519104202.GB13159@amd.home.annexia.org> On Tue, May 19, 2009 at 12:21:11PM +0200, J?n ONDREJ (SAL) wrote: > I think it's less disturbing for user, but may be harder for packager. > Another way can be to add only binary to libguestfs sources, but it's not an > opensource way. No way ... > > I also had a go at backporting the changes to qemu that we ship in > > F-10. qemu in F-10 is based on qemu 0.9, and doesn't include the > > vmchannel patch. The vmchannel patch looks deceptively simple: > > > > http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html > > > > However when I started to backport it, I discovered that the patch > > depends on many extra functions and features added to slirp since > > qemu 0.9. I ended up with a qemu which compiled, but kept > > segfaulting, and it was tricky to diagnose exactly why. > > Hmm, even if I don't use KVM in Fedora 10 for production, may be somebody > uses, so playing with this package in a stable system is not a good idea. > > Backporting changes to qemu-0.9 may be the right way, just we can pray to > not end with totally non-functional qemu (like your segfaulting). If you want to have a go at backporting that patch ... > > Is it really a problem to use the libguestfs and/or qemu packages from > > Fedora 11 builds? You can grab the latest builds out of Koji: > > > > http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=libguestfs > > http://koji.fedoraproject.org/koji/search?match=glob&type=package&terms=qemu > > > > This worked OK for me, although I have now moved to using Fedora > > 11-Preview, and have mostly abandoned Fedora 10. > > There are still problems with qemu-0.10 on some systems. Look at least for > my bugs 499601 or 499596. I also can't install 11-Preview on qemu-0.10 with > in graphics mode. These need to be fixed. I've found and Mark has fixed at least one other bug in qemu 0.10 too. Another alternative for you is to build libguestfs from the devel specfile here: http://cvs.fedoraproject.org/viewvc/devel/libguestfs/ However in general terms I don't think there's a good solution for libguestfs in Fedora 10, unless (a) we have a broken runtime with a good error message that tells people they need to upgrade qemu, and (b) we can backport the vmchannel support into qemu 0.9. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From rjones at redhat.com Tue May 19 14:23:41 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 19 May 2009 15:23:41 +0100 Subject: [fedora-virt] ANNOUNCE: virt-inspector Message-ID: <20090519142341.GA14897@amd.home.annexia.org> Hi, I'm pleased to announce the first 'really working' version of virt-inspector. Source is in the libguestfs source repository: http://git.et.redhat.com/?p=libguestfs.git;a=summary Binaries are in Koji for Fedora 11 / Rawhide: http://koji.fedoraproject.org/koji/buildinfo?buildID=102697 (minimum version 1.0.27 required) This is a tool based around libguestfs which can inspect a virtual machine disk image and tell you some interesting things about what's inside it. Some of the things it can tell you: - What operating system(s) are installed, and what distros and versions. It currently covers RHEL releases, Fedora releases, Debian releases, and has limited support for Windows. - How disk partitions are expected to be mounted (eg. /dev/sda1 -> /boot) - What applications are installed. - What kernel(s) are installed. - What kernel modules are installed. It can produce output in plain text (as a report) or in XML (to feed into other programs). It uses libvirt (if available) to get details of inactive domains, or you can run it directly on disk images. Usage is simple: virt-inspector --xml domain > info.xml # xml report virt-inspector /path/to/disk.img | less # plain text report You can also use it as a wrapper for guestfish, so it mounts the guest disks in the correct places, eg: $ virt-inspector --fish /dev/Guests/CentOS5.img guestfish -a /dev/Guests/CentOS5.img -m /dev/VolGroup00/LogVol00:/ -m /dev/sda1:/boot $ eval `virt-inspector --fish /dev/Guests/CentOS5.img` Welcome to guestfish, the libguestfs filesystem interactive shell for editing virtual machine filesystems. Type: 'help' for help with commands 'quit' to quit the shell > mounts /dev/mapper/VolGroup00-LogVol00 /dev/sda1 > ll /boot/ total 12151 drwxr-xr-x 4 root root 1024 May 11 18:16 . drwxr-xr-x 22 root root 4096 May 19 05:06 .. -rw-r--r-- 1 root root 931546 May 7 11:05 System.map-2.6.18-128.1.10.el5 -rw-r--r-- 1 root root 931457 Jan 21 11:10 System.map-2.6.18-128.el5 -rw-r--r-- 1 root root 67942 May 7 11:05 config-2.6.18-128.1.10.el5 -rw-r--r-- 1 root root 67937 Jan 21 11:10 config-2.6.18-128.el5 [etc] To do: - Lots of testing, particularly with different guests. - Answer higher-level questions, eg. "what network drivers will this guest use?" "does this guest need a Xen hypervisor?" - Generate a libvirt XML config file (partly overlaps the functionality of new 'virsh import|export*' functions: http://www.redhat.com/archives/libvir-list/2009-April/msg00401.html ) Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From markmc at redhat.com Tue May 19 17:32:35 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Tue, 19 May 2009 18:32:35 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users Message-ID: <1242754355.9151.8.camel@blaa> Hey, The idea[1] was discussed here before, so I'll keep this short. We've set up a repository for people running Fedora 11 who would like to test the rawhide/F12 virt packages. To use it, do e.g. $> cat > /etc/yum.repos.d/fedora-virt-preview.repo << EOF [rawvirt] name=Virtualization Rawhide for Fedora 11 baseurl=http://markmc.fedorapeople.org/virt-preview/f11/$basearch/ enabled=1 gpgcheck=0 EOF $> yum update At the moment, it contains the F-12 versions of libvirt and qemu, but as F-12 development continues, it will contain more. I'll send periodic mails to the list detailing the latest updates. Also, this is very much a work-in-progress. The TODO list includes: - get createrepo installed on fedorapeople.org - include debuginfo packages in the repo (need more quota) - find a better location than markmc.fedorapeople.org - enable package builders to upload directly to the repo - automatically do preview build/upload for devel/ builds - allow building against preview packages; e.g. for language bindings Comments most welcome. Help with the TODO list is even more welcome :-) Thanks, Mark. [1] - https://www.redhat.com/archives/fedora-virt/2009-April/msg00154.html From rjones at redhat.com Tue May 19 20:04:56 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 19 May 2009 21:04:56 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <1242754355.9151.8.camel@blaa> References: <1242754355.9151.8.camel@blaa> Message-ID: <20090519200456.GA17760@amd.home.annexia.org> On Tue, May 19, 2009 at 06:32:35PM +0100, Mark McLoughlin wrote: > - enable package builders to upload directly to the repo Until this happens, where are you pulling packages from? I can probably push libguestfs & some virt tools packages built for F-11 to somewhere. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From rjones at redhat.com Tue May 19 21:59:55 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 19 May 2009 22:59:55 +0100 Subject: [fedora-virt] Re: [Fwd: [et-mgmt-tools] ANNOUNCE: virt-inspector] In-Reply-To: <4A1318B4.6000204@garlic.com> References: <4A1318B4.6000204@garlic.com> Message-ID: <20090519215955.GA18440@amd.home.annexia.org> On Tue, May 19, 2009 at 01:38:12PM -0700, Roy wrote: > There doesn't seem to be a "configure" in the tar.gz file I downloaded. > Am I doing something wrong? There's one in the tarball ... actually two, but you can ignore the one in the daemon/ subdirectory: $ zcat libguestfs-1.0.27.tar.gz | tar tf - | grep configure libguestfs-1.0.27/daemon/configure libguestfs-1.0.27/daemon/configure.ac libguestfs-1.0.27/configure libguestfs-1.0.27/configure.ac But if you got libguestfs from the git repository, then you will need to run autogen.sh instead. Just do: ./autogen.sh [any configure args] Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From markmc at redhat.com Wed May 20 07:12:50 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 20 May 2009 08:12:50 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090519200456.GA17760@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> Message-ID: <1242803570.4787.21.camel@blaa> On Tue, 2009-05-19 at 21:04 +0100, Richard W.M. Jones wrote: > On Tue, May 19, 2009 at 06:32:35PM +0100, Mark McLoughlin wrote: > > - enable package builders to upload directly to the repo > > Until this happens, where are you pulling packages from? > > I can probably push libguestfs & some virt tools packages built for > F-11 to somewhere. For now, in devel/ just do: $> make scratch-build TARGET=dist-f11 and send me the task number once it's finished. See the script I use to wrote to pull the build into the repo: http://markmc.fedorapeople.org/virt-preview/upload-build.sh Cheers, Mark. From rjones at redhat.com Wed May 20 12:45:22 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 20 May 2009 13:45:22 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090519200456.GA17760@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> Message-ID: <20090520124522.GA22370@amd.home.annexia.org> Are we carrying build dependencies in this repo too? For libguestfs, it's likely we'd want a few non-virt packages in there too. Probably these build deps: febootstrap fakeroot fakechroot zerofree + any new revs of augeas that turn up, eg. if David adds support for more augeas 'lenses'. perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will be pushing new versions of that into fedora-virt repo. qemu is an important build + runtime dependency, but I assume that new versions of qemu will be pushed here. How do you feel about taking packages like the ones above? Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From rjones at redhat.com Wed May 20 12:50:30 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 20 May 2009 13:50:30 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520124522.GA22370@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> Message-ID: <20090520125030.GA22422@amd.home.annexia.org> Here's a (non-scratch) build of the very latest libguestfs for F-11. You're gonna love this one -- 14 subpackages for each architecture :-) http://koji.fedoraproject.org/koji/taskinfo?taskID=1365770 Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From ondrejj at salstar.sk Wed May 20 12:50:30 2009 From: ondrejj at salstar.sk (=?utf-8?B?SsOhbiBPTkRSRUogKFNBTCk=?=) Date: Wed, 20 May 2009 14:50:30 +0200 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520124522.GA22370@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> Message-ID: <20090520125030.GC24830@salstar.sk> On Wed, May 20, 2009 at 01:45:22PM +0100, Richard W.M. Jones wrote: > Are we carrying build dependencies in this repo too? > > For libguestfs, it's likely we'd want a few non-virt packages in there > too. Probably these build deps: > > febootstrap > fakeroot > fakechroot > zerofree > > + any new revs of augeas that turn up, eg. if David adds support for > more augeas 'lenses'. > > perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will > be pushing new versions of that into fedora-virt repo. > > qemu is an important build + runtime dependency, but I assume that new > versions of qemu will be pushed here. > > How do you feel about taking packages like the ones above? I think a separate repository for libguestfs will be a good choice. This new repository can depend on packages from "Rawhide virt repo for F11 users". Libguestfs has a very wide view, what makes more dependencies. SAL From berrange at redhat.com Wed May 20 12:56:08 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 20 May 2009 13:56:08 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520124522.GA22370@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> Message-ID: <20090520125608.GP29798@redhat.com> On Wed, May 20, 2009 at 01:45:22PM +0100, Richard W.M. Jones wrote: > Are we carrying build dependencies in this repo too? > > For libguestfs, it's likely we'd want a few non-virt packages in there > too. Probably these build deps: > > febootstrap > fakeroot > fakechroot > zerofree Aren't these all already in F11 ? (We're not doing a virt-preview for F10) > + any new revs of augeas that turn up, eg. if David adds support for > more augeas 'lenses'. > > perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will > be pushing new versions of that into fedora-virt repo. > > qemu is an important build + runtime dependency, but I assume that new > versions of qemu will be pushed here. > > How do you feel about taking packages like the ones above? It is hard to describe the line, but if they're directly required by new functionality in order to test a virt tool use, then its reasonable to add them. So if the new libguestfs, needs new APIs from perl-Sys-Virt and new QEMU features then its reasonable to add them. Adding new Augeas to get a few extra lens is not critical to be able to test the core functionality, so that would rule it out unless there's some more critical new capability. The virt-preview repo is intended primarily as an aid to testing / early experimentation. It is not intended for "production" deployment. Regards, Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From rjones at redhat.com Wed May 20 13:03:53 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 20 May 2009 14:03:53 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520125608.GP29798@redhat.com> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <20090520125608.GP29798@redhat.com> Message-ID: <20090520130353.GH13159@amd.home.annexia.org> On Wed, May 20, 2009 at 01:56:08PM +0100, Daniel P. Berrange wrote: > On Wed, May 20, 2009 at 01:45:22PM +0100, Richard W.M. Jones wrote: > > Are we carrying build dependencies in this repo too? > > > > For libguestfs, it's likely we'd want a few non-virt packages in there > > too. Probably these build deps: > > > > febootstrap > > fakeroot > > fakechroot > > zerofree > > Aren't these all already in F11 ? They are - it's just that I won't rev them intensively for F-11 (just for Rawhide). Unless we have them in virt-preview for F-11 in which case I can also rev them intensively there too :-) Maybe I'm a bit confused about what virt-preview is really for. I would normally say to people, just pull the updated packages and dependencies from Rawhide. But there's a danger doing that, eg. if some fundamental soname in Rawhide gets bumped (like libcrypto), then they end up pulling in a huge number of Rawhide packages into their F-11. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From markmc at redhat.com Wed May 20 13:07:23 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 20 May 2009 14:07:23 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520124522.GA22370@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> Message-ID: <1242824843.19243.32.camel@blaa> Hey, On Wed, 2009-05-20 at 13:45 +0100, Richard W.M. Jones wrote: > Are we carrying build dependencies in this repo too? Not a whole lot of point right now; you can't build in koji against packages in this repo. > For libguestfs, it's likely we'd want a few non-virt packages in there > too. Probably these build deps: > > febootstrap > fakeroot > fakechroot > zerofree > > + any new revs of augeas that turn up, eg. if David adds support for > more augeas 'lenses'. That's runtime, right? Sure, if it was needed for a new libguestfs feature it would make sense to add. > perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will > be pushing new versions of that into fedora-virt repo. Yeah; again the problem of not being able to build against newer versions of libvirt, though. > qemu is an important build + runtime dependency, but I assume that new > versions of qemu will be pushed here. Yes, we already have qemu-0.10.50/kvm-85 in there. > How do you feel about taking packages like the ones above? I think the way to look at it is - the repo should contain F-12 packages with new virt features that don't belong in F-11 and, within reason, packages needed to enable those new virt features. By within reason, I mean that e.g. if this was an F-11 preview repo for F-10 a few months ago, we would have had to think hard about whether we wanted to include selinux-policy for sVirt testing. My instinct is that we wouldn't include it because it would be likely to break non-virt things on people. Cheers, Mark. From berrange at redhat.com Wed May 20 13:13:56 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 20 May 2009 14:13:56 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520130353.GH13159@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <20090520125608.GP29798@redhat.com> <20090520130353.GH13159@amd.home.annexia.org> Message-ID: <20090520131356.GQ29798@redhat.com> On Wed, May 20, 2009 at 02:03:53PM +0100, Richard W.M. Jones wrote: > On Wed, May 20, 2009 at 01:56:08PM +0100, Daniel P. Berrange wrote: > > On Wed, May 20, 2009 at 01:45:22PM +0100, Richard W.M. Jones wrote: > > > Are we carrying build dependencies in this repo too? > > > > > > For libguestfs, it's likely we'd want a few non-virt packages in there > > > too. Probably these build deps: > > > > > > febootstrap > > > fakeroot > > > fakechroot > > > zerofree > > > > Aren't these all already in F11 ? > > They are - it's just that I won't rev them intensively for F-11 (just > for Rawhide). Unless we have them in virt-preview for F-11 in which > case I can also rev them intensively there too :-) > > Maybe I'm a bit confused about what virt-preview is really for. It is intended as a way for end users to test & experiance new features from rawhide, without needing to use rawhide for their entire distro. It isn't intended as an alternative to updates/updates-testing. So if there is a new fakechroot release with new features in rawhide that you need for libguestfs, then that's a candidate for virt-preview. If it is just a few bug fixes in fakechroot, then that's a candidate for normal stable updates repo. Spefically 'virt-preview' shouldn't have any expectation of support. You don't want to run your servers on virt-preview - its just as raw as rawhide is, merely scope restricted to virt packages. > I would normally say to people, just pull the updated packages and > dependencies from Rawhide. But there's a danger doing that, eg. if > some fundamental soname in Rawhide gets bumped (like libcrypto), then > they end up pulling in a huge number of Rawhide packages into their > F-11. Yes this is the intended use case - it lets people try out new virt features from rawhide without having the rest of their entire OS broken by non-virt features from rawhide. Regards, Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From berrange at redhat.com Wed May 20 13:19:22 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 20 May 2009 14:19:22 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <1242824843.19243.32.camel@blaa> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <1242824843.19243.32.camel@blaa> Message-ID: <20090520131922.GR29798@redhat.com> On Wed, May 20, 2009 at 02:07:23PM +0100, Mark McLoughlin wrote: > Hey, > > On Wed, 2009-05-20 at 13:45 +0100, Richard W.M. Jones wrote: > > Are we carrying build dependencies in this repo too? > > Not a whole lot of point right now; you can't build in koji against > packages in this repo. > > > For libguestfs, it's likely we'd want a few non-virt packages in there > > too. Probably these build deps: > > > > febootstrap > > fakeroot > > fakechroot > > zerofree > > > > + any new revs of augeas that turn up, eg. if David adds support for > > more augeas 'lenses'. > > That's runtime, right? Sure, if it was needed for a new libguestfs > feature it would make sense to add. > > > perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will > > be pushing new versions of that into fedora-virt repo. > > Yeah; again the problem of not being able to build against newer > versions of libvirt, though. Since we don't have a formal koji build target, and are thus just doing transient scratch builds, I see no reason why we couldn't do a mock build as an alternative, adding in the preview repo to the mock config. Only really need x86_64 and i686, so that's easy enough to do on a single host. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From markmc at redhat.com Wed May 20 13:25:17 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Wed, 20 May 2009 14:25:17 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <20090520125030.GA22422@amd.home.annexia.org> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <20090520125030.GA22422@amd.home.annexia.org> Message-ID: <1242825917.19243.49.camel@blaa> On Wed, 2009-05-20 at 13:50 +0100, Richard W.M. Jones wrote: > Here's a (non-scratch) build of the very latest libguestfs for F-11. > You're gonna love this one -- 14 subpackages for each architecture :-) Um. You've pushed this to F-11 too, though: $> koji latest-pkg dist-f12 --quiet libguestfs libguestfs-1.0.27-2.fc12 dist-f12 rjones $> koji latest-pkg dist-f11-updates-candidate --quiet libguestfs libguestfs-1.0.27-1.fc11.1 dist-f11-updates-candidate rjones The preview repo is for packages we choose not to push to the stable branch (because we want to avoid regressions) which we would like virt testers using F-11 to play with. You could think of it this way - there are different types of users consuming these packages: 1) Users who want things to stay stable and who aren't necessarily expecting new features until they update to F-12 - these are people with just the updates repo enabled 2) Same as (1) but who are willing to help out testing updates for the whole distro in order to catch things before they hit the people in category (1) - these people have the updates and updates-testing repos enabled 3) Mostly the same as (1) or (2), but have a specific interest in testing new virt features and are willing to deal with virt regressions - these people enable the updates, updates-testing and preview repos 4) People who are interested in helping with helping with F-12 development in general, not just virt - these people run rawhide Since you're making libguestfs-1.0.27 available to (1) and (2), they'll naturally be available to (3) too. i.e. it doesn't need to be in the preview repo, right? Cheers, Mark. From rjones at redhat.com Wed May 20 13:26:51 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 20 May 2009 14:26:51 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <1242824843.19243.32.camel@blaa> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <1242824843.19243.32.camel@blaa> Message-ID: <20090520132651.GI13159@amd.home.annexia.org> On Wed, May 20, 2009 at 02:07:23PM +0100, Mark McLoughlin wrote: > Hey, > > On Wed, 2009-05-20 at 13:45 +0100, Richard W.M. Jones wrote: > > Are we carrying build dependencies in this repo too? > > Not a whole lot of point right now; you can't build in koji against > packages in this repo. > > > For libguestfs, it's likely we'd want a few non-virt packages in there > > too. Probably these build deps: > > > > febootstrap > > fakeroot > > fakechroot > > zerofree > > > > + any new revs of augeas that turn up, eg. if David adds support for > > more augeas 'lenses'. > > That's runtime, right? Sure, if it was needed for a new libguestfs > feature it would make sense to add. The ones above are build time, below are runtime. > > perl-Sys-Virt is a runtime dependency, but I'm assuming that Dan will > > be pushing new versions of that into fedora-virt repo. > > Yeah; again the problem of not being able to build against newer > versions of libvirt, though. > > > qemu is an important build + runtime dependency, but I assume that new > > versions of qemu will be pushed here. > > Yes, we already have qemu-0.10.50/kvm-85 in there. > > > How do you feel about taking packages like the ones above? > > I think the way to look at it is - the repo should contain F-12 packages > with new virt features that don't belong in F-11 and, within reason, > packages needed to enable those new virt features. > > By within reason, I mean that e.g. if this was an F-11 preview repo for > F-10 a few months ago, we would have had to think hard about whether we > wanted to include selinux-policy for sVirt testing. My instinct is that > we wouldn't include it because it would be likely to break non-virt > things on people. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Wed May 20 13:28:45 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 20 May 2009 14:28:45 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <1242825917.19243.49.camel@blaa> References: <1242754355.9151.8.camel@blaa> <20090519200456.GA17760@amd.home.annexia.org> <20090520124522.GA22370@amd.home.annexia.org> <20090520125030.GA22422@amd.home.annexia.org> <1242825917.19243.49.camel@blaa> Message-ID: <20090520132845.GJ13159@amd.home.annexia.org> On Wed, May 20, 2009 at 02:25:17PM +0100, Mark McLoughlin wrote: > On Wed, 2009-05-20 at 13:50 +0100, Richard W.M. Jones wrote: > > Here's a (non-scratch) build of the very latest libguestfs for F-11. > > You're gonna love this one -- 14 subpackages for each architecture :-) > > Um. You've pushed this to F-11 too, though: > > $> koji latest-pkg dist-f12 --quiet libguestfs > libguestfs-1.0.27-2.fc12 dist-f12 rjones > $> koji latest-pkg dist-f11-updates-candidate --quiet libguestfs > libguestfs-1.0.27-1.fc11.1 dist-f11-updates-candidate rjones > > The preview repo is for packages we choose not to push to the stable > branch (because we want to avoid regressions) which we would like virt > testers using F-11 to play with. Sure sure ... > You could think of it this way - there are different types of users > consuming these packages: > > 1) Users who want things to stay stable and who aren't necessarily > expecting new features until they update to F-12 - these are > people with just the updates repo enabled > > 2) Same as (1) but who are willing to help out testing updates for > the whole distro in order to catch things before they hit the > people in category (1) - these people have the updates and > updates-testing repos enabled > > 3) Mostly the same as (1) or (2), but have a specific interest in > testing new virt features and are willing to deal with virt > regressions - these people enable the updates, updates-testing and > preview repos > > 4) People who are interested in helping with helping with F-12 > development in general, not just virt - these people run rawhide > > Since you're making libguestfs-1.0.27 available to (1) and (2), they'll > naturally be available to (3) too. > > i.e. it doesn't need to be in the preview repo, right? No this one doesn't need to be. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From markmc at redhat.com Thu May 21 13:47:13 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Thu, 21 May 2009 14:47:13 +0100 Subject: [fedora-virt] ANNOUNCE: Rawhide virt repo for F11 users In-Reply-To: <1242754355.9151.8.camel@blaa> References: <1242754355.9151.8.camel@blaa> Message-ID: <1242913633.28844.46.camel@blaa> On Tue, 2009-05-19 at 18:32 +0100, Mark McLoughlin wrote: > Hey, > The idea[1] was discussed here before, so I'll keep this short. > > We've set up a repository for people running Fedora 11 who would like > to test the rawhide/F12 virt packages. To use it, do e.g. > > $> cat > /etc/yum.repos.d/fedora-virt-preview.repo << EOF > [rawvirt] > name=Virtualization Rawhide for Fedora 11 > baseurl=http://markmc.fedorapeople.org/virt-preview/f11/$basearch/ > enabled=1 > gpgcheck=0 > EOF > $> yum update > > At the moment, it contains the F-12 versions of libvirt and qemu, but > as F-12 development continues, it will contain more. I'll send periodic > mails to the list detailing the latest updates. Latest updates: = libvirt = * Thu May 21 2009 Mark McLoughlin - 0.6.3-10.fc12 - Fix XML attribute escaping (bug #499791) - Fix serious event handling issues causing guests to be destroyed (bug #499698) * Thu May 21 2009 Mark McLoughlin - 0.6.3-9.fc12 - Fix qemu argv detection with latest qemu (bug #501923) = qemu = * Thu May 21 2009 Mark McLoughlin - 2:0.10.50-4.kvm86 - Update to kvm-86 release - ChangeLog here: http://marc.info/?l=kvm&m=124282885729710 Cheers, Mark. From lfarkas at lfarkas.org Fri May 22 09:23:01 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Fri, 22 May 2009 11:23:01 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges Message-ID: <4A166EF5.5000502@lfarkas.org> hi, yesterday i try to rebuild fedora packages for rhel/centos/epel-5 and i've got a few comments. i try to build these packages: bochs-bios-2.3.8-0.6.git36989b0d2.fc11.src.rpm etherboot-5.4.4-13.fc11.src.rpm gtk-vnc-0.3.8-8.fc11.src.rpm libvirt-0.6.3-10.fc11.src.rpm openbios-1.0-1.fc11.src.rpm python-virtinst-0.400.3-8.fc11.src.rpm qemu-0.10.4-5.fc11.src.rpm qemu-0.10.50-4.kvm86.fc12.src.rpm vgabios-0.6-0.6.c.fc12.src.rpm virt-manager-0.7.0-4.fc11.src.rpm virt-viewer-0.0.3-5.fc11.src.rpm - first of all i really don't like to trick used in bochs-bios and etherboot. these can be build on ix86 and has only noarch subpackages. why not just define as noarch for the main packages and and exclude/includearch only for ix86? it'd be cleaner and easier. i saw that you're waiting for a rpm bugfix, but currently (and probably later) it also won't compile on x86_64 in koji since it's exclude all ix86 packages! what's more it _is_ build on x86_64 just do not create any result binary rpm which more annoying! - on the other hand in this case bochs-bios subpackage called bochs-bios-data (since you can't call it bochs-bios cause of the above:-( and of course qemu-system-x86 wrongly requires bochs-bios in stead of bochs-bios-data. it's a bug! - bochs-bios install only /usr/share/bochs/BIOS-bochs-latest /usr/share/bochs/BIOS-bochs-legacy but qemu requires BIOS-bochs-kvm which not exist anywhere. a link should have to be create for any of the above file as BIOS-bochs-kvm either in bochs-bios or in qemu. anyway why bochs-bios pulled out of qemu when it's also in qemu? - libvirt BR qemu. why? which require all qemu package which require openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a dependency hell. imho it'd be useful to clean up! - openbios build only for the given arch, but qemu requires all qemu-system* so it can't be build eg on i386 for i386 since libvirt BR qemu! - in more package the in the subpackage "Group field must be present in package", so group name missing:-( and bacuse of this none of openbios, bochs-bios and etherboot can be build on epel:-( - this /etc/libvirt/qemu/networks/default.xml makes me crazy. since blow up my current guest's configs:-((( -- Levente "Si vis pacem para bellum!" From rjones at redhat.com Fri May 22 09:45:53 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 22 May 2009 10:45:53 +0100 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <4A166EF5.5000502@lfarkas.org> References: <4A166EF5.5000502@lfarkas.org> Message-ID: <20090522094553.GA1697@amd.home.annexia.org> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: > hi, > yesterday i try to rebuild fedora packages for rhel/centos/epel-5 and > i've got a few comments. i try to build these packages: Are you actually going to import them into EL-5? I discussed with lkundrak already doing that for just qemu. [...] > - in more package the in the subpackage "Group field must be present in > package", so group name missing:-( and bacuse of this none of openbios, > bochs-bios and etherboot can be build on epel:-( Has RPM changed so it no longer requires the Group field? It was a pretty useless field anyway. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From berrange at redhat.com Fri May 22 10:06:38 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 22 May 2009 11:06:38 +0100 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <4A166EF5.5000502@lfarkas.org> References: <4A166EF5.5000502@lfarkas.org> Message-ID: <20090522100638.GE11009@redhat.com> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: > - libvirt BR qemu. why? which require all qemu package which require > openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a > dependency hell. imho it'd be useful to clean up! The build process wants QEMU so its a BR. There is no dependancy hell here unless you're using the wrong tools. mock trivially pull in the chain of deps as needed during build, so there's nothing to 'clean up'. > - in more package the in the subpackage "Group field must be present in > package", so group name missing:-( and bacuse of this none of openbios, > bochs-bios and etherboot can be build on epel:-( Just add the 'Group' field to the spec in EPEL branch of CVS. > - this /etc/libvirt/qemu/networks/default.xml makes me crazy. since blow > up my current guest's configs:-((( That will only be created if it doesn't exist, so it will not impact things unless you have created another network with a clashing IP address range, which is trivially avoidable. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lfarkas at lfarkas.org Fri May 22 11:10:31 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Fri, 22 May 2009 13:10:31 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <20090522094553.GA1697@amd.home.annexia.org> References: <4A166EF5.5000502@lfarkas.org> <20090522094553.GA1697@amd.home.annexia.org> Message-ID: <4A168827.7060303@lfarkas.org> Richard W.M. Jones wrote: > On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: >> hi, >> yesterday i try to rebuild fedora packages for rhel/centos/epel-5 and >> i've got a few comments. i try to build these packages: > > Are you actually going to import them into EL-5? I discussed > with lkundrak already doing that for just qemu. most people use my repo on centos for kvm and i just try to update th repo, but there are far too many problems. may be next week i spend another week with this:-( > [...] >> - in more package the in the subpackage "Group field must be present in >> package", so group name missing:-( and bacuse of this none of openbios, >> bochs-bios and etherboot can be build on epel:-( > > Has RPM changed so it no longer requires the Group field? > It was a pretty useless field anyway. but make these package not be able to rebuild:-( or would be useful to update rhel's rpm too... -- Levente "Si vis pacem para bellum!" From lfarkas at lfarkas.org Fri May 22 11:23:06 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Fri, 22 May 2009 13:23:06 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <20090522100638.GE11009@redhat.com> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> Message-ID: <4A168B1A.7050205@lfarkas.org> Daniel P. Berrange wrote: > On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: >> - libvirt BR qemu. why? which require all qemu package which require >> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a >> dependency hell. imho it'd be useful to clean up! > > The build process wants QEMU so its a BR. There is no dependancy hell > here unless you're using the wrong tools. mock trivially pull in the > chain of deps as needed during build, so there's nothing to 'clean up'. i can't build (since i don't have ppc) but i need it for qemu-system-ppc which is needed by qemu which is needed by libvirt:-( are you sure all of these req and br are required? >> - this /etc/libvirt/qemu/networks/default.xml makes me crazy. since blow >> up my current guest's configs:-((( > > That will only be created if it doesn't exist, so it will not impact > things unless you have created another network with a clashing IP > address range, which is trivially avoidable. but if it's not exist then it's created and all guest will suddenly have a new interface which may interfere with the existing network which is very hard to find the reason!:-( -- Levente "Si vis pacem para bellum!" From berrange at redhat.com Fri May 22 11:33:29 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 22 May 2009 12:33:29 +0100 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <4A168B1A.7050205@lfarkas.org> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> Message-ID: <20090522113329.GF11009@redhat.com> On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: > Daniel P. Berrange wrote: > > On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: > >> - libvirt BR qemu. why? which require all qemu package which require > >> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a > >> dependency hell. imho it'd be useful to clean up! > > > > The build process wants QEMU so its a BR. There is no dependancy hell > > here unless you're using the wrong tools. mock trivially pull in the > > chain of deps as needed during build, so there's nothing to 'clean up'. > > i can't build (since i don't have ppc) but i need it for qemu-system-ppc > which is needed by qemu which is needed by libvirt:-( > are you sure all of these req and br are required? You're not making any sense here. You don't need a ppc host, to build qemu-system-ppc. All host architectures can build all QEMU targets, you're not restricted to matching host & qemu target, with the exception of KVM. > >> - this /etc/libvirt/qemu/networks/default.xml makes me crazy. since blow > >> up my current guest's configs:-((( > > > > That will only be created if it doesn't exist, so it will not impact > > things unless you have created another network with a clashing IP > > address range, which is trivially avoidable. > > but if it's not exist then it's created and all guest will suddenly have > a new interface which may interfere with the existing network which is > very hard to find the reason!:-( This only creates a bridge on the host. No guest is connected to this bridge unless you configure it that way. So merely having the config does not affect existing guests Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lfarkas at lfarkas.org Fri May 22 12:04:28 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Fri, 22 May 2009 14:04:28 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <20090522113329.GF11009@redhat.com> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> <20090522113329.GF11009@redhat.com> Message-ID: <4A1694CC.3060308@lfarkas.org> Daniel P. Berrange wrote: > On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: >> Daniel P. Berrange wrote: >>> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: >>>> - libvirt BR qemu. why? which require all qemu package which require >>>> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a >>>> dependency hell. imho it'd be useful to clean up! >>> The build process wants QEMU so its a BR. There is no dependancy hell >>> here unless you're using the wrong tools. mock trivially pull in the >>> chain of deps as needed during build, so there's nothing to 'clean up'. >> i can't build (since i don't have ppc) but i need it for qemu-system-ppc >> which is needed by qemu which is needed by libvirt:-( >> are you sure all of these req and br are required? > > You're not making any sense here. You don't need a ppc host, to build > qemu-system-ppc. All host architectures can build all QEMU targets, > you're not restricted to matching host & qemu target, with the exception > of KVM. i wrote above i can't build openbios-ppc which required by qemu etc...so i can't build libvirt:-( >>>> - this /etc/libvirt/qemu/networks/default.xml makes me crazy. since blow >>>> up my current guest's configs:-((( >>> That will only be created if it doesn't exist, so it will not impact >>> things unless you have created another network with a clashing IP >>> address range, which is trivially avoidable. >> but if it's not exist then it's created and all guest will suddenly have >> a new interface which may interfere with the existing network which is >> very hard to find the reason!:-( > > This only creates a bridge on the host. No guest is connected to this > bridge unless you configure it that way. So merely having the config > does not affect existing guests ok. i don't know since yesterday i wasn't more time for this, but until now there was no such interface in guest as virbr0 and after testing the new packages i have. may be there are other reason for this but that also comes from one of these new packages. -- Levente "Si vis pacem para bellum!" From berrange at redhat.com Fri May 22 12:16:54 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 22 May 2009 13:16:54 +0100 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <4A1694CC.3060308@lfarkas.org> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> <20090522113329.GF11009@redhat.com> <4A1694CC.3060308@lfarkas.org> Message-ID: <20090522121654.GG11009@redhat.com> On Fri, May 22, 2009 at 02:04:28PM +0200, Farkas Levente wrote: > Daniel P. Berrange wrote: > > On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: > >> Daniel P. Berrange wrote: > >>> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: > >>>> - libvirt BR qemu. why? which require all qemu package which require > >>>> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a > >>>> dependency hell. imho it'd be useful to clean up! > >>> The build process wants QEMU so its a BR. There is no dependancy hell > >>> here unless you're using the wrong tools. mock trivially pull in the > >>> chain of deps as needed during build, so there's nothing to 'clean up'. > >> i can't build (since i don't have ppc) but i need it for qemu-system-ppc > >> which is needed by qemu which is needed by libvirt:-( > >> are you sure all of these req and br are required? > > > > You're not making any sense here. You don't need a ppc host, to build > > qemu-system-ppc. All host architectures can build all QEMU targets, > > you're not restricted to matching host & qemu target, with the exception > > of KVM. > > i wrote above i can't build openbios-ppc which required by qemu etc...so > i can't build libvirt:-( Then just disable the qemu-system-ppc bits in QEMU. It really isn't hard to remove the ppc sub-RPM and change the target-list for the QEMU build to turn off ppc. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lfarkas at lfarkas.org Fri May 22 19:41:01 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Fri, 22 May 2009 21:41:01 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <20090522121654.GG11009@redhat.com> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> <20090522113329.GF11009@redhat.com> <4A1694CC.3060308@lfarkas.org> <20090522121654.GG11009@redhat.com> Message-ID: <4A16FFCD.8050803@lfarkas.org> Daniel P. Berrange wrote: > On Fri, May 22, 2009 at 02:04:28PM +0200, Farkas Levente wrote: >> Daniel P. Berrange wrote: >>> On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: >>>> Daniel P. Berrange wrote: >>>>> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: >>>>>> - libvirt BR qemu. why? which require all qemu package which require >>>>>> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a >>>>>> dependency hell. imho it'd be useful to clean up! >>>>> The build process wants QEMU so its a BR. There is no dependancy hell >>>>> here unless you're using the wrong tools. mock trivially pull in the >>>>> chain of deps as needed during build, so there's nothing to 'clean up'. >>>> i can't build (since i don't have ppc) but i need it for qemu-system-ppc >>>> which is needed by qemu which is needed by libvirt:-( >>>> are you sure all of these req and br are required? >>> You're not making any sense here. You don't need a ppc host, to build >>> qemu-system-ppc. All host architectures can build all QEMU targets, >>> you're not restricted to matching host & qemu target, with the exception >>> of KVM. >> i wrote above i can't build openbios-ppc which required by qemu etc...so >> i can't build libvirt:-( > > Then just disable the qemu-system-ppc bits in QEMU. It really isn't hard > to remove the ppc sub-RPM and change the target-list for the QEMU build > to turn off ppc. this means even on a primary platform ix86 these packages can't be rebuild without modification. wouldn't be easier to put back openbios-ppc, vgabios, bochs-bios-data, etherboot into qemu? -- Levente "Si vis pacem para bellum!" From m.a.young at durham.ac.uk Fri May 22 23:26:29 2009 From: m.a.young at durham.ac.uk (M A Young) Date: Sat, 23 May 2009 00:26:29 +0100 (BST) Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels In-Reply-To: References: <20090328163545.GM31725@salstar.sk> <20090414143618.GH351@redhat.com> Message-ID: This build http://koji.fedoraproject.org/koji/taskinfo?taskID=1371327 updates to the latest xen-tip/next version, and is of courase available via the repository http://fedorapeople.org/~myoung/dom0/ Michael Young From pasik at iki.fi Sat May 23 18:16:28 2009 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Sat, 23 May 2009 21:16:28 +0300 Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels In-Reply-To: References: <20090328163545.GM31725@salstar.sk> <20090414143618.GH351@redhat.com> Message-ID: <20090523181628.GF24960@edu.joroinen.fi> On Sat, May 23, 2009 at 12:26:29AM +0100, M A Young wrote: > This build http://koji.fedoraproject.org/koji/taskinfo?taskID=1371327 > updates to the latest xen-tip/next version, and is of courase available > via the repository http://fedorapeople.org/~myoung/dom0/ > I assume this is a version which should have Xen HVM guests working now? "[Xen-devel] [GIT] privcmd fixes --> working HVM": http://lists.xensource.com/archives/html/xen-devel/2009-05/msg00839.html -- Pasi From cochranb at speakeasy.net Sat May 23 19:19:26 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sat, 23 May 2009 15:19:26 -0400 Subject: [fedora-virt] Best Way To Install Ubuntu 8.10 (AMD64) Guest? Message-ID: <4A184C3E.5050001@speakeasy.net> Hi, I have a fully updated Fedora 11 with Virtualization. My laptop is a Dell Latitude E6400, 4 Gb of memory, Intel P9500 dual core processor with virtualization support. I want to install Ubuntu 8.10 Desktop (AMD64) as a guest in Fedora. What is the easiest way of installing it such that I have internet access, USB, and possibly RS-232 serial port access? Thanks! Bob Cochran From m.a.young at durham.ac.uk Sat May 23 19:39:31 2009 From: m.a.young at durham.ac.uk (M A Young) Date: Sat, 23 May 2009 20:39:31 +0100 (BST) Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels In-Reply-To: <20090523181628.GF24960@edu.joroinen.fi> References: <20090328163545.GM31725@salstar.sk> <20090414143618.GH351@redhat.com> <20090523181628.GF24960@edu.joroinen.fi> Message-ID: On Sat, 23 May 2009, Pasi K?rkk?inen wrote: > On Sat, May 23, 2009 at 12:26:29AM +0100, M A Young wrote: >> This build http://koji.fedoraproject.org/koji/taskinfo?taskID=1371327 >> updates to the latest xen-tip/next version, and is of courase available >> via the repository http://fedorapeople.org/~myoung/dom0/ >> > > I assume this is a version which should have Xen HVM guests working now? > > "[Xen-devel] [GIT] privcmd fixes --> working HVM": > http://lists.xensource.com/archives/html/xen-devel/2009-05/msg00839.html Yes, that patch is in, as is the followup patch that takes it out again and puts it somewhere else. Michael Young From pasik at iki.fi Sat May 23 20:42:20 2009 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Sat, 23 May 2009 23:42:20 +0300 Subject: [fedora-virt] Re: [Fedora-xen] Dom0 kernels with HVM guest support In-Reply-To: References: <20090414143618.GH351@redhat.com> <20090523181628.GF24960@edu.joroinen.fi> Message-ID: <20090523204219.GG24960@edu.joroinen.fi> On Sat, May 23, 2009 at 08:39:31PM +0100, M A Young wrote: > On Sat, 23 May 2009, Pasi K?rkk?inen wrote: > > >On Sat, May 23, 2009 at 12:26:29AM +0100, M A Young wrote: > >>This build http://koji.fedoraproject.org/koji/taskinfo?taskID=1371327 > >>updates to the latest xen-tip/next version, and is of courase available > >>via the repository http://fedorapeople.org/~myoung/dom0/ > >> > > > >I assume this is a version which should have Xen HVM guests working now? > > > >"[Xen-devel] [GIT] privcmd fixes --> working HVM": > >http://lists.xensource.com/archives/html/xen-devel/2009-05/msg00839.html > > Yes, that patch is in, as is the followup patch that takes it out again > and puts it somewhere else. > Great. Now we just have to get that blktap2 in and most of the common features are there.. -- Pasi From rjones at redhat.com Sun May 24 08:15:58 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Sun, 24 May 2009 09:15:58 +0100 Subject: [fedora-virt] Best Way To Install Ubuntu 8.10 (AMD64) Guest? In-Reply-To: <4A184C3E.5050001@speakeasy.net> References: <4A184C3E.5050001@speakeasy.net> Message-ID: <20090524081558.GA19272@amd.home.annexia.org> On Sat, May 23, 2009 at 03:19:26PM -0400, Robert L Cochran wrote: > Hi, > > I have a fully updated Fedora 11 with Virtualization. My laptop is a > Dell Latitude E6400, 4 Gb of memory, Intel P9500 dual core processor > with virtualization support. > > I want to install Ubuntu 8.10 Desktop (AMD64) as a guest in Fedora. What > is the easiest way of installing it such that I have internet access, > USB, and possibly RS-232 serial port access? You can install Ubuntu easily from the live CD. Download the ISO from here: http://www.ubuntu.com/getubuntu/download then start up virt-manager and click through to install a new guest from an ISO. Alternately you can do it from virt-install, something along these lines: lvcreate -L 10G -n Ubuntu VG virt-install -v -n Ubuntu \ --accelerate -r 512 \ -c /tmp/ubuntu-9.04-desktop-i386.iso -f /dev/VG/Ubuntu Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From berrange at redhat.com Sun May 24 12:49:09 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Sun, 24 May 2009 13:49:09 +0100 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <4A16FFCD.8050803@lfarkas.org> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> <20090522113329.GF11009@redhat.com> <4A1694CC.3060308@lfarkas.org> <20090522121654.GG11009@redhat.com> <4A16FFCD.8050803@lfarkas.org> Message-ID: <20090524124909.GA10902@redhat.com> On Fri, May 22, 2009 at 09:41:01PM +0200, Farkas Levente wrote: > Daniel P. Berrange wrote: > > On Fri, May 22, 2009 at 02:04:28PM +0200, Farkas Levente wrote: > >> Daniel P. Berrange wrote: > >>> On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: > >>>> Daniel P. Berrange wrote: > >>>>> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: > >>>>>> - libvirt BR qemu. why? which require all qemu package which require > >>>>>> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a > >>>>>> dependency hell. imho it'd be useful to clean up! > >>>>> The build process wants QEMU so its a BR. There is no dependancy hell > >>>>> here unless you're using the wrong tools. mock trivially pull in the > >>>>> chain of deps as needed during build, so there's nothing to 'clean up'. > >>>> i can't build (since i don't have ppc) but i need it for qemu-system-ppc > >>>> which is needed by qemu which is needed by libvirt:-( > >>>> are you sure all of these req and br are required? > >>> You're not making any sense here. You don't need a ppc host, to build > >>> qemu-system-ppc. All host architectures can build all QEMU targets, > >>> you're not restricted to matching host & qemu target, with the exception > >>> of KVM. > >> i wrote above i can't build openbios-ppc which required by qemu etc...so > >> i can't build libvirt:-( > > > > Then just disable the qemu-system-ppc bits in QEMU. It really isn't hard > > to remove the ppc sub-RPM and change the target-list for the QEMU build > > to turn off ppc. > > this means even on a primary platform ix86 these packages can't be > rebuild without modification. > wouldn't be easier to put back openbios-ppc, vgabios, bochs-bios-data, > etherboot into qemu? These packages were split out from QEMU because, they were duplicating functionality in Bochs & QEMU packages. It was also not clear that they were in compliance with the license, because there was no corresponding source to the pre-built binary being shipped. You fundamentally can't build many of these packages on all archs. This last reason is the real key bit. openbios-ppc can only be built from source on a PPC host, so we need to build on PPC, and then include that built binary on a 2nd build on all other archs. The only practical way todo this is if the BIOS is separate from QEMU, otherwise you end up havig to rebuild far too much stuff each time. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From rjones at redhat.com Sun May 24 14:57:28 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Sun, 24 May 2009 15:57:28 +0100 Subject: Libguestfs on Debian (was: Re: [fedora-virt] P2V and grub bootloader) In-Reply-To: <20090501121828.GA10510@amd.home.annexia.org> References: <49FA1813.1030808@conversis.de> <20090430213937.GA8368@amd.home.annexia.org> <20090430221132.GA8394@amd.home.annexia.org> <49FA2568.8010508@conversis.de> <20090430223014.GC8394@amd.home.annexia.org> <49FA4573.8030408@conversis.de> <20090501083423.GA9933@amd.home.annexia.org> <20090501121828.GA10510@amd.home.annexia.org> Message-ID: <20090524145728.GA20045@amd.home.annexia.org> On Fri, May 01, 2009 at 01:18:28PM +0100, Richard W.M. Jones wrote: > (1) Port febootstrap plus its deps to your distro. Approximately > those deps would be: yum, python-rpm, fakeroot, fakechroot >= 2.9. So this is the approach I took, and libguestfs + febootstrap now compiles and passes most of the tests on Debian. I'll try to post some more detailed instructions later, or even get the package into Debian, but for now let me just say that you do need to use the * latest versions of everything from git *, and keep your wits about you. Also you can following postings on my blog: http://rwmj.wordpress.com/ and look at libguestfs bugs if you some specific issue: https://bugzilla.redhat.com/buglist.cgi?component=libguestfs&product=Fedora Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From cochranb at speakeasy.net Sun May 24 16:06:29 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sun, 24 May 2009 12:06:29 -0400 Subject: [Fwd: Re: [fedora-virt] Best Way To Install Ubuntu 8.10 (AMD64) Guest?] Message-ID: <4A197085.5070204@speakeasy.net> Ooops, sorry, I meant this for the Rich and the list. Bob -------- Original Message -------- Subject: Re: [fedora-virt] Best Way To Install Ubuntu 8.10 (AMD64) Guest? Date: Sun, 24 May 2009 12:05:11 -0400 From: Robert L Cochran To: Richard W.M. Jones On 05/24/2009 04:15 AM, Richard W.M. Jones wrote: > On Sat, May 23, 2009 at 03:19:26PM -0400, Robert L Cochran wrote: > >> Hi, >> >> I have a fully updated Fedora 11 with Virtualization. My laptop is a >> Dell Latitude E6400, 4 Gb of memory, Intel P9500 dual core processor >> with virtualization support. >> >> I want to install Ubuntu 8.10 Desktop (AMD64) as a guest in Fedora. What >> is the easiest way of installing it such that I have internet access, >> USB, and possibly RS-232 serial port access? >> > > You can install Ubuntu easily from the live CD. Download the ISO from > here: > > http://www.ubuntu.com/getubuntu/download > > then start up virt-manager and click through to install a new guest > from an ISO. Alternately you can do it from virt-install, something > along these lines: > > lvcreate -L 10G -n Ubuntu VG > virt-install -v -n Ubuntu \ > --accelerate -r 512 \ > -c /tmp/ubuntu-9.04-desktop-i386.iso -f /dev/VG/Ubuntu > > Rich. > Thank you, this worked great with Ubuntu 9.04 (i386). It feels pretty easy to get a virtual machine going. Being a programmer myself, I can well appreciate how much work you and your team had to put in to make it so easily done. Bob From lfarkas at lfarkas.org Sun May 24 21:02:02 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Sun, 24 May 2009 23:02:02 +0200 Subject: [fedora-virt] a few comment about the current virt pacakges In-Reply-To: <20090524124909.GA10902@redhat.com> References: <4A166EF5.5000502@lfarkas.org> <20090522100638.GE11009@redhat.com> <4A168B1A.7050205@lfarkas.org> <20090522113329.GF11009@redhat.com> <4A1694CC.3060308@lfarkas.org> <20090522121654.GG11009@redhat.com> <4A16FFCD.8050803@lfarkas.org> <20090524124909.GA10902@redhat.com> Message-ID: <4A19B5CA.4000507@lfarkas.org> Daniel P. Berrange wrote: > On Fri, May 22, 2009 at 09:41:01PM +0200, Farkas Levente wrote: >> Daniel P. Berrange wrote: >>> On Fri, May 22, 2009 at 02:04:28PM +0200, Farkas Levente wrote: >>>> Daniel P. Berrange wrote: >>>>> On Fri, May 22, 2009 at 01:23:06PM +0200, Farkas Levente wrote: >>>>>> Daniel P. Berrange wrote: >>>>>>> On Fri, May 22, 2009 at 11:23:01AM +0200, Farkas Levente wrote: >>>>>>>> - libvirt BR qemu. why? which require all qemu package which require >>>>>>>> openbios-ppc, vgabios, bochs-bios-data, etherboot-*. so this is a >>>>>>>> dependency hell. imho it'd be useful to clean up! >>>>>>> The build process wants QEMU so its a BR. There is no dependancy hell >>>>>>> here unless you're using the wrong tools. mock trivially pull in the >>>>>>> chain of deps as needed during build, so there's nothing to 'clean up'. >>>>>> i can't build (since i don't have ppc) but i need it for qemu-system-ppc >>>>>> which is needed by qemu which is needed by libvirt:-( >>>>>> are you sure all of these req and br are required? >>>>> You're not making any sense here. You don't need a ppc host, to build >>>>> qemu-system-ppc. All host architectures can build all QEMU targets, >>>>> you're not restricted to matching host & qemu target, with the exception >>>>> of KVM. >>>> i wrote above i can't build openbios-ppc which required by qemu etc...so >>>> i can't build libvirt:-( >>> Then just disable the qemu-system-ppc bits in QEMU. It really isn't hard >>> to remove the ppc sub-RPM and change the target-list for the QEMU build >>> to turn off ppc. >> this means even on a primary platform ix86 these packages can't be >> rebuild without modification. >> wouldn't be easier to put back openbios-ppc, vgabios, bochs-bios-data, >> etherboot into qemu? > > These packages were split out from QEMU because, they were duplicating > functionality in Bochs & QEMU packages. It was also not clear that they > were in compliance with the license, because there was no corresponding > source to the pre-built binary being shipped. You fundamentally can't > build many of these packages on all archs. This last reason is the real > key bit. openbios-ppc can only be built from source on a PPC host, so > we need to build on PPC, and then include that built binary on a 2nd > build on all other archs. The only practical way todo this is if the > BIOS is separate from QEMU, otherwise you end up havig to rebuild far > too much stuff each time. and if qemu do not require qemu-system-ppc which do not require openbios-ppc then those part which will run on ix86 can be build on ix86... -- Levente "Si vis pacem para bellum!" From wangjihai at huawei.com Mon May 25 01:21:37 2009 From: wangjihai at huawei.com (Arthur Wang) Date: Mon, 25 May 2009 09:21:37 +0800 Subject: [fedora-virt] how to build iso from virt-p2v Message-ID: <200905250921375077225@huawei.com> Hello, I have a trouble when build virt-p2v-0.9.9.tar.gz to generate an iso file. Search for help I have install all the packet that virt-p2v depend on, and i run the follow commands successfully configure make rpm But when i enter "make build" command, wait minutes, i got the follow error: Setting maximal mount count to -1 Setting interval between checks to 0 seconds Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: released. Please verify its path and try again make: *** [build] error 1 Could anyone tell me where it want to get repository metadata , and which tools or command it use to get? Arthur Wang Research Department, CPD Huawei Technologies Co., Ltd. Phone: 86-755-28976096 Email: wangjihai at huawei.com 2009-05-25 -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Mon May 25 01:40:01 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Sun, 24 May 2009 21:40:01 -0400 Subject: [fedora-virt] Distributing Fedora 11 Created Virtual Machines Message-ID: <4A19F6F1.40601@speakeasy.net> I have a virtual machine created on Fedora 11 which runs Ubuntu 9.04. I have some more customizations to do with the machine, such as getting it to speak to (FTDI 232R-based) USB devices. When that is done, I would like to copy the virtual machine to a USB flash drive and postal mail that to a teacher. The scenario I'm envisioning is that the teacher can then import the virtual machine and play it on his or her computer -- whatever that computer is, perhaps a Windows machine running VirtualBox or VMWare. Not necessarily a machine running Fedora 11, you see. Now we have my first question: can she run my virtual machine on another virtual machine manager, which is not necessarily a Fedora product? To go on with the story I'd like the teacher to add her customizations to the Ubuntu virtual machine, then she in turn will deploy that to students. They would copy the virtual machine to their computers (whatever those are) and play them on those computers. These will probably be whatever computers the school has an d/or whatever computers the student can afford to pay for. Could this work as well? Suppose the answer is no they cannot play a Fedora 11-created virtual machine on their own virtualization software. Is it possible to take the virtual machine file and clone (part of?) it to a hard drive such that Ubuntu can be booted from the hard drive? I'm having a wonderful time in the sandbox, playing with my Ubuntu guest. Bob Cochran From frankly3d at gmail.com Mon May 25 12:07:53 2009 From: frankly3d at gmail.com (Frank Murphy (Frankly3d)) Date: Mon, 25 May 2009 13:07:53 +0100 Subject: [fedora-virt] Newbie Q? Mailing-List Support. Message-ID: <4A1A8A19.7040001@gmail.com> Is the mailing list support just for people who run guests on Fedora, or can it be used with Fedora Guest on CentOS Base? Thinking of putting a virt box or two together. FRank -- msn: frankly3d skype: frankly3d Mailing-List Reply to: Mailing-List Still Learning, Unicode where possible From cochranb at speakeasy.net Mon May 25 16:56:03 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Mon, 25 May 2009 12:56:03 -0400 Subject: [fedora-virt] Successful Compile of libguestfs 1.0.31 in Fedora 11 Message-ID: <4A1ACDA3.6020601@speakeasy.net> I've been looking at the postings here more closely now that I have a Ubuntu guest running, and I decided to compile libguestfs-1.0.31 in Fedora 11. It took some time for me to get the dependencies installed. Question: what packages do I need for the ruby bindings? What package represents rake? How about OCaml bindings -- what packages are needed to get it to build? When building this file as part of `make`: initramfs.fedora-11.x86_64.img I got an awesome number of errors related to my not being the root user, but I assume those errors are okay since I built as an ordinary user. Also the perl documentation process came up with a large number of errors. After doing `make check` I got this result: Test boot completed after 10 iterations. PASS: test-bootbootboot.sh ================== All 2 tests passed ================== Am I ready for 'make install' -- is libguestfs built sufficiently well for that? Although I would like to get the ruby and OCaml bindings to build. Thanks Bob From markmc at redhat.com Mon May 25 17:06:55 2009 From: markmc at redhat.com (Mark McLoughlin) Date: Mon, 25 May 2009 18:06:55 +0100 Subject: [fedora-virt] Fedora virt status Message-ID: <1243271215.32745.38.camel@blaa> F11 Release =========== The F11 release is imminent. Happily, there are no open bugs remaining on the virtualization blocker list. However, the blocker list for the entire distro still has a significant number of open bugs so Release Engineering has announced a one week slip of the release date: http://www.redhat.com/archives/fedora-devel-list/2009-May/msg01585.html We still have a massive list of bugs on the virt target list for F-11: https://bugzilla.redhat.com/showdependencytree.cgi?id=F11VirtTarget&hide_resolved=1 These bugs haven't been forgotten about and we hope to fix many of them in post-release updates. Please do jump in and help out! F12 Features ============ Dan Berrange has added three new feature pages for F12: http://fedoraproject.org/wiki/Features/VirtAppliances http://fedoraproject.org/wiki/Features/VirtPrivileges http://fedoraproject.org/wiki/Features/VirtTCK the full list of features is available here: http://fedoraproject.org/wiki/Category:F12_Virt_Features Many more features are planned for F12 and we should see more feature pages posted to the wiki as folks begin work on them. Is it 'Raw Virt' or 'Virt Hide'? ================================ As discussed previously on the fedora-virt list, we're going to maintain a 'virt preview' yum repository for Fedora 11 users who wish to test out the latest virtualization bits from Fedora 12: https://www.redhat.com/archives/fedora-virt/2009-May/msg00074.html This repository isn't intended to replace rawhide - e.g. if we had this repository during F-11 development, it wouldn't have helped with sVirt testing. We hope it will allow people who are interested in virtualization - but who would not consider running rawhide - to test some of the latest virtualization features. Work remains in order to ensure that this repository is as useful as possible and volunteers of help are most welcome. Virt Test Day ============= The F11 Virtualization Test Day was held on May 7th and was a great success. Over 30 people turned out and a whole heap of high quality bug reports were filed: https://www.redhat.com/archives/fedora-virt/2009-May/msg00046.html Our hope is that the test cases and wiki structure used for this test day will be re-used for similar test days for F-12 and beyond. FWN === Dale Bewley continues his excellent series of virtualization sections in Fedora Weekly News: http://fedoraproject.org/wiki/FWN/Issue175#Virtualization http://fedoraproject.org/wiki/FWN/Issue176#Virtualization http://fedoraproject.org/wiki/FWN/Issue177#Virtualization FUDCon ====== The 2009 Berlin FUDCon is coming up at the end of June: http://fedoraproject.org/wiki/FUDConBerlin2009 Rich Jones and I will be there giving a talk and we're looking forward to meeting any other Fedora virt folks who happen to be there too. http://fedoraproject.org/wiki/Rich_Jones_and_Mark_McLoughlin_-_Virtualization virt-inspector ============== Rich Jones announced the available of a new tool called virt-inspector: https://www.redhat.com/archives/fedora-virt/2009-May/msg00073.html This is a tool based around libguestfs which can inspect a virtual machine disk image and tell you some interesting things about what's inside it. Some of the things it can tell you: - What operating system(s) are installed, and what distros and versions. It currently covers RHEL releases, Fedora releases, Debian releases, and has limited support for Windows. - How disk partitions are expected to be mounted (eg. /dev/sda1 -> /boot) - What applications are installed. - What kernel(s) are installed. - What kernel modules are installed. It can produce output in plain text (as a report) or in XML (to feed into other programs). This is currently available for F-11 through libguestfs-1.0.27 in updates-testing: https://admin.fedoraproject.org/updates/libguestfs-1.0.27-1.fc11.1 Xen Dom0 Kernels ================ Michael Young continues building Xen Dom0 kernels: http://www.redhat.com/archives/fedora-virt/2009-May/msg00099.html http://www.redhat.com/archives/fedora-virt/2009-May/msg00051.html Upstream, Jeremy Fitzhardinge continues to slug it out with Ingo Molnar to get his patches merged for 2.6.31. Latest progress here: http://lists.xensource.com/archives/html/xen-devel/2009-05/threads.html#00624 Bugs ==== Crikey! Streuth! DOOM-O-METER: 182 open bugs 3 weeks ago, 236 now. Yes, we have bugs. Lots of them! Seriously, we've had a huge surge in bugzilla activity since the test day. I've even left out most of the test day bugs below, since I have already posted details of those to the list. = New Bugs = == kvm == https://bugzilla.redhat.com/499210 OpenSolaris installer segfaults from OOM under KVM, bug not with unaccelerated QEMU Looks like some difference between KVM and TCG is causing the OpenSolaris installer to handle OOM differently. https://bugzilla.redhat.com/346151 Create a kvm user account and kvm group https://bugzilla.redhat.com/497341 Make the /dev/kvm device world accessible to all users by default https://bugzilla.redhat.com/500472 QEMU driver should run all QEMU VMs as non-root system account All bugs related to the VirtPrivileges feature for F-12. https://bugzilla.redhat.com/500866 virtio-net driver does not support ethtool drvinfo query The virtio-net driver should probably support some more ethtool queries. == qemu == https://bugzilla.redhat.com/499060 qemu-kvm with -vga std does not provide widescreen resolution in win xp guest Looks like a vgabios regression since qemu-0.9.1. Glauber pushed a rebase to a newer version from upstream, but that only seems to make matters worse. https://bugzilla.redhat.com/501545 "qemu-kvm -vga std" broken with vgabios 0.6c Details on how 0.6c makes things worse; it's been unpushed from F-11 but we've left it in F-12 for now. https://bugzilla.redhat.com/499542 qemu graphics corruption with -vga vmware Looks like -vga vmware is totally broken at the moment, that's assuming it isn't a guest driver issue. https://bugzilla.redhat.com/502058 qemu -no-kvm guest hangs at during timer setup; works with noapic https://bugzilla.redhat.com/502440 qemu -no-kvm guest oops during IO-APIC setup Looks like two possibly related APIC issue when running qemu-kvm -no-kvm. This preventing us from doing a "make check" during libguestfs build. https://bugzilla.redhat.com/501264 qemu segfaults for -net socket,listen=localhost:4567 Fixed upstream, Glauber has proposed the fix for the stable branch. https://bugzilla.redhat.com/499596 32 bit KVM guest hangs enabling NX protection; booting with -cpu qemu32 works We're pinning our hopes on the cpuid trimming patches fixing this, but it's not totally clear they will. https://bugzilla.redhat.com/501131 qemu segfault when VNC client disconnects It looks like this is an issue with the way qemu's vnc code handles I/O errrors. A hacky patch has been built for the reporter to verify that theory. https://bugzilla.redhat.com/500863 qemu -kernel/-initrd with 4Gb memory and F6 image results in "initramfs bad gzip magic" Daire Byrne, aka "-kernel man", found yet another bug with qemu-kvm -kernel. Glauber promptly fixed this upstream and it will be in the qemu-0.10.5 update we hope to push next week. https://bugzilla.redhat.com/501935 qemu-kvm -kernel should parse "vga=" cmdline option Daire also discovered an oddity whereby vga= option is ignored if you boot with -kernel. It turns out the bootloader implements this option and, so, qemu should implement it too. https://bugzilla.redhat.com/500317 Vista install fails in qemu guest (without kvm) == libvirt == https://bugzilla.redhat.com/499223 Add virt-install --virt-type option and make kvm the default In F-12, the plan is to no longer require --accelerate to be specified in order to use kvm. https://bugzilla.redhat.com/499970 F12: port libvirt to PolicyKit 1.0 Repeat after me: "We loves ABI changes". https://bugzilla.redhat.com/501934 libvirt bridge should not have IPv6 disabled https://bugzilla.redhat.com/499669 libvirt QEMU driver is using old pci_add/pci_del syntax PCI hotplug isn't currently working because qemu-kvm's monitor command syntax has changed and hasn't been updated to use that syntax. https://bugzilla.redhat.com/501550 virtio console device has lame PCI class info ajax reports that virtio-console's PCI class is bogus and proposes a patch to fix it. The patch has now been posted upstream. https://bugzilla.redhat.com/501328 Add support for virtio console in libvirt QEMU driver https://bugzilla.redhat.com/501720 upstart does not launch a login process on /dev/hvc0 in all cases The plan here is to enable virtio console for Linux guests by default and have them unconditionally configure it as a console. This will give us an "always available" console like we have for Xen. == virtinst == https://bugzilla.redhat.com/500320 Vista needs >15G; add "minimum disk size" to python-virtinst OS_TYPES table? https://bugzilla.redhat.com/502088 virt-clone.log does not get closed on exec == virt-manager == https://bugzilla.redhat.com/499812 host screensaver kicks in while virt-manager guest console is grabbed Dan reckons this is a problem for gnome-screensaver, but most likely this is just a fundamental limitation of X grabs and virt-manager will have to explicitly inhibit the screensaver. https://bugzilla.redhat.com/501922 virt-manager: NFS install of guest complains about insufficient privilege Another side effect of us not running virt-manager as root by default anymore. We should be able to use libvirt's storage management APIs to fix this. https://bugzilla.redhat.com/500940 virt-manager should delete failed guest images https://bugzilla.redhat.com/502204 virt-manager's dialog to connect an existing CD-ROM to an ISO does not use storage pool interface https://bugzilla.redhat.com/502333 virt-manager should show guest disk size in Details == libguestfs == https://bugzilla.redhat.com/501878 guestfish built-in commands like 'alloc' and 'help' don't autocomplete https://bugzilla.redhat.com/501882 /tmp/libguestfs temporary directory is not cleaned up https://bugzilla.redhat.com/501883 javadoc messed up in libguestfs java documentation https://bugzilla.redhat.com/502074 libguestfs checksum test (of SHA1) fails on i586 (running under qemu sans kvm) https://bugzilla.redhat.com/502309 libguestfs build failure - tests occasionally fail with 'ranlib: './libmlguestfs.a': No such file' Rich moved his libguestfs TODO list to bugzilla. These are the ones that are particularly interesting for Fedora, the rest are in the 'Virtualization Tools' bugzilla product. == xen == https://bugzilla.redhat.com/499850 RFE: Implement the Xen PV-on-HVM drivers We want to re-gain support for Xen paravirt devices in HVM guests, but we're not quite clear yet that anyone is going to get to it any time soon. https://bugzilla.redhat.com/501720 upstart does not launch a login process on /dev/hvc0 in all cases Fedora isn't always setting up a tty for Xen PV guests any more. https://bugzilla.redhat.com/458385 Running gdb in F-9 x86_64 guest on RHEL 5.2 x86_64 host crashes guest Jeremy Fitzhardinge has reportedly fixed this upstream so we should apply those patches to F-11. https://bugzilla.redhat.com/499611 Hotplugging vcpus to an F-11 64-bit Xen domU doesn't work Justin Forbes has proposed a udev script to automatically online CPUs as they are added. = Resolved Bugs = == kvm == https://bugzilla.redhat.com/492838 kvm: PTHREAD_PRIO_INHERIT mutexes fail to unlock sometimes with 32 bit i686/PAE guest kernel, works on i686/non-PAE Avi heroically figured out this bug after much faffing about on everyone else's part. It turns out to need two separate fixes for the paravirt MMU. https://bugzilla.redhat.com/499307 BUG() in kvm_handle_fault_on_reboot() on amd64 host during RHEL5.3 guest install There are over a hundred of these logged to kerneloops.org. Turns out to be caused by the VirtualBox kernel module. WONTFIX. == qemu == https://bugzilla.redhat.com/498405 qemu disk image corruption with qcow2/virtio Avi looked into this and came up with a workaround patch. Much discussion ensued upstream with Avi, Kevin and Gleb as they grappled to understand the code and in the end Gleb came up with a simple fix. The fix was included in qemu-kvm-0.10.4 which is now available in F11 updates-stable. https://bugzilla.redhat.com/499666 Reboot fails when using qemu -kernel Glauber posted a patch to qem-devel for this and built the fix as an F-11 update. https://bugzilla.redhat.com/494075 openbios bug causes qemu-system-ppc "invalid/unsupported opcode" failure This seems to be due to a change in binutils in F11. Pavel Roskin fixed the problem in upstream openbios and a openbios-1.0-1.fc11 update has been pushed with the fix. We expect there are more issues to be resolved here. == libvirt == https://bugzilla.redhat.com/493692 Creating new VM; SELinux prevents opening iso image of install media libvirt wasn't correctly re-labelling shared or read-only images when starting a guest. Dan Berrange came up with a patch for this issue and built just in time for the test day. https://bugzilla.redhat.com/460649 libvirtd requires restart in order to detect new capabilities like KVM support Cole came up with a patch for this, posted upstream and built in F-11, also in advance of the test day. https://bugzilla.redhat.com/499791 virt-manager does not escape special characters in ISO file name libvirt gets XML attribute escaping wrong. DV has fixed upstream and that fix is available in libvirt-0.6.2-9.fc11. https://bugzilla.redhat.com/499584 info balloon is repeated in qemu log once per second Fixed in libvirt-0.6.2-10.fc11 by only logging monitor output to the debug logs, not the domain log file. https://bugzilla.redhat.com/501912 libvirt/qemu does not activate bridge if no IP address is assigned This little oversight was fixed upstream recently and the fix has been cherry-picked for F-11 in libvirt-0.6.2-11.fc11. https://bugzilla.redhat.com/499698 'virsh destroy' destroys multiple VMs Dan Berrange posted a fix for this to the bug report which is now upstream and in F-11 updates-testing. == virt-manager == https://bugzilla.redhat.com/491683 virt-manager no longer auto adjusts to guest screen size Dan Berrange tracked this down to a recent change which explictly sets the window size in virt-manager. The fix has been pushed to update-testing in virt-manager-0.7.0-5.fc11. https://bugzilla.redhat.com/499145 virt-manager appears to ignore arch selection Cole implemented this and and the patch has been include in the virt-manager-0.7.0-5.fc11 update. https://bugzilla.redhat.com/499589 virt-manager console widget doesn't know how to collect a 'username' for authentication with VNC Dan's patch to implement this is in virt-manager-0.7.0-5.fc11. https://bugzilla.redhat.com/495128 "New VM" dialog box misspells "operating" as "opertaing" Another minor victory in the War on Spelling Mistakes. Even the translations have been purged of this evil typo. https://bugzilla.redhat.com/499685 RFE - include virt type in virt-manager Guest->Details tab It's very hard for users to tell whether they're guest is actually using KVM. This new UI added by Cole for F-12 will make it easier. == libguestfs == https://bugzilla.redhat.com/501761 libguestfs requires qemu, not qemu-system-x86 or qemu-system-ppc Rich continues his crusade to over-inflate the dependencies of Fedora packages. == xen == https://bugzilla.redhat.com/499621 F-11 Xen 64-bit domU stack traces when unmounting a disk Justin backported the fix for this to the F-11 kernel. https://bugzilla.redhat.com/499592 F-11 Xen 64-bit domU cannot be started with > 2047MB of memory Chris Lalancette came up with a fix for this issue and Justin pushed it to F-11. Jeremy Fitzhardinge plans to push a slightly different patch upstream. == anaconda == https://bugzilla.redhat.com/496298 F11 Beta :: under KVM anaconda detect cdrom media because udev doesn't probe for filesystem type This blocker appears to have finally been stomped. https://bugzilla.redhat.com/499194 anaconda fails to install grub in KVM guest It seems anaconda was briefly broken wrt. installing grub in a KVM guest for a while because of /dev/ being unmounted. Fixed in later builds. = Ongoing Bugs = == qemu == https://bugzilla.redhat.com/496627 qemu-kvm es1370 sound card emulation segfault https://bugzilla.redhat.com/495964 qemu locks up at shutdown with sdl audio driver Tom London confirmed that the pulseaudio backend resolves this issue. However, qemu will not default to the pulseaudio backend even if it is enabled. This needs to be addressed upstream. https://bugzilla.redhat.com/477035 EFI BIOS support in qemu Peter Jones has been looking into this and keeping notes here: https://fedoraproject.org/wiki/User:Pjones/KvmEFI https://bugzilla.redhat.com/494541 [KVM] E1000 PXE boot fails with "No IP address" It looks like building with PXE_DHCP_STRICT helps at least the reporter, so an update with this enabled has been pushed to updates-testing. == libvirt == https://bugzilla.redhat.com/490191 virt-manager cannot open VM serial console because of /dev/pty permissions Dan Berrange comments that this can't be fixed until virt-manager uses qemu:///session by default. See also the VirtPrivileges F12 feature. == virt-manager == https://bugzilla.redhat.com/478418 virt-manager prevents key combinations like alt-f from being sent to the guest https://bugzilla.redhat.com/499362 virt-viewer prevents key combinations like alt-f from being sent to the guest Dan Berrange describes how virt-manager and virt-viewer needs to intercept, and block, various key combinations before they are processed by gtk+ so that they can be sent to the guest. He has fixed this for virt-viewer, but the update needs pushing. From cochranb at speakeasy.net Tue May 26 01:14:25 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Mon, 25 May 2009 21:14:25 -0400 Subject: [fedora-virt] Hostdev XML Not Working (Libvirt) Message-ID: <4A1B4271.2020903@speakeasy.net> Why doesn't the hostdev xml shown below for a USB device work? I've shown the entire block for context. /usr/bin/qemu-kvm The USB device of interest to me is an FTDI FT232RL chip. When I plug the board into a USB port on the host, I want it to be passed through to my Ubuntu guest so that it can be seen as /dev/ttyUSB0. The guest doesn't see it. I took the product and vendor id numbers from the /var/log/messages output, and these might be in hexadecimal (and I should be writing e.g. . I might also be getting the physical handling of the device wrong: maybe it needs to be already plugged in at the time the Ubuntu guest is started up, or maybe I should start the guest first and then plug in the device? Advice appreciated. Bob Cochran From berrange at redhat.com Tue May 26 09:35:12 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 26 May 2009 10:35:12 +0100 Subject: [fedora-virt] Distributing Fedora 11 Created Virtual Machines In-Reply-To: <4A19F6F1.40601@speakeasy.net> References: <4A19F6F1.40601@speakeasy.net> Message-ID: <20090526093512.GB31808@redhat.com> On Sun, May 24, 2009 at 09:40:01PM -0400, Robert L Cochran wrote: > I have a virtual machine created on Fedora 11 which runs Ubuntu 9.04. I > have some more customizations to do with the machine, such as getting it > to speak to (FTDI 232R-based) USB devices. When that is done, I would > like to copy the virtual machine to a USB flash drive and postal mail > that to a teacher. The scenario I'm envisioning is that the teacher can > then import the virtual machine and play it on his or her computer -- > whatever that computer is, perhaps a Windows machine running VirtualBox > or VMWare. Not necessarily a machine running Fedora 11, you see. Now we > have my first question: can she run my virtual machine on another > virtual machine manager, which is not necessarily a Fedora product? There's no easy answer to that question. it really depends how you configured the virtual machine, and the OS installed inside it. For example if you configured 'virtio' disks, then you'll likely have trouble making it work in VMWare or VirtualBox. To a certain extent Linux can cope with hardware changing & automatically reconfigure itself, but it depends on exactly how well the distro in question is put together. The only way to be sure of success is to actually test your viurtual machine image in VirtalBox & VMWare yourself and fixing any portability problems you encounter. > To go on with the story I'd like the teacher to add her customizations > to the Ubuntu virtual machine, then she in turn will deploy that to > students. They would copy the virtual machine to their computers > (whatever those are) and play them on those computers. These will > probably be whatever computers the school has an d/or whatever computers > the student can afford to pay for. Could this work as well? Again you're looking at the same problem of whether the guest OS will automatically reconfigure itself for the different hardware or not, but with the added problem that you'll be unable to test it yourself. You also have the issue of how exactly they'll get the image off the USB stick and onto the physical machine. This is not an altogether impossible problem, but do be aware of the scope of potential problems you may encounter. For Fedora we distribute a Live CD image which is put together such that it'll boot successfull under any common hardware, physical or virtual. It also includes a mini installer, which lets you copy the LiveCD image over to the real OS. Rather than building a custom virtual machine image and then copying around the disk image, I think you'd be better off taking a Live CD image, and tailoring what's included in that. Since that already has been designed with hardware portability in mind. > Suppose the answer is no they cannot play a Fedora 11-created virtual > machine on their own virtualization software. Is it possible to take the > virtual machine file and clone (part of?) it to a hard drive such that > Ubuntu can be booted from the hard drive? Ubuntu questions are best asked on Ubuntu mailing lists. For Fedora we have a Live CD image that can be tailored to do the kind of thing you are after, more or less just be tweaking the kickstart file used to build it. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From ruben at rubenkerkhof.com Tue May 26 10:04:07 2009 From: ruben at rubenkerkhof.com (Ruben Kerkhof) Date: Tue, 26 May 2009 12:04:07 +0200 Subject: [fedora-virt] Backport of zfs patches. Message-ID: Hi all, I've hit a few issues with booting an Opensolaris domU (build 114) with the current Xen in Fedora. Would it be possible to backport patches hg19323 and hg19322? Otherwise I can wait for Xen 3.4 to hit rawhide, what's the timeframe for that? Kind regards, Ruben Kerkhof From rjones at redhat.com Tue May 26 14:54:22 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 15:54:22 +0100 Subject: [fedora-virt] how to build iso from virt-p2v In-Reply-To: <200905250921375077225@huawei.com> References: <200905250921375077225@huawei.com> Message-ID: <20090526145422.GA30514@amd.home.annexia.org> On Mon, May 25, 2009 at 09:21:37AM +0800, Arthur Wang wrote: > Hello, > I have a trouble when build virt-p2v-0.9.9.tar.gz to generate an iso file. Search for help > > I have install all the packet that virt-p2v depend on, and i run the follow commands successfully > configure > make rpm > But when i enter "make build" command, wait minutes, i got the follow error: > Setting maximal mount count to -1 > Setting interval between checks to 0 seconds > Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: released. Please verify its path and try again > make: *** [build] error 1 > Could anyone tell me where it want to get repository metadata , and which tools or command it use to get? Just to check that you do have network access where you are trying to build, right? The other things to change are the CD base version, eg: ./configure --enable-cd-base=10 which will base everything on Fedora 10. Have a look also at the generated kickstart file, p2v.ks, to see what it contains. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From rjones at redhat.com Tue May 26 14:56:26 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 15:56:26 +0100 Subject: [fedora-virt] Distributing Fedora 11 Created Virtual Machines In-Reply-To: <4A19F6F1.40601@speakeasy.net> References: <4A19F6F1.40601@speakeasy.net> Message-ID: <20090526145626.GB30514@amd.home.annexia.org> In addition to Dan's answer, I'm looking into this area too. There is no easy way to do it at the moment, but currently a virtual machine image is more likely to work if it's fully virtualized, ie. uses no virtio or other PV drivers. Sadly such a virtual machine will also be much slower. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Tue May 26 14:56:50 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 15:56:50 +0100 Subject: [fedora-virt] Newbie Q? Mailing-List Support. In-Reply-To: <4A1A8A19.7040001@gmail.com> References: <4A1A8A19.7040001@gmail.com> Message-ID: <20090526145650.GC30514@amd.home.annexia.org> On Mon, May 25, 2009 at 01:07:53PM +0100, Frank Murphy (Frankly3d) wrote: > Is the mailing list support just for people who run guests on Fedora, > or can it be used with Fedora Guest on CentOS Base? > > Thinking of putting a virt box or two together. Yup, this is a good place to post such questions. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html From rjones at redhat.com Tue May 26 15:01:17 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 16:01:17 +0100 Subject: [fedora-virt] Successful Compile of libguestfs 1.0.31 in Fedora 11 In-Reply-To: <4A1ACDA3.6020601@speakeasy.net> References: <4A1ACDA3.6020601@speakeasy.net> Message-ID: <20090526150117.GD30514@amd.home.annexia.org> On Mon, May 25, 2009 at 12:56:03PM -0400, Robert L Cochran wrote: > I've been looking at the postings here more closely now that I have a > Ubuntu guest running, and I decided to compile libguestfs-1.0.31 in > Fedora 11. It took some time for me to get the dependencies installed. > Question: what packages do I need for the ruby bindings? What package > represents rake? How about OCaml bindings -- what packages are needed to > get it to build? If you're just trying to get libguestfs to run on Fedora 11, then I strongly suggest starting off with the packages we are building for Fedora 11. You should be able to get them from updates, ie: yum --enablerepo=updates install libguestfs guestfish ocaml-libguestfs-devel [etc...] and if that doesn't work, you can get them from Koji: http://koji.fedoraproject.org/koji/packageinfo?packageID=8391 Select the latest package with '.fc11' in the name. > When building this file as part of `make`: > > initramfs.fedora-11.x86_64.img > > I got an awesome number of errors related to my not being the root user, > but I assume those errors are okay since I built as an ordinary user. You'll get a few errors. Lots of these errors go away if you use the newest possible version of fakechroot (eg. 2.9-22). > Also the perl documentation process came up with a large number of > errors. > > After doing `make check` I got this result: > > Test boot completed after 10 iterations. > PASS: test-bootbootboot.sh > ================== > All 2 tests passed > ================== That looks very good. > Am I ready for 'make install' -- is libguestfs built sufficiently well > for that? Although I would like to get the ruby and OCaml bindings to > build. 'make install' ought to work. We never use it because we build RPMs and install those. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From rjones at redhat.com Tue May 26 15:02:51 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 16:02:51 +0100 Subject: [fedora-virt] Hostdev XML Not Working (Libvirt) In-Reply-To: <4A1B4271.2020903@speakeasy.net> References: <4A1B4271.2020903@speakeasy.net> Message-ID: <20090526150251.GE30514@amd.home.annexia.org> On Mon, May 25, 2009 at 09:14:25PM -0400, Robert L Cochran wrote: > Why doesn't the hostdev xml shown below for a USB device work? I've > shown the entire block for context. > > > > > > hostdev was only added to libvirt recently -- earlier this month -- so my best guess is that you need a newer version of libvirt. Get it from Fedora updates or Koji. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From rjones at redhat.com Tue May 26 15:03:39 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 16:03:39 +0100 Subject: [fedora-virt] Backport of zfs patches. In-Reply-To: References: Message-ID: <20090526150339.GF30514@amd.home.annexia.org> On Tue, May 26, 2009 at 12:04:07PM +0200, Ruben Kerkhof wrote: > Hi all, > > I've hit a few issues with booting an Opensolaris domU (build 114) > with the current Xen in Fedora. > Would it be possible to backport patches hg19323 and hg19322? > Otherwise I can wait for Xen 3.4 to hit rawhide, what's the timeframe > for that? Best thing might be to file some bugs containing links to the patches you need: http://bugzilla.redhat.com/ Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Tue May 26 15:04:52 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 16:04:52 +0100 Subject: [fedora-virt] Successful Compile of libguestfs 1.0.31 in Fedora 11 In-Reply-To: <20090526150117.GD30514@amd.home.annexia.org> References: <4A1ACDA3.6020601@speakeasy.net> <20090526150117.GD30514@amd.home.annexia.org> Message-ID: <20090526150452.GG30514@amd.home.annexia.org> On Tue, May 26, 2009 at 04:01:17PM +0100, Richard W.M. Jones wrote: > On Mon, May 25, 2009 at 12:56:03PM -0400, Robert L Cochran wrote: > > I got an awesome number of errors related to my not being the root user, > > but I assume those errors are okay since I built as an ordinary user. > > You'll get a few errors. Lots of these errors go away if you use the > newest possible version of fakechroot (eg. 2.9-22). BTW if the test suite passes, then it's unlikely that the errors are important enough to worry about. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From ruben at rubenkerkhof.com Tue May 26 17:57:40 2009 From: ruben at rubenkerkhof.com (Ruben Kerkhof) Date: Tue, 26 May 2009 19:57:40 +0200 Subject: [fedora-virt] Backport of zfs patches. In-Reply-To: <20090526150339.GF30514@amd.home.annexia.org> References: <20090526150339.GF30514@amd.home.annexia.org> Message-ID: On Tue, May 26, 2009 at 5:03 PM, Richard W.M. Jones wrote: > On Tue, May 26, 2009 at 12:04:07PM +0200, Ruben Kerkhof wrote: >> Hi all, >> >> I've hit a few issues with booting an Opensolaris domU (build 114) >> with the current Xen in Fedora. >> Would it be possible to backport patches hg19323 and hg19322? >> Otherwise I can wait for Xen 3.4 to hit rawhide, what's the timeframe >> for that? > > Best thing might be to file some bugs containing links to > the patches you need: > > http://bugzilla.redhat.com/ > > Rich. Done, thanks. https://bugzilla.redhat.com/502644 Ruben From cochranb at speakeasy.net Tue May 26 21:00:01 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Tue, 26 May 2009 17:00:01 -0400 Subject: [Fwd: Re: [fedora-virt] Hostdev XML Not Working (Libvirt)] Message-ID: <4A1C5851.9050808@speakeasy.net> -------- Original Message -------- Subject: Re: [fedora-virt] Hostdev XML Not Working (Libvirt) Date: Tue, 26 May 2009 16:51:04 -0400 From: Robert L Cochran To: Richard W.M. Jones On 05/26/2009 11:02 AM, Richard W.M. Jones wrote: > On Mon, May 25, 2009 at 09:14:25PM -0400, Robert L Cochran wrote: > >> Why doesn't the hostdev xml shown below for a USB device work? I've >> shown the entire block for context. >> >> >> >> >> >> >> > > hostdev was only added to libvirt recently -- earlier this month -- so > my best guess is that you need a newer version of libvirt. Get it > from Fedora updates or Koji. > > Rich. > > I guess I need to explore the world of Koji. I don't know how to grab Koji updates. I think I'm set up to get Fedora 11 updates off of yum. I generally hunt around for new updates at least twice daily. $ rpm -q libvirt libvirt-0.6.2-8.fc11.x86_64 Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From cochranb at speakeasy.net Tue May 26 21:09:20 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Tue, 26 May 2009 17:09:20 -0400 Subject: [fedora-virt] Successful Compile of libguestfs 1.0.31 in Fedora 11 In-Reply-To: <20090526150117.GD30514@amd.home.annexia.org> References: <4A1ACDA3.6020601@speakeasy.net> <20090526150117.GD30514@amd.home.annexia.org> Message-ID: <4A1C5A80.2030901@speakeasy.net> On 05/26/2009 11:01 AM, Richard W.M. Jones wrote: > On Mon, May 25, 2009 at 12:56:03PM -0400, Robert L Cochran wrote: > >> I've been looking at the postings here more closely now that I have a >> Ubuntu guest running, and I decided to compile libguestfs-1.0.31 in >> Fedora 11. It took some time for me to get the dependencies installed. >> Question: what packages do I need for the ruby bindings? What package >> represents rake? How about OCaml bindings -- what packages are needed to >> get it to build? >> > > If you're just trying to get libguestfs to run on Fedora 11, then I > strongly suggest starting off with the packages we are building for > Fedora 11. > > You should be able to get them from updates, ie: > > yum --enablerepo=updates install libguestfs guestfish ocaml-libguestfs-devel [etc...] > > and if that doesn't work, you can get them from Koji: > > http://koji.fedoraproject.org/koji/packageinfo?packageID=8391 > > Select the latest package with '.fc11' in the name. > > >> When building this file as part of `make`: >> >> initramfs.fedora-11.x86_64.img >> >> I got an awesome number of errors related to my not being the root user, >> but I assume those errors are okay since I built as an ordinary user. >> > > You'll get a few errors. Lots of these errors go away if you use the > newest possible version of fakechroot (eg. 2.9-22). > > >> Also the perl documentation process came up with a large number of >> errors. >> >> After doing `make check` I got this result: >> >> Test boot completed after 10 iterations. >> PASS: test-bootbootboot.sh >> ================== >> All 2 tests passed >> ================== >> > > That looks very good. > > >> Am I ready for 'make install' -- is libguestfs built sufficiently well >> for that? Although I would like to get the ruby and OCaml bindings to >> build. >> > > 'make install' ought to work. We never use it because we build RPMs > and install those. > > Rich. > > Okay, thanks for pointing me to Koji. Can I set up a yum repo for it? How do I code it? Also, can I set up a repo for Fedora 12 updates in Fedora 11, using Mark's suggested repo coding? Or has the location for Fedora 12 updates changed to Koji? If I take in Fedora 12 versions, will I risk my virtual machines dropping dead? Thanks Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From rjones at redhat.com Tue May 26 21:20:53 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 22:20:53 +0100 Subject: [Fwd: Re: [fedora-virt] Hostdev XML Not Working (Libvirt)] In-Reply-To: <4A1C5851.9050808@speakeasy.net> References: <4A1C5851.9050808@speakeasy.net> Message-ID: <20090526212053.GA32646@amd.home.annexia.org> On Tue, May 26, 2009 at 05:00:01PM -0400, Robert L Cochran wrote: [...] Before looking at Koji, make sure updated packages aren't in Fedora updates-testing or in the new Rawhide virt repo: yum --enablerepo=updates-testing install [...] or see here: http://www.redhat.com/archives/fedora-virt/2009-May/msg00074.html > I guess I need to explore the world of Koji. I don't know how to grab > Koji updates. There are two ways to do it. Manually, go here: http://koji.fedoraproject.org/koji/ Type the package name you want into the box in the top right corner, for example 'libvirt' takes you here: http://koji.fedoraproject.org/koji/packageinfo?packageID=91 then click on the version you want to try out, download the RPMs for your architecture, and install them manually (including resolving any dependencies manually). The alternate, automatic way to do this is to create a file /etc/yum.repos.d/koji.repo which should contain: [koji] name=Koji baseurl=http://koji.fedoraproject.org/static-repos/dist-f12-build-current/x86_64/ enabled=0 gpgcheck=0 keepcache=0 Replace 'f12' with your base distro, and 'x86_64' with your architecture. You can browse the Koji static repos here: http://koji.fedoraproject.org/static-repos/ I _do not_ recommend that you enable this repository by default, but instead enable it just when you need it, thus: yum --enablerepo=koji install [...] Installing packages from Koji can be quite dangerous to your system because you're on the bleeding edge of the bleeding edge, testing packages that have just been built before anyone else. Maybe even packages that the maintainer knows are broken and doesn't want to 'escape' into the wild. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Tue May 26 21:24:01 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 26 May 2009 22:24:01 +0100 Subject: [fedora-virt] Successful Compile of libguestfs 1.0.31 in Fedora 11 In-Reply-To: <20090526150117.GD30514@amd.home.annexia.org> References: <4A1ACDA3.6020601@speakeasy.net> <20090526150117.GD30514@amd.home.annexia.org> Message-ID: <20090526212401.GB32646@amd.home.annexia.org> On Tue, May 26, 2009 at 04:01:17PM +0100, Richard W.M. Jones wrote: > yum --enablerepo=updates install libguestfs guestfish ocaml-libguestfs-devel [etc...] That might be 'yum --enablerepo=updates-testing [...]' Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From cochranb at speakeasy.net Tue May 26 21:50:03 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Tue, 26 May 2009 17:50:03 -0400 Subject: [fedora-virt] Distributing Fedora 11 Created Virtual Machines In-Reply-To: <20090526093512.GB31808@redhat.com> References: <4A19F6F1.40601@speakeasy.net> <20090526093512.GB31808@redhat.com> Message-ID: <4A1C640B.40302@speakeasy.net> On 05/26/2009 05:35 AM, Daniel P. Berrange wrote: > On Sun, May 24, 2009 at 09:40:01PM -0400, Robert L Cochran wrote: > >> I have a virtual machine created on Fedora 11 which runs Ubuntu 9.04. I >> have some more customizations to do with the machine, such as getting it >> to speak to (FTDI 232R-based) USB devices. When that is done, I would >> like to copy the virtual machine to a USB flash drive and postal mail >> that to a teacher. The scenario I'm envisioning is that the teacher can >> then import the virtual machine and play it on his or her computer -- >> whatever that computer is, perhaps a Windows machine running VirtualBox >> or VMWare. Not necessarily a machine running Fedora 11, you see. Now we >> have my first question: can she run my virtual machine on another >> virtual machine manager, which is not necessarily a Fedora product? >> > > There's no easy answer to that question. it really depends how you > configured the virtual machine, and the OS installed inside it. For > example if you configured 'virtio' disks, then you'll likely have > trouble making it work in VMWare or VirtualBox. To a certain extent > Linux can cope with hardware changing& automatically reconfigure > itself, but it depends on exactly how well the distro in question > is put together. The only way to be sure of success is to actually > test your viurtual machine image in VirtalBox& VMWare yourself > and fixing any portability problems you encounter. > > >> To go on with the story I'd like the teacher to add her customizations >> to the Ubuntu virtual machine, then she in turn will deploy that to >> students. They would copy the virtual machine to their computers >> (whatever those are) and play them on those computers. These will >> probably be whatever computers the school has an d/or whatever computers >> the student can afford to pay for. Could this work as well? >> > > Again you're looking at the same problem of whether the guest OS will > automatically reconfigure itself for the different hardware or not, > but with the added problem that you'll be unable to test it yourself. > You also have the issue of how exactly they'll get the image off the > USB stick and onto the physical machine. > > > This is not an altogether impossible problem, but do be aware of the > scope of potential problems you may encounter. For Fedora we > distribute a Live CD image which is put together such that it'll > boot successfull under any common hardware, physical or virtual. > It also includes a mini installer, which lets you copy the LiveCD > image over to the real OS. Rather than building a custom virtual > machine image and then copying around the disk image, I think you'd > be better off taking a Live CD image, and tailoring what's included > in that. Since that already has been designed with hardware portability > in mind. > > >> Suppose the answer is no they cannot play a Fedora 11-created virtual >> machine on their own virtualization software. Is it possible to take the >> virtual machine file and clone (part of?) it to a hard drive such that >> Ubuntu can be booted from the hard drive? >> > > Ubuntu questions are best asked on Ubuntu mailing lists. For Fedora we > have a Live CD image that can be tailored to do the kind of thing you > are after, more or less just be tweaking the kickstart file used to > build it. > > Daniel > Thank you, Dan, for taking the time to answer this for me. I guess I'm going to have to greatly simplify this because I've never met the other person, and I'm not sure how much fiddling and adjusting he will accept from me. And there's always the issue that I know little of Fedora's virtualization internals. I think my best bet might be to set Ubuntu up on a laptop and then loan him the laptop for a while. I'll see if he can accept Fedora 11, too. A little cloning here, a little cloning there... Bob From wangjihai at huawei.com Wed May 27 01:44:25 2009 From: wangjihai at huawei.com (Arthur Wang) Date: Wed, 27 May 2009 09:44:25 +0800 Subject: [fedora-virt] how to build iso from virt-p2v References: <200905250921375077225@huawei.com> Message-ID: <200905270944249344637@huawei.com> Hello, The p2v.ks contains the follows: lang en_US.UTF-8 keyboard us timezone US/Eastern auth --useshadow --enablemd5 selinux --enforcing firewall --disabled # This should enable the serial console. However it requires # livecd-tools 018 or above (is ignored on earlier versions). bootloader --append="console=tty0 console=ttyS0,9600n8" # Basic Fedora repository and updates. repo --name=released --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/releases/9/Everything/i386/os/ repo --name=updates --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/updates/9/i386/ #repo --name=released --mirrorlist=http://mirrors.fedoraproject.org/mirrorlist?repo=fedora-9&arch=i386 # A local repository if selected by ./configure --enable-localrepo=... # The p2vrepo subdirectory, containing virt-p2v RPM. repo --name=p2v --baseurl=file:///home/w00135343/virt-p2v--devel/p2vrepo %packages --excludedocs bash kernel syslinux passwd policycoreutils chkconfig authconfig rootfiles # For unicode console support. kbd # dd, sleep, sync, etc. coreutils # For device mapping device-mapper util-linux-ng module-init-tools gzip kpartx # For remote communications openssh-clients nc # For the shell script grep sed And i guess it is the repo command to cause this error, for the two line as follows repo --name=released --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/releases/9/Everything/i386/os/ repo --name=updates --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/updates/9/i386/ the reason is that in my company i have to use proxy for network access , but i don't know which tools repo command comes from , and how to set its proxy. Could you give me some suggestion? thank you. 2009-05-27 Sender? Richard W.M. Jones Send time? 2009-05-26 23:22:17 Receiver? Arthur Wang Copyto? Fedora-virt Topic? Re: [fedora-virt] how to build iso from virt-p2v On Mon, May 25, 2009 at 09:21:37AM +0800, Arthur Wang wrote: > Hello, > I have a trouble when build virt-p2v-0.9.9.tar.gz to generate an iso file. Search for help > > I have install all the packet that virt-p2v depend on, and i run the follow commands successfully > configure > make rpm > But when i enter "make build" command, wait minutes, i got the follow error: > Setting maximal mount count to -1 > Setting interval between checks to 0 seconds > Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: released. Please verify its path and try again > make: *** [build] error 1 > Could anyone tell me where it want to get repository metadata , and which tools or command it use to get? Just to check that you do have network access where you are trying to build, right? The other things to change are the CD base version, eg: ./configure --enable-cd-base=10 which will base everything on Fedora 10. Have a look also at the generated kickstart file, p2v.ks, to see what it contains. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From charles at dyfis.net Wed May 27 02:05:38 2009 From: charles at dyfis.net (Charles Duffy) Date: Tue, 26 May 2009 21:05:38 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 Message-ID: <4A1C9FF2.2040805@dyfis.net> Howdy. CentOS 5's stock kernel has ext3 modularized; however, while make-initramfs.sh allows ext2.ko and ext4.ko modules to be included in the generated initramfs, no such exception is made for ext3.ko (or jbd.ko, on which it depends), leading to the following: ==== 5/140 test_zerofree_0 mount: /dev/hda1 on /: mount: unknown filesystem type 'ext3' test_zerofree_0 FAILED ==== Notably, ext4 is still ext4dev as of this kernel. From rjones at redhat.com Wed May 27 07:18:02 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 08:18:02 +0100 Subject: [fedora-virt] how to build iso from virt-p2v In-Reply-To: <200905270944249344637@huawei.com> References: <200905250921375077225@huawei.com> <200905270944249344637@huawei.com> Message-ID: <20090527071802.GG1267@amd.home.annexia.org> On Wed, May 27, 2009 at 09:44:25AM +0800, Arthur Wang wrote: > And i guess it is the repo command to cause this error, for the two line as follows > repo --name=released --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/releases/9/Everything/i386/os/ > repo --name=updates --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/updates/9/i386/ > the reason is that in my company i have to use proxy for network access , but i don't know which tools repo command comes from , and how to set its proxy. Could you give me some suggestion? thank you. This is a kickstart script, and the 'repo' command is part of kickstart. See: http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5/html/Installation_Guide/ch-kickstart2.html http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5/html/Installation_Guide/s1-kickstart2-options.html I actually have no idea how to specify a proxy for kickstart - it doesn't seem to be covered in the manual. However you might try setting the environment variables $http_proxy and/or $ftp_proxy to point to your internal proxy. eg: export http_proxy=http://11.22.33.44:3128/ Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From rjones at redhat.com Wed May 27 07:33:46 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 08:33:46 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <4A1C9FF2.2040805@dyfis.net> References: <4A1C9FF2.2040805@dyfis.net> Message-ID: <20090527073346.GA2492@amd.home.annexia.org> On Tue, May 26, 2009 at 09:05:38PM -0500, Charles Duffy wrote: > Howdy. > > CentOS 5's stock kernel has ext3 modularized; however, while > make-initramfs.sh allows ext2.ko and ext4.ko modules to be included in > the generated initramfs, no such exception is made for ext3.ko (or > jbd.ko, on which it depends), leading to the following: > > ==== > 5/140 test_zerofree_0 > mount: /dev/hda1 on /: mount: unknown filesystem type 'ext3' > test_zerofree_0 FAILED > ==== > > Notably, ext4 is still ext4dev as of this kernel. Does it work if you modify make-initramfs to add the extra module? If so then let me know and I'll add that to the list. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From rjones at redhat.com Wed May 27 07:37:38 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 08:37:38 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527073346.GA2492@amd.home.annexia.org> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> Message-ID: <20090527073738.GB2492@amd.home.annexia.org> On Wed, May 27, 2009 at 08:33:46AM +0100, Richard W.M. Jones wrote: > On Tue, May 26, 2009 at 09:05:38PM -0500, Charles Duffy wrote: > > Howdy. > > > > CentOS 5's stock kernel has ext3 modularized; however, while > > make-initramfs.sh allows ext2.ko and ext4.ko modules to be included in > > the generated initramfs, no such exception is made for ext3.ko (or > > jbd.ko, on which it depends), leading to the following: > > > > ==== > > 5/140 test_zerofree_0 > > mount: /dev/hda1 on /: mount: unknown filesystem type 'ext3' > > test_zerofree_0 FAILED > > ==== > > > > Notably, ext4 is still ext4dev as of this kernel. > > Does it work if you modify make-initramfs to add the extra module? > > If so then let me know and I'll add that to the list. BTW that file is generated so you have to do: edit make-initramfs.sh.in make make-initramfs.sh make For some reason the dependencies don't always rebuild the script, even though we added extra dependencies to handle this situation ... Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From kraxel at redhat.com Wed May 27 07:55:29 2009 From: kraxel at redhat.com (Gerd Hoffmann) Date: Wed, 27 May 2009 09:55:29 +0200 Subject: [fedora-virt] Distributing Fedora 11 Created Virtual Machines In-Reply-To: <20090526145626.GB30514@amd.home.annexia.org> References: <4A19F6F1.40601@speakeasy.net> <20090526145626.GB30514@amd.home.annexia.org> Message-ID: <4A1CF1F1.9020705@redhat.com> On 05/26/09 16:56, Richard W.M. Jones wrote: > In addition to Dan's answer, I'm looking into this area too. > > There is no easy way to do it at the moment, but currently a virtual > machine image is more likely to work if it's fully virtualized, > ie. uses no virtio or other PV drivers. Sadly such a virtual machine > will also be much slower. You can get the guest boot from almost anything with a few simple rules. First make sure you don't use device names in /etc/fstab. Mount by label/uuid or use lvm, so the guest can find the root filesystem no matter where it lives (ide/scsi/virtio/...). The default fedora partitioning scheme (lvm) should work fine. Second make sure the initrd loads the drivers needed. You can stick a list into /etc/sysconfig/mkinitrd (MODULES="..."). There isn't that much possible storage hardware one can find in a virtual machine (ide, a few (3?) scsi adapters, virtio, xenbus), so simply loading all possible drivers doesn't waste that much memory ... cheers, Gerd From rjones at redhat.com Wed May 27 14:13:05 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 15:13:05 +0100 Subject: [fedora-virt] Re: p2v question In-Reply-To: <9A0C4F19-FBAE-4588-B1A6-0AD9AAE91C60@gmail.com> References: <9A0C4F19-FBAE-4588-B1A6-0AD9AAE91C60@gmail.com> Message-ID: <20090527141305.GA4642@amd.home.annexia.org> On Wed, May 27, 2009 at 04:03:50PM +0200, Kevin Stabel wrote: > I subscribed to et-mgmt-tools and sent an inquiry already but i don't Ah I wasn't paying much attention to that list. I'm looking at fedora-virt mostly. > see anything coming back to me (including my own post) ... so if you > don't mind, i would like to ask you directly ... if that's okay. > > Can i convert or use the resulting images with vmware in any way? I've > tried qemu-tools but that won't work, it only creates ide disks and i > can't convert those. We have a separate tool in Fedora called virt-convert. I don't know specifically if it will work on the virt-p2v images, but worth a try. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.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 From crobinso at redhat.com Wed May 27 14:18:37 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 27 May 2009 10:18:37 -0400 Subject: [fedora-virt] Hostdev XML Not Working (Libvirt) In-Reply-To: <20090526150251.GE30514@amd.home.annexia.org> References: <4A1B4271.2020903@speakeasy.net> <20090526150251.GE30514@amd.home.annexia.org> Message-ID: <4A1D4BBD.4020501@redhat.com> Richard W.M. Jones wrote: > On Mon, May 25, 2009 at 09:14:25PM -0400, Robert L Cochran wrote: >> Why doesn't the hostdev xml shown below for a USB device work? I've >> shown the entire block for context. >> >> >> >> >> >> > > hostdev was only added to libvirt recently -- earlier this month -- so > my best guess is that you need a newer version of libvirt. Get it > from Fedora updates or Koji. > PCI was only added within the past few months: USB hostdev support has been available since September (Libvirt 0.4.5). - Cole From crobinso at redhat.com Wed May 27 14:23:02 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 27 May 2009 10:23:02 -0400 Subject: [fedora-virt] Hostdev XML Not Working (Libvirt) In-Reply-To: <4A1B4271.2020903@speakeasy.net> References: <4A1B4271.2020903@speakeasy.net> Message-ID: <4A1D4CC6.1010102@redhat.com> Robert L Cochran wrote: > Why doesn't the hostdev xml shown below for a USB device work? I've > shown the entire block for context. > > > /usr/bin/qemu-kvm > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > The USB device of interest to me is an FTDI FT232RL chip. When I plug > the board into a USB port on the host, I want it to be passed through to > my Ubuntu guest so that it can be seen as /dev/ttyUSB0. The guest > doesn't see it. I took the product and vendor id numbers from the > /var/log/messages output, and these might be in hexadecimal (and I > should be writing e.g. . I might also be getting the > physical handling of the device wrong: maybe it needs to be already > plugged in at the time the Ubuntu guest is started up, or maybe I should > start the guest first and then plug in the device? > > Advice appreciated. > > Bob Cochran > For future reference, easiest way to get device product + vendor info is to use 'lsusb'. I think you need to use the hex format you mention above to get this to work correctly ( ...). Also, make sure that the device is still listed in the domain xml when you use 'virsh dumpxml ': this will guarantee that libvirt is at least recognizing the new xml. You can also check /var/log/libvirt/qemu/.log to see what QEMU command line libvirt is generating, which helps debugging these issues. - Cole From rjones at redhat.com Wed May 27 14:27:04 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 15:27:04 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> Message-ID: <20090527142704.GB4642@amd.home.annexia.org> [Please keep fedora-virt CC'd on these ...] On Wed, May 27, 2009 at 09:18:33AM -0500, Charles Duffy wrote: > On Wed, May 27, 2009 at 2:37 AM, Richard W.M. Jones wrote: > > Does it work if you modify make-initramfs to add the extra module? > > I can't give a full reply until I'm back at the office (no remote access > atm), but editing make-initramfs.sh.in did indeed resolve that particular > issue. Could you check this patch to see if it should work: http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=213988115a8fd25266050228ac5184264f56baf8 > However, the logic for skipping the zerofree test if > /usr/sbin/zerofree wasn't found inside the guest wasn't working, so it still > proceeded to fail even with working ext3 support. I'll take a look at this one on my CentOS machine ... > Also, the CentOS 5 %check script was trying to use test-command without > compiling it first. That'll be a bug. Do you mean the EPEL package or the contributed CentOS spec file (in contrib/)? Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html From charles at dyfis.net Wed May 27 14:33:39 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 09:33:39 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527142704.GB4642@amd.home.annexia.org> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> Message-ID: On Wed, May 27, 2009 at 9:27 AM, Richard W.M. Jones wrote: > [Please keep fedora-virt CC'd on these ...] Oops -- using webmail, not fully accustomed to the UI. Could you check this patch to see if it should work: > > http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h= > 213988115a8fd25266050228ac5184264f56baf8 It's effectively the same thing I was doing, so it looks good. > > > However, the logic for skipping the zerofree test if > > /usr/sbin/zerofree wasn't found inside the guest wasn't working, so it > still > > proceeded to fail even with working ext3 support. > > I'll take a look at this one on my CentOS machine ... Thanks! > > Also, the CentOS 5 %check script was trying to use test-command without > > compiling it first. > > That'll be a bug. Do you mean the EPEL package or the contributed > CentOS spec file (in contrib/)? The latter. -------------- next part -------------- An HTML attachment was scrubbed... URL: From rjones at redhat.com Wed May 27 15:05:12 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 16:05:12 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> Message-ID: <20090527150512.GA5472@amd.home.annexia.org> On Wed, May 27, 2009 at 09:33:39AM -0500, Charles Duffy wrote: > On Wed, May 27, 2009 at 9:27 AM, Richard W.M. Jones wrote: > > That'll be a bug. Do you mean the EPEL package or the contributed > > CentOS spec file (in contrib/)? > > The latter. I guess I ought to mark that deprecated ... Have a look at the EPEL spec file, and the EPEL builds in general: http://cvs.fedoraproject.org/viewvc/rpms/libguestfs/EL-5/ BTW at the moment libguestfs from git is actually broken on CentOS 5.3. I'm going to commit some fixes for it as I fix them. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Wed May 27 16:03:21 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 17:03:21 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> Message-ID: <20090527160321.GB5472@amd.home.annexia.org> I checked in some fixes for CentOS to the git repo. I'll try to build an EPEL package later this afternoon. On Wed, May 27, 2009 at 09:18:33AM -0500, Charles Duffy wrote: > However, the logic for skipping the zerofree test if > /usr/sbin/zerofree wasn't found inside the guest wasn't working, so it still > proceeded to fail even with working ext3 support. > Also, the CentOS 5 %check script was trying to use test-command without > compiling it first. The above can be fixed by doing: make SUBDIRS="" check Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From rjones at redhat.com Wed May 27 17:29:33 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 18:29:33 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527150512.GA5472@amd.home.annexia.org> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> Message-ID: <20090527172933.GA6243@amd.home.annexia.org> On Wed, May 27, 2009 at 04:05:12PM +0100, Richard W.M. Jones wrote: > http://cvs.fedoraproject.org/viewvc/rpms/libguestfs/EL-5/ I haven't actually managed to complete an EPEL build, _but_ I'm reasonably confident that libguestfs 1.0.34 + the updated spec file above should work. It even runs the tests ... Note that if you enable the epel-testing repository you'll get qemu-0.10.5 which works with libguestfs w/o any patches or special builds/wrappers required. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From charles at dyfis.net Wed May 27 19:34:27 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 14:34:27 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527172933.GA6243@amd.home.annexia.org> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> Message-ID: <4A1D95C3.5030602@dyfis.net> Richard W.M. Jones wrote: > Note that if you enable the epel-testing repository you'll get > qemu-0.10.5 which works with libguestfs w/o any patches or special > builds/wrappers required. > Out of curiosity -- what do you need from qemu-0.10.5 (as opposed to qemu-kvm-0.10.4, or kvm-86)? From rjones at redhat.com Wed May 27 19:45:53 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 20:45:53 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <4A1D95C3.5030602@dyfis.net> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> Message-ID: <20090527194553.GC5472@amd.home.annexia.org> On Wed, May 27, 2009 at 02:34:27PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> Note that if you enable the epel-testing repository you'll get >> qemu-0.10.5 which works with libguestfs w/o any patches or special >> builds/wrappers required. >> > Out of curiosity -- what do you need from qemu-0.10.5 (as opposed to > qemu-kvm-0.10.4, or kvm-86)? Several things - In the qemu 0.10 series in general, we need the upstream vmchannel support: http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html qemu 0.10.4 wouldn't even boot the libguestfs appliance. For whatever unknown reason it does appear that qemu 0.10.5 fixes this. I don't know what particular change fixes this. Generally KVM works, again specific versions sometimes have problems. Really the best thing to do is to run 'make check' [or for EPEL, 'make SUBDIRS= check'] and if that passes then the qemu/kvm version is good. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From charles at dyfis.net Wed May 27 19:52:05 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 14:52:05 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527194553.GC5472@amd.home.annexia.org> References: <4A1C9FF2.2040805@dyfis.net> <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> <20090527194553.GC5472@amd.home.annexia.org> Message-ID: <4A1D99E5.8060605@dyfis.net> Richard W.M. Jones wrote: > qemu 0.10.4 wouldn't even boot the libguestfs appliance. For whatever > unknown reason it does appear that qemu 0.10.5 fixes this. I don't > know what particular change fixes this. > qemu-kvm-0.10.4 (as opposed to upstream qemu-0.10.4, which I haven't tried) has passed "make check" for me with a prior libguestfs; I'll be interested to see what happens with this one. BTW, I'm having a bit of trouble with the %{buildnonet} mechanism used by the newer spec file. Is there a script hiding somewhere for populating /var/cache/yum appropriately? As it is, mine doesn't have all the packages called for, and febootstrap is failing as a result. From rjones at redhat.com Wed May 27 20:03:23 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 May 2009 21:03:23 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <4A1D99E5.8060605@dyfis.net> References: <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> <20090527194553.GC5472@amd.home.annexia.org> <4A1D99E5.8060605@dyfis.net> Message-ID: <20090527200248.GD5472@amd.home.annexia.org> On Wed, May 27, 2009 at 02:52:05PM -0500, Charles Duffy wrote: > BTW, I'm having a bit of trouble with the %{buildnonet} mechanism used > by the newer spec file. Is there a script hiding somewhere for > populating /var/cache/yum appropriately? As it is, mine doesn't have all > the packages called for, and febootstrap is failing as a result. You've tried setting buildnonet to 0 (ie. the opposite of what's in the specfile)? It should then download the required packages from the EPEL repositories, although I've not tested that option in a while ... Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From charles at dyfis.net Wed May 27 20:21:40 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 15:21:40 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527200248.GD5472@amd.home.annexia.org> References: <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> <20090527194553.GC5472@amd.home.annexia.org> <4A1D99E5.8060605@dyfis.net> <20090527200248.GD5472@amd.home.annexia.org> Message-ID: <4A1DA0D4.2070607@dyfis.net> Richard W.M. Jones wrote: > You've tried setting buildnonet to 0 (ie. the opposite of what's in > the specfile)? It should then download the required packages from the > EPEL repositories, although I've not tested that option in a while ... > Ahh -- I was running rpmbuild --define 'buildnonet 0', but the spec file doesn't actually respect any preset value. Might I suggest "%global buildnonet 0%{!?buildwithnet:1}"? Setting buildnonet to 0 does refer to remote yum sources as expected -- but that doesn't go happily either, it smells to me like it's trying to refer to Fedora-devel packages, which the older EL yum can't handle: http://fedora.secsup.org/linux/development/x86_64/os/repodata/a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2: [Errno -3] Error performing checksum Trying other mirror. a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2 | 10 MB 00:02 http://mirrors.reflected.net/fedora/linux/development/x86_64/os/repodata/a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2: [Errno -3] Error performing checksum Trying other mirror. [...and so on, for other mirrors attempted] From cochranb at speakeasy.net Wed May 27 21:28:36 2009 From: cochranb at speakeasy.net (Robert L Cochran) Date: Wed, 27 May 2009 17:28:36 -0400 Subject: [fedora-virt] Hostdev XML Not Working (Libvirt) In-Reply-To: <4A1D4CC6.1010102@redhat.com> References: <4A1B4271.2020903@speakeasy.net> <4A1D4CC6.1010102@redhat.com> Message-ID: <4A1DB084.8020702@speakeasy.net> On 05/27/2009 10:23 AM, Cole Robinson wrote: > Robert L Cochran wrote: > >> Why doesn't the hostdev xml shown below for a USB device work? I've >> shown the entire block for context. >> >> >> /usr/bin/qemu-kvm >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> The USB device of interest to me is an FTDI FT232RL chip. When I plug >> the board into a USB port on the host, I want it to be passed through to >> my Ubuntu guest so that it can be seen as /dev/ttyUSB0. The guest >> doesn't see it. I took the product and vendor id numbers from the >> /var/log/messages output, and these might be in hexadecimal (and I >> should be writing e.g.. I might also be getting the >> physical handling of the device wrong: maybe it needs to be already >> plugged in at the time the Ubuntu guest is started up, or maybe I should >> start the guest first and then plug in the device? >> >> Advice appreciated. >> >> Bob Cochran >> >> > > For future reference, easiest way to get device product + vendor info is > to use 'lsusb'. > > I think you need to use the hex format you mention above to get this to > work correctly ( ...). Also, make sure that the > device is still listed in the domain xml when you use 'virsh > dumpxml': this will guarantee that libvirt is at least > recognizing the new xml. > > You can also check /var/log/libvirt/qemu/.log to see what QEMU > command line libvirt is generating, which helps debugging these issues. > > - Cole > > > Cole, Thanks for your response. I modified the file /etc/libvirt/qemu/Ubuntu9.xml so that it looks like this: Ubuntu9 cb8c3d15-fe92-1bf2-5676-03370743e815 1572864 1572864 1 hvm destroy restart restart /usr/bin/qemu-kvm My interest is getting the Ubuntu machine to recognize the FTDI device indicated in the entry you see above. I did not start the virtual machine or actually plug the USB device of interest into an available port at this point. Instead I ran 'virsh' as you suggested above. Here is the output: [root at deafeng3 qemu]# virsh dumpxml Ubuntu9 Ubuntu9 cb8c3d15-fe92-1bf2-5676-03370743e815 1572864 1572864 1 hvm destroy restart restart /usr/bin/qemu-kvm At this point, I plugged in the USB device I am interested in and again ran 'virsh', repeating the command above. Here is the output: [root at deafeng3 qemu]# virsh dumpxml Ubuntu9 Ubuntu9 cb8c3d15-fe92-1bf2-5676-03370743e815 1572864 1572864 1 hvm destroy restart restart /usr/bin/qemu-kvm I'm not sure what device 0103:1771 is. Here is the output of lsusb. Notice the device on bus 007 device 003. This is what I want the Ubuntu machine to see and manage. [root at deafeng3 qemu]# lsusb Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 007 Device 002: ID 0403:6001 Future Technology Devices International, Ltd FT232 USB-Serial (UART) IC Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 001 Device 006: ID 0c45:63f8 Microdia Bus 001 Device 011: ID 045e:0084 Microsoft Corp. Basic Optical Mouse Bus 001 Device 010: ID 046d:c315 Logitech, Inc. Classic New Touch Keyboard Bus 001 Device 009: ID 051d:0002 American Power Conversion Uninterruptible Power Supply Bus 001 Device 012: ID 413c:8153 Dell Computer Corp. Bus 001 Device 013: ID 413c:8154 Dell Computer Corp. Bus 001 Device 003: ID 413c:2513 Dell Computer Corp. Bus 001 Device 008: ID 0a5c:4500 Broadcom Corp. Bus 001 Device 007: ID 413c:8149 Dell Computer Corp. Bus 001 Device 004: ID 413c:2513 Dell Computer Corp. Bus 001 Device 002: ID 0424:2512 Standard Microsystems Corp. Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 005 Device 002: ID 0a5c:5800 Broadcom Corp. Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 008 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Lastly, you wondered what command line qemu is using. I checked /var/log/libvirt/qemu and at the very start of the file I see this: LC_ALL=C PATH=/sbin:/usr/sbin:/bin:/usr/bin /usr/bin/qemu-kvm -S -M pc -m 1536 -smp 1 -name Ubuntu9 -uuid cb8c3d15-fe92-1bf2-5676-03370743e815 -monitor pty -pidfile /var/run/libvirt/qemu//Ubuntu9.pid -no-reboot -boot d -drive file=/home/rlc/Download/ubuntu-9.04-desktop-i386.iso,if=ide,media=cdrom,index=2 -drive file=/var/lib/libvirt/images/Ubuntu9.img,if=virtio,index=0 -net nic,macaddr=54:52:00:74:32:2f,vlan=0,model=virtio -net tap,fd=18,script=,vlan=0,ifname=vnet0 -serial pty -parallel none -usb -vnc 127.0.0.1:0 char device redirected to /dev/pts/1 char device redirected to /dev/pts/2 info cpus * CPU #0: pc=0x00000000000ffff0 thread_id=3476 cont balloon 1536 info balloon balloon: actual=1536 info balloon balloon: actual=1536 info balloon balloon: actual=1536 info balloon balloon: actual=1536 info balloon balloon: actual=1536 [above two lines repeat for a few thousand lines] Advice appreciated. How can I get the USB device to work? By the way, I thought I have an installed virtual machine on my hard drive, so I'm a little puzzled that the reference -drive file=/home/rlc/Download/ubuntu-9.04-desktop-i386.iso still exists in the qemu-kvm command line. Am I still starting up from the iso image? I'm confused. Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From charles at dyfis.net Wed May 27 21:03:07 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 16:03:07 -0500 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <20090527200248.GD5472@amd.home.annexia.org> References: <20090527073346.GA2492@amd.home.annexia.org> <20090527073738.GB2492@amd.home.annexia.org> <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> <20090527194553.GC5472@amd.home.annexia.org> <4A1D99E5.8060605@dyfis.net> <20090527200248.GD5472@amd.home.annexia.org> Message-ID: <4A1DAA8B.9050107@dyfis.net> Updating the %else clause to contain %define extra --with-repo=centos-5 --with-mirror=http://mirror.centos.org/centos-5/5.3/os/%{_arch}/ --with-updates=none ...works around the failure with buildnonet==0. From charles at dyfis.net Wed May 27 21:23:35 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 16:23:35 -0500 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <20090508073741.GB12649@amd.home.annexia.org> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> <4A0383D0.5060800@dyfis.net> <4A038862.6000805@dyfis.net> <20090508073741.GB12649@amd.home.annexia.org> Message-ID: <4A1DAF57.2090302@dyfis.net> Richard W.M. Jones wrote: > On Thu, May 07, 2009 at 08:18:26PM -0500, Charles Duffy wrote: > >> Charles Duffy wrote: >> >>> After this point, the only remaining issue is that the ocaml bindings >>> aren't being built, resulting in a packaging failure: >>> >>> RPM build errors: >>> File not found: >>> /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs >>> File not found by glob: >>> /var/tmp/libguestfs-1.0.19-1-root/usr/lib64/ocaml/guestfs/*.a >>> [...] >>> >> As it turns out, the issue was not that the ocaml bindings aren't being >> built, but that they were being installed under /usr/lib/ocaml as >> opposed to /usr/lib64/ocaml. >> > > Strange one - this works for me on Fedora x86_64 .. > > I'll look into this a bit more when I come to add libguestfs to EPEL. You were indicating that you were having trouble reproducing this issue? It's easy enough to work around (s@%{_libdir}@/usr/lib*@), but I'm still seeing this: error: Installed (but unpackaged) file(s) found: /usr/lib/ocaml/guestfs/META /usr/lib/ocaml/guestfs/guestfs.cmi /usr/lib/ocaml/guestfs/guestfs.cmx /usr/lib/ocaml/guestfs/guestfs.mli /usr/lib/ocaml/guestfs/libmlguestfs.a /usr/lib/ocaml/guestfs/mlguestfs.a /usr/lib/ocaml/guestfs/mlguestfs.cma /usr/lib/ocaml/guestfs/mlguestfs.cmxa /usr/lib/ocaml/stublibs/dllmlguestfs.so /usr/lib/ocaml/stublibs/dllmlguestfs.so.owner RPM build errors: File not found: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.a File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.cmxa File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.cmx File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.mli File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/stublibs/*.so File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/stublibs/*.so.owner File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.a File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.cmxa File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.cmx File not found by glob: /var/tmp/libguestfs-1.0.34-1.3-root/usr/lib64/ocaml/guestfs/*.mli Installed (but unpackaged) file(s) found: /usr/lib/ocaml/guestfs/META /usr/lib/ocaml/guestfs/guestfs.cmi /usr/lib/ocaml/guestfs/guestfs.cmx /usr/lib/ocaml/guestfs/guestfs.mli /usr/lib/ocaml/guestfs/libmlguestfs.a /usr/lib/ocaml/guestfs/mlguestfs.a /usr/lib/ocaml/guestfs/mlguestfs.cma /usr/lib/ocaml/guestfs/mlguestfs.cmxa /usr/lib/ocaml/stublibs/dllmlguestfs.so /usr/lib/ocaml/stublibs/dllmlguestfs.so.owner From charles at dyfis.net Wed May 27 22:09:20 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 17:09:20 -0500 Subject: [fedora-virt] virt-inspector calling guestfs_umount_all() before launch()? Message-ID: <4A1DBA10.90601@dyfis.net> This is virt-inspector-1.0.34-1.3 on CentOS 5: $ virt-inspector --ro-fish --xml da.qcow2 qemu: loading initrd (0x33a576d bytes) at 0x0000000014c4a000 umount_all: guestfs_umount_all: call launch() before using this function at /usr/bin/virt-inspector line 375. From charles at dyfis.net Wed May 27 22:16:40 2009 From: charles at dyfis.net (Charles Duffy) Date: Wed, 27 May 2009 17:16:40 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1DBA10.90601@dyfis.net> References: <4A1DBA10.90601@dyfis.net> Message-ID: <4A1DBBC8.5060805@dyfis.net> Charles Duffy wrote: > This is virt-inspector-1.0.34-1.3 on CentOS 5: > > $ virt-inspector --ro-fish --xml da.qcow2 > qemu: loading initrd (0x33a576d bytes) at 0x0000000014c4a000 > umount_all: guestfs_umount_all: call launch() before using this > function at /usr/bin/virt-inspector line 375. Running with LIBGUESTFS_DEBUG=1, it appears that this happened due to the guest trying to do ext3 journal recovery against a read-only image. If telling libguestfs to run in read-only mode implied passing the -snapshot flag to qemu, and passing the snapshot=on option to each -drive, this should allow changes to be *temporarily* created (stored to unlinked copy-on-write files for the duration of the run), mooting issues of this kind. From rjones at redhat.com Thu May 28 08:07:49 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 09:07:49 +0100 Subject: [fedora-virt] libguestfs - no ext3 support building on CentOS 5 In-Reply-To: <4A1DA0D4.2070607@dyfis.net> References: <20090527142704.GB4642@amd.home.annexia.org> <20090527150512.GA5472@amd.home.annexia.org> <20090527172933.GA6243@amd.home.annexia.org> <4A1D95C3.5030602@dyfis.net> <20090527194553.GC5472@amd.home.annexia.org> <4A1D99E5.8060605@dyfis.net> <20090527200248.GD5472@amd.home.annexia.org> <4A1DA0D4.2070607@dyfis.net> Message-ID: <20090528080749.GH1267@amd.home.annexia.org> On Wed, May 27, 2009 at 03:21:40PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> You've tried setting buildnonet to 0 (ie. the opposite of what's in >> the specfile)? It should then download the required packages from the >> EPEL repositories, although I've not tested that option in a while ... >> > Ahh -- I was running rpmbuild --define 'buildnonet 0', but the spec file > doesn't actually respect any preset value. Might I suggest "%global > buildnonet 0%{!?buildwithnet:1}"? > > Setting buildnonet to 0 does refer to remote yum sources as expected -- > but that doesn't go happily either, it smells to me like it's trying to > refer to Fedora-devel packages, which the older EL yum can't handle: > > http://fedora.secsup.org/linux/development/x86_64/os/repodata/a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2: > [Errno -3] Error performing checksum > Trying other mirror. > a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2 > > | 10 MB 00:02 > http://mirrors.reflected.net/fedora/linux/development/x86_64/os/repodata/a639a7f3e2a4b53872dc403af7c8991e0bd0f441eea36d57cc321c0a69f4c415-primary.sqlite.bz2: > [Errno -3] Error performing checksum > Trying other mirror. > [...and so on, for other mirrors attempted] Yes, that's because that path isn't tested. The actual configure command which works is: ./configure --with-repo=centos-5 \ --with-mirror=http://mirror.centos.org/centos-5/5.3/os/i386/ \ --with-updates=none I updated the spec file to reflect this. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html From rjones at redhat.com Thu May 28 08:10:49 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 09:10:49 +0100 Subject: [fedora-virt] Re: libguestfs RPMs for RHEL 5.3 In-Reply-To: <4A1DAF57.2090302@dyfis.net> References: <49FE3151.5040600@dyfis.net> <20090507194006.GA6062@amd.home.annexia.org> <4A0383D0.5060800@dyfis.net> <4A038862.6000805@dyfis.net> <20090508073741.GB12649@amd.home.annexia.org> <4A1DAF57.2090302@dyfis.net> Message-ID: <20090528081049.GI1267@amd.home.annexia.org> On Wed, May 27, 2009 at 04:23:35PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> On Thu, May 07, 2009 at 08:18:26PM -0500, Charles Duffy wrote: >>> Charles Duffy wrote: >>> As it turns out, the issue was not that the ocaml bindings aren't >>> being built, but that they were being installed under /usr/lib/ocaml >>> as opposed to /usr/lib64/ocaml. >>> >> >> Strange one - this works for me on Fedora x86_64 .. >> >> I'll look into this a bit more when I come to add libguestfs to EPEL. > You were indicating that you were having trouble reproducing this issue? > It's easy enough to work around (s@%{_libdir}@/usr/lib*@), but I'm still > seeing this: > > error: Installed (but unpackaged) file(s) found: > /usr/lib/ocaml/guestfs/META > /usr/lib/ocaml/guestfs/guestfs.cmi > /usr/lib/ocaml/guestfs/guestfs.cmx > /usr/lib/ocaml/guestfs/guestfs.mli > /usr/lib/ocaml/guestfs/libmlguestfs.a > /usr/lib/ocaml/guestfs/mlguestfs.a > /usr/lib/ocaml/guestfs/mlguestfs.cma > /usr/lib/ocaml/guestfs/mlguestfs.cmxa > /usr/lib/ocaml/stublibs/dllmlguestfs.so > /usr/lib/ocaml/stublibs/dllmlguestfs.so.owner That's the correct location on 32 bit archs. What's the ocamlfind install command that it actually runs? Are you passing --libdir=%{_libdir} to the configure? Note that I've built this successfully on the EL-5 builders on both i386 and x86_64 so it should just work. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html From rjones at redhat.com Thu May 28 08:39:30 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 09:39:30 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1DBBC8.5060805@dyfis.net> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> Message-ID: <20090528083930.GA13154@amd.home.annexia.org> On Wed, May 27, 2009 at 05:16:40PM -0500, Charles Duffy wrote: > Charles Duffy wrote: >> This is virt-inspector-1.0.34-1.3 on CentOS 5: >> >> $ virt-inspector --ro-fish --xml da.qcow2 >> qemu: loading initrd (0x33a576d bytes) at 0x0000000014c4a000 >> umount_all: guestfs_umount_all: call launch() before using this >> function at /usr/bin/virt-inspector line 375. > Running with LIBGUESTFS_DEBUG=1, it appears that this happened due to > the guest trying to do ext3 journal recovery against a read-only image. ext3 tries to do a journal recovery if we mount -o ro ? > If telling libguestfs to run in read-only mode implied passing the > -snapshot flag to qemu, and passing the snapshot=on option to each > -drive, this should allow changes to be *temporarily* created (stored to > unlinked copy-on-write files for the duration of the run), mooting > issues of this kind. I want to get the ability to mount drives completely read-only into qemu. Turns out that it's a bit more complicated than just opening the underlying file O_RDONLY - we have to communicate this all the way through IDE / virtio drivers. But using snapshots actually isn't such a bad idea. Also works with previous and current versions of qemu too. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ From rjones at redhat.com Thu May 28 08:57:33 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 09:57:33 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528083930.GA13154@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> Message-ID: <20090528085733.GA13234@amd.home.annexia.org> Charles, can you see if this fixes the problem? It's completely untested, and might even segfault the Perl bindings because of BZ #501892 (which really needs to be fixed). Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ -------------- next part -------------- diff --git a/inspector/virt-inspector.pl b/inspector/virt-inspector.pl index 4ee0e08..1a1f94f 100755 --- a/inspector/virt-inspector.pl +++ b/inspector/virt-inspector.pl @@ -213,6 +213,7 @@ if (-e $ARGV[0]) { # We've now got the list of @images, so feed them to libguestfs. my $g = Sys::Guestfs->new (); +$g->config ("-snapshot", undef); $g->add_drive ($_) foreach @images; $g->launch (); $g->wait_ready (); From charles at dyfis.net Thu May 28 14:40:24 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 09:40:24 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528083930.GA13154@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> Message-ID: On Thu, May 28, 2009 at 3:39 AM, Richard W.M. Jones wrote: > > Running with LIBGUESTFS_DEBUG=1, it appears that this happened due to > > the guest trying to do ext3 journal recovery against a read-only image. > > ext3 tries to do a journal recovery if we mount -o ro ? Yup. Quoting from Stephen Tweedie's ext3 readme: If you use an older version of e2fsck from e2fsprogs-1.17 or later, then you can now run e2fsck quite happily on the filesystem, but *only as long as the filesystem was unmounted cleanly*. If it wasn't, then you'll need to get the kernel code to recover the journal from the disk by mounting the filesystem *(even a readonly mount will cause a journal recovery to happen)* and umounting it again (or, for the root filesystem, remounting it readonly with "mount -o remount,ro /"). But using snapshots actually isn't such a bad idea. Also works with > previous and current versions of qemu too. > It'd almost be tempting to provide optional support for snapshots in read-write mode, letting the user do a commit to persist their changes or to exit without one to discard them. -------------- next part -------------- An HTML attachment was scrubbed... URL: From charles at dyfis.net Thu May 28 16:50:00 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 11:50:00 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528085733.GA13234@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> Message-ID: <4A1EC0B8.8020908@dyfis.net> Richard W.M. Jones wrote: > Charles, can you see if this fixes the problem? > > It's completely untested, and might even segfault the Perl bindings > because of BZ #501892 (which really needs to be fixed). > It doesn't fix things completely, due to what appears to be a qemu bug -- apparently, -snapshot doesn't work correctly with -drive, and snapshot=on needs to be used instead. I've tested this by setting LIBGUESTFS_QEMU to a wrapper which adds ,cache=off,snapshot=on to each -drive parameter, which works correctly; using only -snapshot fails as below. The generated command line correctly includes -snapshot, and qemu correctly creates a read-only file to store changes to, but we still fail (albeit differently) on attempted journal replay: hda: max request size: 128KiB hda: 2016 sectors (1 MB) w/256KiB Cache, CHS=2/16/63, (U)DMA hda: cache flushes supported hda:<4>hda: dma_timer_expiry: dma status == 0x61 hda: DMA timeout error sock_read_event: 0x9ec8f0 g->state = 1, fd = 3, events = 0x1 stdout_event: 0x9ec8f0: child process died wait_ready: guestfs_wait_ready failed, see earlier error messages at /usr/bin/virt-inspector line 219. closing guestfs handle 0x9ec8f0 (state 0) Manually creating a new, writable backing store (qemu-img create -b da.qcow2 -f qcow2 da.tmp.qcow2) and running against that (with --force), virt-inspector works correctly, just as it does with snapshot=on passed to the individual -drive options and no global -snapshot flag at all. One other issue with that patch is that it passes an empty string rather than no secondary parameter at all in place of the null: /usr/bin/qemu-kvm -snapshot '' -drive file=da.qcow2 [...] From rjones at redhat.com Thu May 28 16:58:30 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 17:58:30 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1EC0B8.8020908@dyfis.net> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EC0B8.8020908@dyfis.net> Message-ID: <20090528165830.GC14389@amd.home.annexia.org> On Thu, May 28, 2009 at 11:50:00AM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> Charles, can you see if this fixes the problem? >> >> It's completely untested, and might even segfault the Perl bindings >> because of BZ #501892 (which really needs to be fixed). I'm still fixing this one ... > It doesn't fix things completely, due to what appears to be a qemu bug > -- apparently, -snapshot doesn't work correctly with -drive, and > snapshot=on needs to be used instead. I've tested this by setting > LIBGUESTFS_QEMU to a wrapper which adds ,cache=off,snapshot=on to each > -drive parameter, which works correctly; using only -snapshot fails as > below. > > The generated command line correctly includes -snapshot, and qemu > correctly creates a read-only file to store changes to, but we still > fail (albeit differently) on attempted journal replay: > > hda: max request size: 128KiB > hda: 2016 sectors (1 MB) w/256KiB Cache, CHS=2/16/63, (U)DMA > hda: cache flushes supported > hda:<4>hda: dma_timer_expiry: dma status == 0x61 > hda: DMA timeout error > sock_read_event: 0x9ec8f0 g->state = 1, fd = 3, events = 0x1 > stdout_event: 0x9ec8f0: child process died > wait_ready: guestfs_wait_ready failed, see earlier error messages at > /usr/bin/virt-inspector line 219. > closing guestfs handle 0x9ec8f0 (state 0) > > Manually creating a new, writable backing store (qemu-img create -b > da.qcow2 -f qcow2 da.tmp.qcow2) and running against that (with --force), > virt-inspector works correctly, just as it does with snapshot=on passed > to the individual -drive options and no global -snapshot flag at all. > > One other issue with that patch is that it passes an empty string rather > than no secondary parameter at all in place of the null: > > /usr/bin/qemu-kvm -snapshot '' -drive file=da.qcow2 [...] I think that's the same bug in the Perl bindings actually. The C code looks correct. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From rjones at redhat.com Thu May 28 19:22:04 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 20:22:04 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528165830.GC14389@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EC0B8.8020908@dyfis.net> <20090528165830.GC14389@amd.home.annexia.org> Message-ID: <20090528192204.GA16476@amd.home.annexia.org> On Thu, May 28, 2009 at 05:58:30PM +0100, Richard W.M. Jones wrote: > On Thu, May 28, 2009 at 11:50:00AM -0500, Charles Duffy wrote: > > Richard W.M. Jones wrote: > >> It's completely untested, and might even segfault the Perl bindings > >> because of BZ #501892 (which really needs to be fixed). > > I'm still fixing this one ... This ridiculously large patch fixes 501892 and a number of other bugs in the language bindings, and also adds detailed regression tests so hopefully we won't end up adding any bugs like this back in future: http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=babc0846cc911b01a58a7385d30ad25889b7175a > > One other issue with that patch is that it passes an empty string rather > > than no secondary parameter at all in place of the null: > > > > /usr/bin/qemu-kvm -snapshot '' -drive file=da.qcow2 [...] > > I think that's the same bug in the Perl bindings actually. The > C code looks correct. The Perl binding was turning undef into '', incorrectly of course. This is fixed by the above patch. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my programming blog: http://rwmj.wordpress.com Fedora now supports 75 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From charles at dyfis.net Thu May 28 21:03:57 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 16:03:57 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528085733.GA13234@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> Message-ID: <4A1EFC3D.700@dyfis.net> Richard W.M. Jones wrote: > Charles, can you see if this fixes the problem? > > It's completely untested, and might even segfault the Perl bindings > because of BZ #501892 (which really needs to be fixed). > With 1.0.35, this works correctly. From rjones at redhat.com Thu May 28 21:13:53 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 22:13:53 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1EFC3D.700@dyfis.net> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EFC3D.700@dyfis.net> Message-ID: <20090528211353.GA16996@amd.home.annexia.org> On Thu, May 28, 2009 at 04:03:57PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> Charles, can you see if this fixes the problem? >> >> It's completely untested, and might even segfault the Perl bindings >> because of BZ #501892 (which really needs to be fixed). >> > > With 1.0.35, this works correctly. 1.0.35 had a broken test. 1.0.36 is on its way ... I've also built 1.0.35 in EPEL now, so it should be possible to use that directly once it is signed and in the EPEL repositories. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html From charles at dyfis.net Thu May 28 21:19:21 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 16:19:21 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528211353.GA16996@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EFC3D.700@dyfis.net> <20090528211353.GA16996@amd.home.annexia.org> Message-ID: <4A1EFFD9.1090309@dyfis.net> Richard W.M. Jones wrote: > On Thu, May 28, 2009 at 04:03:57PM -0500, Charles Duffy wrote: > >> Richard W.M. Jones wrote: >> >>> Charles, can you see if this fixes the problem? >>> >>> It's completely untested, and might even segfault the Perl bindings >>> because of BZ #501892 (which really needs to be fixed). >>> >>> >> With 1.0.35, this works correctly. >> > > 1.0.35 had a broken test. 1.0.36 is on its way ... > > I've also built 1.0.35 in EPEL now, so it should be possible to use > that directly once it is signed and in the EPEL repositories. > Just curious -- do you intend to push the patch ("-snapshot" in virt-inspector) into the tree? From rjones at redhat.com Thu May 28 21:53:20 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 22:53:20 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1EFFD9.1090309@dyfis.net> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EFC3D.700@dyfis.net> <20090528211353.GA16996@amd.home.annexia.org> <4A1EFFD9.1090309@dyfis.net> Message-ID: <20090528215320.GB16996@amd.home.annexia.org> On Thu, May 28, 2009 at 04:19:21PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> On Thu, May 28, 2009 at 04:03:57PM -0500, Charles Duffy wrote: >> >>> Richard W.M. Jones wrote: >>> >>>> Charles, can you see if this fixes the problem? >>>> >>>> It's completely untested, and might even segfault the Perl bindings >>>> because of BZ #501892 (which really needs to be fixed). >>>> >>> With 1.0.35, this works correctly. >>> >> >> 1.0.35 had a broken test. 1.0.36 is on its way ... >> >> I've also built 1.0.35 in EPEL now, so it should be possible to use >> that directly once it is signed and in the EPEL repositories. >> > Just curious -- do you intend to push the patch ("-snapshot" in > virt-inspector) into the tree? I guess we can, but is there a better way? Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From charles at dyfis.net Thu May 28 22:01:02 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 17:01:02 -0500 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <20090528215320.GB16996@amd.home.annexia.org> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EFC3D.700@dyfis.net> <20090528211353.GA16996@amd.home.annexia.org> <4A1EFFD9.1090309@dyfis.net> <20090528215320.GB16996@amd.home.annexia.org> Message-ID: <4A1F099E.7050004@dyfis.net> Richard W.M. Jones wrote: > I guess we can, but is there a better way? > The guests could use "blockdev --setro" prior to any read-only mount calls to prevent the ext3 code from being able to replay its journal. If the journal replay is necessary for the filesystem's metadata to be consistent, though, I'm not sure that this is really "better". From rjones at redhat.com Thu May 28 22:07:25 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 28 May 2009 23:07:25 +0100 Subject: [fedora-virt] Re: virt-inspector calling guestfs_umount_all() before launch()? In-Reply-To: <4A1F099E.7050004@dyfis.net> References: <4A1DBA10.90601@dyfis.net> <4A1DBBC8.5060805@dyfis.net> <20090528083930.GA13154@amd.home.annexia.org> <20090528085733.GA13234@amd.home.annexia.org> <4A1EFC3D.700@dyfis.net> <20090528211353.GA16996@amd.home.annexia.org> <4A1EFFD9.1090309@dyfis.net> <20090528215320.GB16996@amd.home.annexia.org> <4A1F099E.7050004@dyfis.net> Message-ID: <20090528220725.GA17240@amd.home.annexia.org> On Thu, May 28, 2009 at 05:01:02PM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> I guess we can, but is there a better way? >> > > The guests could use "blockdev --setro" prior to any read-only mount > calls to prevent the ext3 code from being able to replay its journal. Ah well ... I discovered that 'blockdev --setro' doesn't work on Fedora 10+. It only works on RHEL 5. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From charles at dyfis.net Thu May 28 22:41:44 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 17:41:44 -0500 Subject: [fedora-virt] Running shell script fragments via guestfish "command" Message-ID: <4A1F1328.2010609@dyfis.net> What's the Right Way to do the below? > command "bash -c 'for D in avahi-daemon avahi-dnsconfd haldaemon; do /sbin/chkconfig $D off; done; true'" libguestfs: error: D: -c: line 0: unexpected EOF while looking for matching `'' D: -c: line 1: syntax error: unexpected end of file From charles at dyfis.net Thu May 28 23:04:16 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 18:04:16 -0500 Subject: [fedora-virt] libguestfs: mkdir-p failing on "directory already exists" Message-ID: <4A1F1870.9020208@dyfis.net> To properly behave as "mkdir -p", mkdir-p should consider the case where the target directory already exists successful. libguestfs: error: mkdir -p: /root/.ssh: File exists From charles at dyfis.net Fri May 29 00:19:06 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 19:19:06 -0500 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session Message-ID: <4A1F29FA.1020100@dyfis.net> I have a rather large set of RPMs and such on my host I want to install on my guest using libguestfs. The normal way to do this would be to upload the package(s), install them, then (optionally) remove the RPMs. However, I don't particularly want to bloat the qcow2 file with the changes made via uploading files which are only going to be deleted when no longer in use. Someday (*sigh*) we'll have 9p-over-virtio support built into qemu; until then, a few ways to get around this present themselves: - Add a temporary file as a disk to the guestfs session on startup; within the guest, create a new filesystem there, transfer temporary files to it, and delete when done. Pro: No external dependencies. Con: Not sure of a safe and correct way to find the name of the temporary disk [to ensure that the mkfs operation being scripted refers only to it], particularly when trying to script everything into a single guestfish invocation with a variable number of other disks exposed (ie. using a command line generated by virt-inspect). Upload needs to be file-by-file (or have everything bundled in a tarball beforehand). - Run a tmpfs mount on the guest; upload RPMs there. Pro: Fast. No need to worry about device names. Con: Potential for running out of memory. Upload needs to be file-by-file (or have everything bundled in a tarball beforehand). - Stuff the files to be added into a cramfs filesystem, running mkcramfs on the host. Pro: Relatively fast. Easy to transfer a full directory structure to the guest at one stroke. Con: Device names an issue. Requires cramfs.ko to be added to the whitelist in appliance/update-initramfs.sh.in. - Support a FUSE-based mechanism for exporting files, ie. ccgfs-over-vmchannel Pro: Potential to be the cleanest approach from an end-user perspective. Con: Implementation difficulty; no Fedora or EPEL packages are available for ccgfs. From redhat-lists at anselcomputers.com Fri May 29 04:38:46 2009 From: redhat-lists at anselcomputers.com (Michael Ansel) Date: Fri, 29 May 2009 00:38:46 -0400 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <4A1F29FA.1020100@dyfis.net> References: <4A1F29FA.1020100@dyfis.net> Message-ID: <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> On Thu, May 28, 2009 at 8:19 PM, Charles Duffy wrote: > I have a rather large set of RPMs and such on my host I want to install on > my guest using libguestfs. The normal way to do this would be to upload the > package(s), install them, then (optionally) remove the RPMs. > > However, I don't particularly want to bloat the qcow2 file with the changes > made via uploading files which are only going to be deleted when no longer > in use. Someday (*sigh*) we'll have 9p-over-virtio support built into qemu; > until then, a few ways to get around this present themselves: > What about setting up a temporary NFS export on the host and mounting that on the guest? I've done installs using yum/rpm directly from the NFS mount, eliminating the need to ever copy the RPMs into anything more than working memory. If network security is an issue, I'm sure you could use a combination of iptables rules and NFS export options to limit access to that single box as well (though, I haven't done that one). Here's to hoping I didn't miss something blatantly obvious... Michael From charles at dyfis.net Fri May 29 04:51:03 2009 From: charles at dyfis.net (Charles Duffy) Date: Thu, 28 May 2009 23:51:03 -0500 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> Message-ID: <4A1F69B7.6040900@dyfis.net> Michael Ansel wrote: > On Thu, May 28, 2009 at 8:19 PM, Charles Duffy wrote: > >> I have a rather large set of RPMs and such on my host I want to install on >> my guest using libguestfs. The normal way to do this would be to upload the >> package(s), install them, then (optionally) remove the RPMs. >> >> However, I don't particularly want to bloat the qcow2 file with the changes >> made via uploading files which are only going to be deleted when no longer >> in use. Someday (*sigh*) we'll have 9p-over-virtio support built into qemu; >> until then, a few ways to get around this present themselves: >> > > What about setting up a temporary NFS export on the host and mounting > that on the guest? I've done installs using yum/rpm directly from the > NFS mount, eliminating the need to ever copy the RPMs into anything > more than working memory. If network security is an issue, I'm sure > you could use a combination of iptables rules and NFS export options > to limit access to that single box as well (though, I haven't done > that one). As libguestfs (and the software I'm building with it) runs unprivileged and unattended, an added requirement of NFS server support on the host would mean a pretty massive increase in security exposure and administrative load. (Also, the libguestfs appliance doesn't included nfs-utils at present, and I don't know that NFS works over qemu's SLiRP-based "-net user" mode anyhow). If I were going to go that route, I'd probably lean towards using qemu's existing SMB server support -- but there would still be changes to the guest needed, making it less than pain-free. [For the moment, I ended up using squashfs; cramfs, mentioned in my previous email, was a Really Bad Idea on account of the size limitations] From rjones at redhat.com Fri May 29 06:21:19 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 07:21:19 +0100 Subject: [fedora-virt] Running shell script fragments via guestfish "command" In-Reply-To: <4A1F1328.2010609@dyfis.net> References: <4A1F1328.2010609@dyfis.net> Message-ID: <20090529062119.GA19219@amd.home.annexia.org> On Thu, May 28, 2009 at 05:41:44PM -0500, Charles Duffy wrote: > What's the Right Way to do the below? > > > command "bash -c 'for D in avahi-daemon avahi-dnsconfd haldaemon; > do /sbin/chkconfig $D off; done; true'" > libguestfs: error: D: -c: line 0: unexpected EOF while looking for > matching `'' > D: -c: line 1: syntax error: unexpected end of file The problem is that guestfish's list splitting code sucks, so the above turns into the equivalent of this API call: guestfs_command (g, "bash", "-c", "'for", "D", "in", "avahi-daemon", "avahi-dnsconfd", "haldaemon;", "do", "/sbin/chkconfig", "$D", "off;", "done;", "true'"); Please file a bug about this. In the meantime you can write what you really wanted to write using a language binding such as Perl. Or if you need to use guestfish then write a small shell script to do the operations you need and use the 'upload' and 'chmod' commands to upload it, chmod it and run it inside the guest. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From rjones at redhat.com Fri May 29 06:21:44 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 07:21:44 +0100 Subject: [fedora-virt] libguestfs: mkdir-p failing on "directory already exists" In-Reply-To: <4A1F1870.9020208@dyfis.net> References: <4A1F1870.9020208@dyfis.net> Message-ID: <20090529062144.GB19219@amd.home.annexia.org> On Thu, May 28, 2009 at 06:04:16PM -0500, Charles Duffy wrote: > To properly behave as "mkdir -p", mkdir-p should consider the case where > the target directory already exists successful. > > libguestfs: error: mkdir -p: /root/.ssh: File exists Yes, please file a bug report about that so we can track it. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Fri May 29 06:26:15 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 07:26:15 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <4A1F29FA.1020100@dyfis.net> References: <4A1F29FA.1020100@dyfis.net> Message-ID: <20090529062615.GC19219@amd.home.annexia.org> On Thu, May 28, 2009 at 07:19:06PM -0500, Charles Duffy wrote: > - Add a temporary file as a disk to the guestfs session on startup; > within the guest, create a new filesystem there, transfer temporary > files to it, and delete when done. > > Pro: No external dependencies. > Con: Not sure of a safe and correct way to find the name of the > temporary disk [to ensure that the mkfs operation being scripted refers > only to it], particularly when trying to script everything into a single > guestfish invocation with a variable number of other disks exposed (ie. > using a command line generated by virt-inspect). Upload needs to be > file-by-file (or have everything bundled in a tarball beforehand). Or you could prepare the disk image in advance using a separate guestfish operation: http://et.redhat.com/~rjones/libguestfs/recipes.html#tar2vm When you attach a new disk it has a predictable name, ie. /dev/hda, /dev/hdb and so on. If you know that your guest has only one real disk image, then your prepared image will always be /dev/hdb. This is what our V2V scripts do in order to upload new RPMs into the host, although at the moment it's not very well tested. > - Run a tmpfs mount on the guest; upload RPMs there. > > Pro: Fast. No need to worry about device names. > Con: Potential for running out of memory. Upload needs to be > file-by-file (or have everything bundled in a tarball beforehand). The amount of memory available is quite limited. > - Stuff the files to be added into a cramfs filesystem, running mkcramfs > on the host. > > Pro: Relatively fast. Easy to transfer a full directory structure to the > guest at one stroke. > Con: Device names an issue. Requires cramfs.ko to be added to the > whitelist in appliance/update-initramfs.sh.in. We should add this anyway -- please file a bug in bugzilla. > - Support a FUSE-based mechanism for exporting files, ie. > ccgfs-over-vmchannel > > Pro: Potential to be the cleanest approach from an end-user perspective. > Con: Implementation difficulty; no Fedora or EPEL packages are available > for ccgfs. And the other option would be: Implement the FTP server! Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From rjones at redhat.com Fri May 29 06:27:35 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 07:27:35 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> Message-ID: <20090529062735.GD19219@amd.home.annexia.org> On Fri, May 29, 2009 at 12:38:46AM -0400, Michael Ansel wrote: > On Thu, May 28, 2009 at 8:19 PM, Charles Duffy wrote: > > I have a rather large set of RPMs and such on my host I want to install on > > my guest using libguestfs. The normal way to do this would be to upload the > > package(s), install them, then (optionally) remove the RPMs. > > > > However, I don't particularly want to bloat the qcow2 file with the changes > > made via uploading files which are only going to be deleted when no longer > > in use. Someday (*sigh*) we'll have 9p-over-virtio support built into qemu; > > until then, a few ways to get around this present themselves: > > > > > What about setting up a temporary NFS export on the host and mounting > that on the guest? I've done installs using yum/rpm directly from the > NFS mount, eliminating the need to ever copy the RPMs into anything > more than working memory. If network security is an issue, I'm sure > you could use a combination of iptables rules and NFS export options > to limit access to that single box as well (though, I haven't done > that one). The original libguestfs plan was to use NFS for all file transfers. The reason why we don't intend to do this anymore is twofold: NFS requires root on the host side, and NFS requires either matching UIDs or complex UID-UID mappings. If we were going to implement any protocol, it'd be FTP, because FTP clients are common, can be run non-root, and don't need complicated UID mappings. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From redhat-lists at anselcomputers.com Fri May 29 07:34:21 2009 From: redhat-lists at anselcomputers.com (Michael Ansel) Date: Fri, 29 May 2009 03:34:21 -0400 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529062735.GD19219@amd.home.annexia.org> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> Message-ID: <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> On Fri, May 29, 2009 at 2:27 AM, Richard W.M. Jones wrote: > The original libguestfs plan was to use NFS for all file transfers. > The reason why we don't intend to do this anymore is twofold: NFS > requires root on the host side, and NFS requires either matching UIDs > or complex UID-UID mappings. ?If we were going to implement any > protocol, it'd be FTP, because FTP clients are common, can be run > non-root, and don't need complicated UID mappings. Okay, I think I'm not completely understanding how libguestfs works, so that is probably why I'm getting confused. *adds that to to-do list* I was looking at it figuring: 1) root (UID=0, everywhere) is the only one installing packages, so the UID mapping lines up perfectly every time 2) you only need root access on the host once when you are building the VM; you don't need NFS after it is already built I have a feeling this will all become clear to me once I do some research *self-chastisement for speaking before reading*, but for now I need sleep, so that reading shall come tomorrow. Also, on a side note: Is there a reason all of my replies are automatically directed at individuals vs the list? (I keep having to retype fedora-virt at redhat.com) Or is this another one of those things where I didn't read something I was supposed to... Thanks! Michael From charles at dyfis.net Fri May 29 07:46:28 2009 From: charles at dyfis.net (Charles Duffy) Date: Fri, 29 May 2009 02:46:28 -0500 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> Message-ID: <4A1F92D4.4090302@dyfis.net> Michael Ansel wrote: > 1) root (UID=0, everywhere) is the only one installing packages, so > the UID mapping lines up perfectly every time libguestfs in general, and host filesystem access in particular, is good for more than just installing packages, though admittedly that's my immediate use case. > 2) you only need root access on the host once when you are building > the VM; you don't need NFS after it is already built > Even if there were no use case for access to files off the host other than early installation -- Why is it acceptable for libguestfs to require root access *ever*, when so many alternate transport mechanisms (9p, ccgfs, FTP, qemu's built in smb, etc) are available which wouldn't create this constraint? If I'm building a piece of software invoked by untrusted users in a lab environment, I don't want that software to need, or have any means to access, root. Ever. For anything, at all. By the way -- I'm tired too (normal wakeup time in ~3.25 hours), so please excuse my tone if it's a bit harsh. From rjones at redhat.com Fri May 29 09:44:34 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 10:44:34 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> Message-ID: <20090529094434.GA4185@thinkpad.fab.redhat.com> On Fri, May 29, 2009 at 03:34:21AM -0400, Michael Ansel wrote: > 1) root (UID=0, everywhere) is the only one installing packages, so > the UID mapping lines up perfectly every time No - the concept of UIDs in libguestfs is non-intuitive. Firstly the normal mode of operation is to run libguestfs as non-root. There are plenty of reasons why this is a good idea, I think Charles covered a few of them. So everything we do is oriented around making sure that we don't need to be root to do libguestfs operations. However, libguestfs runs qemu/kvm (also as non-root) but that boots a mini appliance. *Inside* the appliance the appliance kernel thinks everything is running as root. It's not really from the point of view of the host machine, but inside the appliance that's how it looks. If we were to put an NFS server inside the appliance and tunnel the connections out to the host, then the host would have to mount the drive as root (or we'd have to fake it somehow - non-trivial with NFS). If we put an FTP server inside the appliance, things are better. FTP clients are widely available, either standalone or as libraries for many programming languages. They just need to be able to make a TCP connection, so there is no requirement for special privileges. On the server side (inside the appliance) the FTP server is running as "root" so it can make arbitrary changes to the filesystem. > 2) you only need root access on the host once when you are building > the VM; you don't need NFS after it is already built With libguestfs currently, you never need root at all, not to build libguestfs, not to run it, not for any operations using it. Rich. From ask at develooper.com Fri May 29 10:13:10 2009 From: ask at develooper.com (=?ISO-8859-1?Q?Ask_Bj=F8rn_Hansen?=) Date: Fri, 29 May 2009 03:13:10 -0700 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529094434.GA4185@thinkpad.fab.redhat.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> Message-ID: <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> On May 29, 2009, at 2:44, Richard W.M. Jones wrote: > If we put an FTP server inside the appliance, things are better. FTP > clients are widely available, either standalone or as libraries for > many programming languages. They just need to be able to make a TCP > connection, so there is no requirement for special privileges. On the > server side (inside the appliance) the FTP server is running as "root" > so it can make arbitrary changes to the filesystem. Please make FTP die already! :-) How about HTTP (WebDAV) or - better - just use the ssh file transfer protocol? - ask -- http://develooper.com/ - http://askask.com/ From charles at dyfis.net Fri May 29 10:50:47 2009 From: charles at dyfis.net (Charles Duffy) Date: Fri, 29 May 2009 05:50:47 -0500 Subject: [fedora-virt] Installing kernel packages under libguestfs Message-ID: <4A1FBE07.9010405@dyfis.net> I'm trying to upgrade kernel packages via libguestfs -- but grubby and mkinitrd don't deal well at all with having a less-than-fully-populated /dev, either failing to operate at all or producing an initrd with LVM support omitted. Before I spend too much more time on this one -- any hints or recipes? Thanks! From rjones at redhat.com Fri May 29 11:10:05 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 12:10:05 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> Message-ID: <20090529111005.GA6904@thinkpad.fab.redhat.com> On Fri, May 29, 2009 at 03:13:10AM -0700, Ask Bj?rn Hansen wrote: > On May 29, 2009, at 2:44, Richard W.M. Jones wrote: >> If we put an FTP server inside the appliance, things are better. FTP >> clients are widely available, either standalone or as libraries for >> many programming languages. They just need to be able to make a TCP >> connection, so there is no requirement for special privileges. On the >> server side (inside the appliance) the FTP server is running as "root" >> so it can make arbitrary changes to the filesystem. > > Please make FTP die already! :-) > > How about HTTP (WebDAV) or - better - just use the ssh file transfer > protocol? ssh isn't too widely supported (from programming language libraries). Does libssh2 even support scp? WebDAV is possible, but it comes down to how much crap it will depend upon, all of which needs to go into the appliance. FTP is widely supported and well understood, and the servers are old, small, self-contained code. I really think it's better for this, and not just because I once wrote an FTP server ... Rich. From charles at dyfis.net Fri May 29 11:13:05 2009 From: charles at dyfis.net (Charles Duffy) Date: Fri, 29 May 2009 06:13:05 -0500 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529111005.GA6904@thinkpad.fab.redhat.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> Message-ID: <4A1FC341.3010708@dyfis.net> Richard W.M. Jones wrote: > FTP is widely supported and well understood, and the servers are old, > small, self-contained code. I really think it's better for this, and > not just because I once wrote an FTP server ... > Just to weigh in -- my only concern with FTP is that it's limited in what metadata it will transport. ccgfs, for instance, supports full POSIX ACLs and xattrs. Certainly, that's overkill for the common use cases here -- but maybe something to keep in mind. From rjones at redhat.com Fri May 29 11:13:14 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 12:13:14 +0100 Subject: [fedora-virt] Installing kernel packages under libguestfs In-Reply-To: <4A1FBE07.9010405@dyfis.net> References: <4A1FBE07.9010405@dyfis.net> Message-ID: <20090529111314.GB4185@thinkpad.fab.redhat.com> On Fri, May 29, 2009 at 05:50:47AM -0500, Charles Duffy wrote: > I'm trying to upgrade kernel packages via libguestfs -- but grubby and > mkinitrd don't deal well at all with having a less-than-fully-populated > /dev, either failing to operate at all or producing an initrd with LVM > support omitted. > > Before I spend too much more time on this one -- any hints or recipes? Quite probably we need to add more /dev devices. For reasons unknown, the base filesystem RPM in Red Hat distributions doesn't set up any device nodes. 'mock' has to do this itself. We do the same thing (copying code from 'mock') in make-initramfs. Can you identify which ones are needed? Try adding them to the make-initramfs '/init' script. As ever, once you get more details open a BZ so we can track this issue. Rich. From berrange at redhat.com Fri May 29 11:18:42 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 29 May 2009 12:18:42 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <4A1FC341.3010708@dyfis.net> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> <4A1FC341.3010708@dyfis.net> Message-ID: <20090529111842.GF29375@redhat.com> On Fri, May 29, 2009 at 06:13:05AM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: > >FTP is widely supported and well understood, and the servers are old, > >small, self-contained code. I really think it's better for this, and > >not just because I once wrote an FTP server ... > > > Just to weigh in -- my only concern with FTP is that it's limited in > what metadata it will transport. ccgfs, for instance, supports full > POSIX ACLs and xattrs. Certainly, that's overkill for the common use > cases here -- but maybe something to keep in mind. I don't think you'd want to transfer ACLs/xattrs from the host into the guest because that implies some fairly close alignment between user / permissioning model between guest & host, cf why NFS isn't a good idea for this. It likeyl be more useful just let the uploaded files get the default ACL/xattr and ue a guestfs command to change them after the fact if needed Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From berrange at redhat.com Fri May 29 11:20:31 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 29 May 2009 12:20:31 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529111005.GA6904@thinkpad.fab.redhat.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> Message-ID: <20090529112031.GG29375@redhat.com> On Fri, May 29, 2009 at 12:10:05PM +0100, Richard W.M. Jones wrote: > On Fri, May 29, 2009 at 03:13:10AM -0700, Ask Bj?rn Hansen wrote: > > On May 29, 2009, at 2:44, Richard W.M. Jones wrote: > >> If we put an FTP server inside the appliance, things are better. FTP > >> clients are widely available, either standalone or as libraries for > >> many programming languages. They just need to be able to make a TCP > >> connection, so there is no requirement for special privileges. On the > >> server side (inside the appliance) the FTP server is running as "root" > >> so it can make arbitrary changes to the filesystem. > > > > Please make FTP die already! :-) > > > > How about HTTP (WebDAV) or - better - just use the ssh file transfer > > protocol? > > ssh isn't too widely supported (from programming language libraries). > Does libssh2 even support scp? > > WebDAV is possible, but it comes down to how much crap it will depend > upon, all of which needs to go into the appliance. > > FTP is widely supported and well understood, and the servers are old, > small, self-contained code. I really think it's better for this, and > not just because I once wrote an FTP server ... I tend to agree with this - ssh is overkill here - the host <->guest channel is already secure by virtue of using vmchannel, so using ssh is not really required. FTP is a really easy to implement, low overhead protocol that'd fit nicely here. I can't really think of what WebDAV would do for this use case that FTP can't do in a simpler manner. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From rjones at redhat.com Fri May 29 11:21:31 2009 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 29 May 2009 12:21:31 +0100 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <4A1FC341.3010708@dyfis.net> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> <4A1FC341.3010708@dyfis.net> Message-ID: <20090529112131.GB6904@thinkpad.fab.redhat.com> On Fri, May 29, 2009 at 06:13:05AM -0500, Charles Duffy wrote: > Richard W.M. Jones wrote: >> FTP is widely supported and well understood, and the servers are old, >> small, self-contained code. I really think it's better for this, and >> not just because I once wrote an FTP server ... >> > Just to weigh in -- my only concern with FTP is that it's limited in > what metadata it will transport. ccgfs, for instance, supports full > POSIX ACLs and xattrs. Certainly, that's overkill for the common use > cases here -- but maybe something to keep in mind. In addition to Dan's reply, as long as 'tar' supports it, you should be able to upload it using the 'tar-in' or 'tgz-in' commands: http://et.redhat.com/~rjones/libguestfs/guestfs.3.html#guestfs_tgz_in I have also added support for cramfs & squashfs to the appliance, which is yet another way to 'get stuff inside' without needing time- consuming uploads. Rich. From armbru at redhat.com Fri May 29 12:57:44 2009 From: armbru at redhat.com (Markus Armbruster) Date: Fri, 29 May 2009 14:57:44 +0200 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529112031.GG29375@redhat.com> (Daniel P. Berrange's message of "Fri\, 29 May 2009 12\:20\:31 +0100") References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> <20090529112031.GG29375@redhat.com> Message-ID: <87vdnkrrqv.fsf@pike.pond.sub.org> "Daniel P. Berrange" writes: > On Fri, May 29, 2009 at 12:10:05PM +0100, Richard W.M. Jones wrote: >> On Fri, May 29, 2009 at 03:13:10AM -0700, Ask Bj?rn Hansen wrote: >> > On May 29, 2009, at 2:44, Richard W.M. Jones wrote: >> >> If we put an FTP server inside the appliance, things are better. FTP >> >> clients are widely available, either standalone or as libraries for >> >> many programming languages. They just need to be able to make a TCP >> >> connection, so there is no requirement for special privileges. On the >> >> server side (inside the appliance) the FTP server is running as "root" >> >> so it can make arbitrary changes to the filesystem. >> > >> > Please make FTP die already! :-) >> > >> > How about HTTP (WebDAV) or - better - just use the ssh file transfer >> > protocol? >> >> ssh isn't too widely supported (from programming language libraries). >> Does libssh2 even support scp? >> >> WebDAV is possible, but it comes down to how much crap it will depend >> upon, all of which needs to go into the appliance. >> >> FTP is widely supported and well understood, and the servers are old, >> small, self-contained code. I really think it's better for this, and >> not just because I once wrote an FTP server ... > > I tend to agree with this - ssh is overkill here - the host <->guest > channel is already secure by virtue of using vmchannel, so using ssh > is not really required. FTP is a really easy to implement, low overhead > protocol that'd fit nicely here. I can't really think of what WebDAV > would do for this use case that FTP can't do in a simpler manner. > > Daniel Sounds like even TFTP would do. From redhat-lists at anselcomputers.com Fri May 29 17:03:18 2009 From: redhat-lists at anselcomputers.com (Michael Ansel) Date: Fri, 29 May 2009 13:03:18 -0400 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529094434.GA4185@thinkpad.fab.redhat.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> Message-ID: <98eec0f70905291003k35ff229fg449a80769738aa62@mail.gmail.com> On Fri, May 29, 2009 at 5:44 AM, Richard W.M. Jones wrote: > On Fri, May 29, 2009 at 03:34:21AM -0400, Michael Ansel wrote: >> 1) root (UID=0, everywhere) is the only one installing packages, so >> the UID mapping lines up perfectly every time > > No - the concept of UIDs in libguestfs is non-intuitive. > > Firstly the normal mode of operation is to run libguestfs as non-root. > There are plenty of reasons why this is a good idea, I think Charles > covered a few of them. ?So everything we do is oriented around making > sure that we don't need to be root to do libguestfs operations. > > However, libguestfs runs qemu/kvm (also as non-root) but that boots a > mini appliance. ?*Inside* the appliance the appliance kernel thinks > everything is running as root. ?It's not really from the point of view > of the host machine, but inside the appliance that's how it looks. > >> 2) you only need root access on the host once when you are building >> the VM; you don't need NFS after it is already built > > With libguestfs currently, you never need root at all, not to build > libguestfs, not to run it, not for any operations using it. > First, I just want to say, you guys are amazing: when do you sleep?! I also want to apologize to the list for my somewhat mis- and un-informed comments last night/early this morning. I do however think I have figured out where my confusion was: I was under the impression that running a virtual machine required root access and libguestfs worked by launching said VM (in order to "run commands in the context of the guest"). It seems I was grossly mistaken on both: virtual machines can run as non-root (though, KVM is currently root only -- possibly changing), and libguestfs does only a miniature launch of the system (? I'm still a little unclear on how the virtual appliance in libguestfs works). I also did not realize that libguestfs was designed with an explicit goal of not requiring root access for anything. In hindsight, I *knew* you didn't have to be root to launch a VM: I've done it countless times to test boot CDs; for some reason (probably because I was up later than normal), I just forgot that fact. Again, I'm very sorry for my ill-informed comments, and have officially decided I'm no longer allowed to post on mailing lists after midnight! Michael Ansel From ask at develooper.com Fri May 29 17:40:24 2009 From: ask at develooper.com (=?ISO-8859-1?Q?Ask_Bj=F8rn_Hansen?=) Date: Fri, 29 May 2009 10:40:24 -0700 Subject: [fedora-virt] libguestfs best practices: Exposing files from the host for the duration of a session In-Reply-To: <20090529111005.GA6904@thinkpad.fab.redhat.com> References: <4A1F29FA.1020100@dyfis.net> <98eec0f70905282138l2e96b368p19ff2f1128771e84@mail.gmail.com> <20090529062735.GD19219@amd.home.annexia.org> <98eec0f70905290034k25ab466fl6682a6afb6e78451@mail.gmail.com> <20090529094434.GA4185@thinkpad.fab.redhat.com> <1AC5E844-83F3-43F5-80E9-AD2B5752AE6A@develooper.com> <20090529111005.GA6904@thinkpad.fab.redhat.com> Message-ID: On May 29, 2009, at 4:10, Richard W.M. Jones wrote: >> just use the ssh file transfer protocol? > > ssh isn't too widely supported (from programming language libraries). > Does libssh2 even support scp? libcurl does. On the server side running sshd (and having sftp-server installed) seems pretty manageable. > WebDAV is possible, but it comes down to how much crap it will depend > upon, all of which needs to go into the appliance. nginx has a reasonable WebDAV implementation and few dependencies IIRC. I don't think WebDAV is worth it though; the only real benefit is that it'd make it easy to make a "browse this disk from your web browser" feature. > FTP is widely supported and well understood, and the servers are old, > small, self-contained code. I really think it's better for this, and > not just because I once wrote an FTP server ... As long as it's just host->guest communication I agree, it makes no difference. If it's ever extended to remote host -> guest communication FTP will just bring tears and misery. - ask