RFC v2 of this series This series implements a basic test job for OVMF guest. The test case will install an OVMF guest and try to boot it. It is also possible to derive a debian HVM installation test case from ts-ovmf-debian-install - the only difference should be the BIOS option. I tested running the test script without specifying BIOS, it worked fine. I just didn''t figure out a sensible way to organize test cases and run jobs. OVMF will need to build for amd64 target and used with QEMU upstream. Not very sure whether I did the right thing to make-flight and sg-run-job though. Wei. Changes in v2: * factor out preseed_base * make installation CD work with seabios Wei Liu (6): TestSupport.pm: add bios option to guest config file Build OVMF for Xen-unstable Debian.pm: factor out preseed_base Introduce ts-ovmf-debian-install make-flight: OVMF test filght sg-run-job: OVMF job Osstest/Debian.pm | 129 +++++++++++++++------------- Osstest/TestSupport.pm | 5 ++ make-flight | 16 +++- sg-run-job | 6 ++ ts-ovmf-debian-install | 218 ++++++++++++++++++++++++++++++++++++++++++++++++ ts-xen-build | 6 +- 6 files changed, 319 insertions(+), 61 deletions(-) create mode 100755 ts-ovmf-debian-install -- 1.7.10.4
Wei Liu
2013-Dec-11 16:11 UTC
[PATCH RFC V2 1/6] TestSupport.pm: add bios option to guest config file
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
---
Osstest/TestSupport.pm | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/Osstest/TestSupport.pm b/Osstest/TestSupport.pm
index 233d453..dbf9061 100644
--- a/Osstest/TestSupport.pm
+++ b/Osstest/TestSupport.pm
@@ -1424,6 +1424,11 @@ END
$cfg .=
"device_model_version=''$devmodel''\n";
}
+ my $bios = $xopts{''Bios''};
+ if (defined $bios) {
+ $cfg .= "bios=''$bios''\n";
+ }
+
my $cfgpath= prepareguest_part_xencfg($ho, $gho, $ram_mb, \%xopts, $cfg);
target_cmd_root($ho, <<END);
(echo $passwd; echo $passwd) | vncpasswd $gho->{Guest}.vncpw
--
1.7.10.4
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
make-flight | 10 +++++++++-
ts-xen-build | 6 +++++-
2 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/make-flight b/make-flight
index 65e3abc..8d900e2 100755
--- a/make-flight
+++ b/make-flight
@@ -88,6 +88,14 @@ if [ x$buildflight = x ]; then
*) enable_xend=false;;
esac
+ case "$xenbranch" in
+ xen-3.*-testing) enable_ovmf=false;;
+ xen-4.0-testing) enable_ovmf=false;;
+ xen-4.1-testing) enable_ovmf=false;;
+ xen-4.2-testing) enable_ovmf=false;;
+ *) enable_ovmf=true;
+ esac
+
eval "
arch_runvars=\"\$ARCH_RUNVARS_$arch\"
"
@@ -95,7 +103,7 @@ if [ x$buildflight = x ]; then
build_hostflags=share-build-$suite-$arch,arch-$arch,suite-$suite,purpose-build
./cs-job-create $flight build-$arch build \
- arch=$arch enable_xend=$enable_xend \
+ arch=$arch enable_xend=$enable_xend enable_ovmf=$enable_ovmf \
tree_qemu=$TREE_QEMU \
tree_qemuu=$TREE_QEMU_UPSTREAM \
tree_xen=$TREE_XEN \
diff --git a/ts-xen-build b/ts-xen-build
index ebb7d6f..74d17f0 100755
--- a/ts-xen-build
+++ b/ts-xen-build
@@ -125,14 +125,18 @@ sub build () {
'' buildconfigs/enable-xen-config
END
my $xend_opt= $r{enable_xend} =~ m/true/ ? "--enable-xend" :
"--disable-xend";
+ my $ovmf_opt= $r{enable_ovmf} =~ m/true/ ? "--enable-ovmf" :
"--disable-ovmf";
buildcmd_stamped_logged(300, ''configure'',
<<END,<<END,<<END);
if test -f configure; then
if grep -q -- $xend_opt tools/configure ; then
xend=$xend_opt
fi
+ if grep -q -- $ovmf_opt tools/configure ; then
+ ovmf=$ovmf_opt
+ fi
END
- ./configure --sysconfdir=/etc \$xend
+ ./configure --sysconfdir=/etc \$xend \$ovmf
END
fi
END
--
1.7.10.4
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
Osstest/Debian.pm | 129 +++++++++++++++++++++++++++++------------------------
1 file changed, 70 insertions(+), 59 deletions(-)
diff --git a/Osstest/Debian.pm b/Osstest/Debian.pm
index e51a233..54bc843 100644
--- a/Osstest/Debian.pm
+++ b/Osstest/Debian.pm
@@ -33,6 +33,7 @@ BEGIN {
@ISA = qw(Exporter);
@EXPORT = qw(debian_boot_setup
%preseed_cmds
+ preseed_base
preseed_create
preseed_hook_command preseed_hook_installscript
di_installcmdline_core
@@ -432,7 +433,73 @@ sub di_installcmdline_core ($$;@) {
if defined $debconf_priority;
return @cl;
-}
+}
+
+sub preseed_base ($) {
+ my ($suite) = @_;
+
+ return <<''END'';
+d-i mirror/suite string $suite
+
+d-i debian-installer/locale string en_GB
+d-i console-keymaps-at/keymap select gb
+d-i keyboard-configuration/xkb-keymap string en_GB
+
+#d-i debconf/frontend string readline
+
+d-i mirror/country string manual
+d-i mirror/http/proxy string
+
+d-i clock-setup/utc boolean true
+d-i time/zone string Europe/London
+d-i clock-setup/ntp boolean true
+
+d-i partman-md/device_remove_md boolean true
+d-i partman-lvm/device_remove_lvm boolean true
+d-i partman-partitioning/confirm_write_new_label boolean true
+d-i partman/choose_partition select finish
+d-i partman/confirm boolean true
+d-i partman-lvm/confirm boolean true
+
+d-i partman/confirm_nooverwrite true
+d-i partman-lvm/confirm_nooverwrite true
+d-i partman-md/confirm_nooverwrite true
+d-i partman-crypto/confirm_nooverwrite true
+
+#d-i netcfg/disable_dhcp boolean true
+d-i netcfg/get_nameservers string $c{NetNameservers}
+#d-i netcfg/get_netmask string \$c{NetNetmask}
+#d-i netcfg/get_gateway string \$c{NetGateway}
+d-i netcfg/confirm_static boolean true
+d-i netcfg/get_domain string $c{TestHostDomain}
+d-i netcfg/wireless_wep string
+
+d-i passwd/root-password password xenroot
+d-i passwd/root-password-again password xenroot
+d-i passwd/user-fullname string FLOSS Xen Test
+d-i passwd/username string osstest
+d-i passwd/user-password password osstest
+d-i passwd/user-password-again password osstest
+
+console-common console-data/keymap/policy select Don''t touch
keymap
+console-data console-data/keymap/policy select Don''t touch
keymap
+console-data console-data/keymap/family select qwerty
+console-data console-data/keymap/template/layout select British
+
+popularity-contest popularity-contest/participate boolean false
+tasksel tasksel/first multiselect standard, web-server
+
+d-i grub-installer/only_debian boolean true
+
+d-i finish-install/keep-consoles boolean true
+d-i finish-install/reboot_in_progress note
+d-i cdrom-detect/eject boolean false
+
+d-i mirror/http/hostname string $c{DebianMirrorHost}
+d-i mirror/http/directory string /$c{DebianMirrorSubpath}
+
+END
+}
sub preseed_create ($$;@) {
my ($ho, $sfx, %xopts) = @_;
@@ -619,45 +686,12 @@ END
my $extra_packages = join(",",@extra_packages);
- my $preseed_file= (<<END);
-d-i mirror/suite string $suite
-
-d-i debian-installer/locale string en_GB
-d-i console-keymaps-at/keymap select gb
-d-i keyboard-configuration/xkb-keymap string en_GB
-
-#d-i debconf/frontend string readline
-
-d-i mirror/country string manual
-d-i mirror/http/proxy string
-
-d-i clock-setup/utc boolean true
-d-i time/zone string Europe/London
-d-i clock-setup/ntp boolean true
+ my $preseed_file= preseed_base($suite);
+ $preseed_file .= (<<END);
d-i partman-auto/method string lvm
#d-i partman-auto/method string regular
-d-i partman-md/device_remove_md boolean true
-d-i partman-lvm/device_remove_lvm boolean true
-d-i partman-partitioning/confirm_write_new_label boolean true
-d-i partman/choose_partition select finish
-d-i partman/confirm boolean true
-d-i partman-lvm/confirm boolean true
-
-d-i partman/confirm_nooverwrite true
-d-i partman-lvm/confirm_nooverwrite true
-d-i partman-md/confirm_nooverwrite true
-d-i partman-crypto/confirm_nooverwrite true
-
-#d-i netcfg/disable_dhcp boolean true
-d-i netcfg/get_nameservers string $c{NetNameservers}
-#d-i netcfg/get_netmask string \$c{NetNetmask}
-#d-i netcfg/get_gateway string \$c{NetGateway}
-d-i netcfg/confirm_static boolean true
-d-i netcfg/get_domain string $c{TestHostDomain}
-d-i netcfg/wireless_wep string
-
#d-i partman-auto/init_automatically_partition select regular
d-i partman-auto/disk string $disk
@@ -685,32 +719,9 @@ d-i partman-auto/expert_recipe string \\
lv_name{ dummy } \\
.
-d-i passwd/root-password password xenroot
-d-i passwd/root-password-again password xenroot
-d-i passwd/user-fullname string FLOSS Xen Test
-d-i passwd/username string osstest
-d-i passwd/user-password password osstest
-d-i passwd/user-password-again password osstest
-
-console-common console-data/keymap/policy select Don''t touch
keymap
-console-data console-data/keymap/policy select Don''t touch
keymap
-console-data console-data/keymap/family select qwerty
-console-data console-data/keymap/template/layout select British
-
-popularity-contest popularity-contest/participate boolean false
-tasksel tasksel/first multiselect standard, web-server
d-i pkgsel/include string openssh-server, ntp, ntpdate, $extra_packages
-d-i grub-installer/only_debian boolean true
-
-d-i finish-install/keep-consoles boolean true
-d-i finish-install/reboot_in_progress note
-d-i cdrom-detect/eject boolean false
-
-d-i mirror/http/hostname string $c{DebianMirrorHost}
-d-i mirror/http/directory string /$c{DebianMirrorSubpath}
-
$xopts{ExtraPreseed}
END
--
1.7.10.4
This is OVMF guest test case. It resembles ts-redhat-install:
1. prepare a auto install CD
2. install OVMF guest
3. test if the guets is up
The installtion CD is also bootable with seabios.
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
ts-ovmf-debian-install | 218 ++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 218 insertions(+)
create mode 100755 ts-ovmf-debian-install
diff --git a/ts-ovmf-debian-install b/ts-ovmf-debian-install
new file mode 100755
index 0000000..2e19348
--- /dev/null
+++ b/ts-ovmf-debian-install
@@ -0,0 +1,218 @@
+#!/usr/bin/perl -w
+# This is part of "osstest", an automated testing framework for Xen.
+# Copyright (C) 2009-2013 Citrix Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+use strict qw(vars);
+use DBI;
+use Osstest;
+use Osstest::Debian;
+use Osstest::TestSupport;
+
+tsreadconfig();
+
+our $stage=0;
+if (@ARGV && $ARGV[0] =~ m/^--stage(\d+)$/) { $stage=$1; shift @ARGV; }
+
+our ($whhost,$gn) = @ARGV;
+$whhost ||= ''host'';
+$gn ||= ''ovmf'';
+
+our $ho= selecthost($whhost);
+
+our $ram_mb= 768;
+our $disk_mb= 10000;
+
+our $guesthost= "$gn.guest.osstest";
+our $gho;
+
+our $xl= toolstack()->{Command};
+
+
+sub preseed () {
+
+ my $preseed_file = preseed_base(''wheezy'');
+
+ $preseed_file .= (<<END);
+d-i netcfg/get_hostname string debian-hvm
+
+d-i partman-auto/disk string /dev/xvda
+d-i partman-auto/method string regular
+
+d-i partman-auto/expert_recipe string \\
+ boot-root :: \\
+ 512 50 512 vfat \\
+ \$primary{ } \$bootable{ } \\
+ method{ efi } format{ } \\
+ use_filesystem{ } filesystem{ vfat } \\
+ mountpoint{ /boot/efi } \\
+ . \\
+ 5000 50 5000 ext4 \\
+ method{ format } format{ } \\
+ use_filesystem{ } filesystem{ ext4 } \\
+ mountpoint{ / } \\
+ . \\
+ 512 30 100% linux-swap \\
+ method{ swap } format{ } \\
+ .
+
+d-i apt-setup/use_mirror boolean false
+d-i apt-setup/another boolean false
+d-i apt-setup/non-free boolean false
+d-i apt-setup/contrib boolean false
+d-i apt-setup/cdrom/set-first boolean false
+
+d-i pkgsel/include string openssh-server, ntp, ntpdate,
+
+d-i preseed/early_command string
+
+d-i preseed/late_command string \\
+ in-target mkdir -p /boot/efi/EFI/boot; \\
+ in-target cp /boot/efi/EFI/debian/grubx64.efi /boot/efi/EFI/boot/bootx64.efi
;\\
+ in-target mkdir -p /root/.ssh; \\
+ cp /authorized_keys /target/root/.ssh
+
+d-i clock-setup/ntp-server string ntp.uk.xensource.com
+
+END
+ return $preseed_file;
+}
+
+sub grub_cfg () {
+
+ return <<"END";
+set default="0"
+set timeout=5
+
+menuentry ''OVMF guest auto Install'' {
+ linux /install.amd/vmlinuz console=vga console=ttyS0,115200n8
preseed/file=/preseed.cfg
+ initrd /install.amd/initrd.gz
+}
+END
+}
+
+sub isolinux_cfg () {
+ return <<"END";
+ default autoinstall
+ prompt 0
+ timeout 0
+
+ label autoinstall
+ kernel /install.amd/vmlinuz
+ append video=vesa:ywrap,mtrr vga=788 console=ttyS0,115200n8
preseed/file=/preseed.cfg initrd=/install.amd/initrd.gz
+END
+}
+
+our $emptyiso= ''/root/empty.iso'';
+
+sub prep () {
+ target_install_packages_norec($ho, qw(lvm2 rsync xorriso genisoimage));
+
+ my $isotimeout= 600;
+
+ $gho= prepareguest($ho, $gn, $guesthost, 22,
+ $disk_mb + 1,
+ 100);
+
+ my $newiso= ''/root/newiso'';
+ my $emptydir= ''/root/empty-dir'';
+ my $initrddir= ''/root/initrd-dir'';
+
+ my @isogen_base= qw(-R -J -T);
+ my @xorriso_opts= qw(-as mkisofs
+ -r
+ -b isolinux/isolinux.bin
+ -c isolinux/boot.cat
+ -no-emul-boot
+ -boot-load-size 4
+ -boot-info-table
+ -eltorito-alt-boot
+ -e boot/grub/efi.img
+ -no-emul-boot);
+
+ target_putfilecontents_root_stash($ho, 10, preseed(),
+ "/root/guest-preseed");
+ target_putfilecontents_root_stash($ho, 10, authorized_keys(),
+ "/root/guest-authkeys");
+ target_cmd_root($ho, <<END, 60);
+ mkdir -p $emptydir
+ genisoimage -o $emptyiso @isogen_base $emptydir/.
+END
+
+ more_prepareguest_hvm($ho,$gho, $ram_mb, $disk_mb,
+ OnReboot => ''preserve'',
+ Bios => ''ovmf'',
+ PostImageHook => sub {
+ target_cmd_root($ho, <<END, $isotimeout);
+ set -x
+ umount /mnt ||:
+ rm -rf $newiso
+ mount -o loop -r $gho->{Rimage} /mnt
+ mkdir $newiso
+ cp -a /mnt/. $newiso/.
+ umount /mnt
+ rm -rf $initrddir
+ mkdir $initrddir
+ cd $initrddir
+ gzip -d \< $newiso/install.amd/initrd.gz | cpio --extract
--make-directories --no-absolute-filename
+ cp /root/guest-preseed preseed.cfg
+ cp /root/guest-authkeys authorized_keys
+ find . | cpio -H newc --create | gzip -9 \>
$newiso/install.amd/initrd.gz
+ cd -
+ rm -rf $initrddir
+ cd $newiso
+ md5sum `find -follow -type f` \> md5sum.txt
+ cd -
+END
+ target_putfilecontents_root_stash($ho, 10, grub_cfg(),
+ "$newiso/debian/boot/grub/grub.cfg");
+
+ target_putfilecontents_root_stash($ho, 10, isolinux_cfg(),
+ "$newiso/isolinux/isolinux.cfg");
+
+ target_cmd_root($ho, <<END, $isotimeout);
+ xorriso @xorriso_opts -o $gho->{Rimage} $newiso/.
+END
+ });
+}
+
+sub start () {
+ target_cmd_root($ho, "$xl create $gho->{CfgPath}", 100);
+}
+
+sub rewrite_config_nocd () {
+ guest_editconfig($ho, $gho, sub {
+ if (m/^\s*disk\s*\=/ .. /\]/) {
+ s/\Q$gho->{Rimage}\E/$emptyiso/;
+ }
+ s/^on_reboot.*/on_reboot=''restart''/;
+ });
+}
+
+if (!$stage) {
+ prep();
+ start();
+} else {
+ $gho= selectguest($gn,$gho);
+}
+if ($stage<2) {
+ guest_await_reboot($ho,$gho,2000);
+ guest_destroy($ho,$gho);
+}
+
+rewrite_config_nocd();
+start();
+guest_await_dhcp_tcp($gho,300);
+guest_check_up($gho);
--
1.7.10.4
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
make-flight | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/make-flight b/make-flight
index 8d900e2..51aebe2 100755
--- a/make-flight
+++ b/make-flight
@@ -418,7 +418,13 @@ for xenarch in ${TEST_ARCHES- i386 amd64 armhf } ; do
test-win xl $qemuu_runvar \
win_image=win7-x64.iso \
all_hostflags=$most_hostflags,hvm
+ fi
+ if [ $xenarch = amd64 -a "x$qemuu_suffix" = "x-qemuu"
]; then
+ job_create_test test-$xenarch$kern-$dom0arch-xl$qemuu_suffix-ovmf-amd64 \
+ test-debianhvm xl $qemuu_runvar \
+ ovmf_image=debian-7.2.0-amd64-CD-1.iso \
+ all_hostflags=$most_hostflags,hvm
fi
if [ $xenarch = amd64 -a $dom0arch = i386 -a "$kern" =
"" ]; then
--
1.7.10.4
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
sg-run-job | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/sg-run-job b/sg-run-job
index f91da85..29e4376 100755
--- a/sg-run-job
+++ b/sg-run-job
@@ -264,6 +264,12 @@ proc run-job/test-rhelhvm {} {
test-guest-nomigr redhat
}
+proc need-hosts/test-debianhvm {} { return host }
+proc run-job/test-debianhvm {} {
+ run-ts . = ts-ovmf-debian-install
+ test-guest ovmf
+}
+
proc need-hosts/test-pair {} { return {src_host dst_host} }
proc run-job/test-pair {} {
run-ts . = ts-debian-install dst_host
--
1.7.10.4
Dario Faggioli
2013-Dec-12 08:49 UTC
Re: [PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install
On mer, 2013-12-11 at 16:11 +0000, Wei Liu wrote:> diff --git a/ts-ovmf-debian-install b/ts-ovmf-debian-install > new file mode 100755 > index 0000000..2e19348 > --- /dev/null > +++ b/ts-ovmf-debian-install> +sub start () { > + target_cmd_root($ho, "$xl create $gho->{CfgPath}", 100); > +} > +Something like "toolstack()->{Command}" instead of "xl" ? I don''t know how important this is, though, as I don''t think this has much chances to work with xend, has it?> +rewrite_config_nocd(); > +start(); > +guest_await_dhcp_tcp($gho,300); > +guest_check_up($gho); >Does this mean that the guest remains up after the end of the test? This is also probably not a big deal, but, shouldn''t we tear it down? Regards, Dario -- <<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
On Thu, Dec 12, 2013 at 09:49:11AM +0100, Dario Faggioli wrote:> On mer, 2013-12-11 at 16:11 +0000, Wei Liu wrote: > > diff --git a/ts-ovmf-debian-install b/ts-ovmf-debian-install > > new file mode 100755 > > index 0000000..2e19348 > > --- /dev/null > > +++ b/ts-ovmf-debian-install > > > +sub start () { > > + target_cmd_root($ho, "$xl create $gho->{CfgPath}", 100); > > +} > > + > Something like "toolstack()->{Command}" instead of "xl" ? >$xl is in fact toolstack()->{Command}.> I don''t know how important this is, though, as I don''t think this has > much chances to work with xend, has it? >No. Xend is deprecated so I don''t bother fiddling with that. However this doesn''t prevent people using Xend from using OVMF. It''s just firmware after all.> > +rewrite_config_nocd(); > > +start(); > > +guest_await_dhcp_tcp($gho,300); > > +guest_check_up($gho); > > > Does this mean that the guest remains up after the end of the test? This > is also probably not a big deal, but, shouldn''t we tear it down? >I copied this from ts-redhat-install. My guess would be we might need to check if there''s anything wrong with DomU, especially when test fails. Destroying DomU will wipe out useful information. Wei.> Regards, > Dario > > -- > <<This happens because I choose it to happen!>> (Raistlin Majere) > ----------------------------------------------------------------- > Dario Faggioli, Ph.D, http://about.me/dario.faggioli > Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) >
Dario Faggioli
2013-Dec-12 12:24 UTC
Re: [PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install
On gio, 2013-12-12 at 12:16 +0000, Wei Liu wrote:> On Thu, Dec 12, 2013 at 09:49:11AM +0100, Dario Faggioli wrote: > > On mer, 2013-12-11 at 16:11 +0000, Wei Liu wrote: > > > diff --git a/ts-ovmf-debian-install b/ts-ovmf-debian-install > > > new file mode 100755 > > > index 0000000..2e19348 > > > --- /dev/null > > > +++ b/ts-ovmf-debian-install > > > > > +sub start () { > > > + target_cmd_root($ho, "$xl create $gho->{CfgPath}", 100); > > > +} > > > + > > Something like "toolstack()->{Command}" instead of "xl" ? > > > > $xl is in fact toolstack()->{Command}. >I know.> > I don''t know how important this is, though, as I don''t think this has > > much chances to work with xend, has it? > > > > No. Xend is deprecated so I don''t bother fiddling with that. >Ok. I was also thinking that keeping these things general and abstract enough would be better if, at some point, we''d like to intdocude other toolstack (e.g., virsh/libvirt). However, changes will be needed anyway, e.g., to the "create" part... so, yeah, I guess this is a non issue.> > Does this mean that the guest remains up after the end of the test? This > > is also probably not a big deal, but, shouldn''t we tear it down? > > > > I copied this from ts-redhat-install. >You''re right, sorry for this. I thought I see gests being teared down somewhere, but just finished running a (similar) test myself, and the guest is up. Sorry again for the noise. :-) Regards, Dario -- <<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
On Thu, Dec 12, 2013 at 01:24:19PM +0100, Dario Faggioli wrote:> On gio, 2013-12-12 at 12:16 +0000, Wei Liu wrote: > > On Thu, Dec 12, 2013 at 09:49:11AM +0100, Dario Faggioli wrote: > > > On mer, 2013-12-11 at 16:11 +0000, Wei Liu wrote: > > > > diff --git a/ts-ovmf-debian-install b/ts-ovmf-debian-install > > > > new file mode 100755 > > > > index 0000000..2e19348 > > > > --- /dev/null > > > > +++ b/ts-ovmf-debian-install > > > > > > > +sub start () { > > > > + target_cmd_root($ho, "$xl create $gho->{CfgPath}", 100); > > > > +} > > > > + > > > Something like "toolstack()->{Command}" instead of "xl" ? > > > > > > > $xl is in fact toolstack()->{Command}. > > > I know. > > > > I don''t know how important this is, though, as I don''t think this has > > > much chances to work with xend, has it? > > > > > > > No. Xend is deprecated so I don''t bother fiddling with that. > > > Ok. I was also thinking that keeping these things general and abstract > enough would be better if, at some point, we''d like to intdocude other > toolstack (e.g., virsh/libvirt). However, changes will be needed anyway, > e.g., to the "create" part... so, yeah, I guess this is a non issue. >Correct. If we are to support more toolstack there''s certainly more work to do...> > > Does this mean that the guest remains up after the end of the test? This > > > is also probably not a big deal, but, shouldn''t we tear it down? > > > > > > > I copied this from ts-redhat-install. > > > You''re right, sorry for this. I thought I see gests being teared down > somewhere, but just finished running a (similar) test myself, and the > guest is up. > > Sorry again for the noise. :-) >Not at all. It''s not noise. Thanks for reviewing. Wei.> Regards, > Dario > > -- > <<This happens because I choose it to happen!>> (Raistlin Majere) > ----------------------------------------------------------------- > Dario Faggioli, Ph.D, http://about.me/dario.faggioli > Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) >
Wei Liu writes ("[PATCH RFC V2 2/6] Build OVMF for
Xen-unstable"):> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
...> + *) enable_ovmf=true;
This patch has made it into the live version of osstest and can be
dropped from your series.
Ian.
Ian Jackson
2013-Dec-12 15:25 UTC
Re: [PATCH RFC V2 3/6] Debian.pm: factor out preseed_base
Wei Liu writes ("[PATCH RFC V2 3/6] Debian.pm: factor out
preseed_base"):> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
On Thu, Dec 12, 2013 at 03:21:08PM +0000, Ian Jackson wrote:> Wei Liu writes ("[PATCH RFC V2 2/6] Build OVMF for Xen-unstable"): > > Signed-off-by: Wei Liu <wei.liu2@citrix.com> > ... > > + *) enable_ovmf=true; > > This patch has made it into the live version of osstest and can be > dropped from your series. >I think there''s one bit missing in the above patch, we should disable ovmf build for 4.3-testing branch as well. I''m quite surprised it passed the push gate... Wei.> Ian.
On Thu, 2013-12-12 at 15:34 +0000, Wei Liu wrote:> On Thu, Dec 12, 2013 at 03:21:08PM +0000, Ian Jackson wrote: > > Wei Liu writes ("[PATCH RFC V2 2/6] Build OVMF for Xen-unstable"): > > > Signed-off-by: Wei Liu <wei.liu2@citrix.com> > > ... > > > + *) enable_ovmf=true; > > > > This patch has made it into the live version of osstest and can be > > dropped from your series. > > > > I think there''s one bit missing in the above patch, we should disable > ovmf build for 4.3-testing branch as well. > > I''m quite surprised it passed the push gate...Push gate only tests unstable...> > Wei. > > > Ian.
Wei Liu writes ("[PATCH RFC V2 4/6] Introduce
ts-ovmf-debian-install"):> This is OVMF guest test case. It resembles ts-redhat-install:
Thanks. This is coming along but it still has too much clone-and-hack
in it. All duplicated code should be factored out into TestSupport.pm
or Debian.pm.
> + $preseed_file .= (<<END);
> +d-i netcfg/get_hostname string debian-hvm
This string should presumably be the guest ident or something.
> +d-i apt-setup/use_mirror boolean false
I think you need to use the mirror as specified for host installs, so
I think you need to move the mirror stuff into preseed_base.
> +d-i pkgsel/include string openssh-server, ntp, ntpdate,
This is mostly common with preseed_create, so should be there. You
will probably need to pass the xopts to preseed_base.
> +d-i preseed/early_command string
> +
> +d-i preseed/late_command string \\
> + in-target mkdir -p /boot/efi/EFI/boot; \\
> + in-target cp /boot/efi/EFI/debian/grubx64.efi
/boot/efi/EFI/boot/bootx64.efi ;\\
Yuk.
> + in-target mkdir -p /root/.ssh; \\
> + cp /authorized_keys /target/root/.ssh
There''s already machinery for doing this (rather differently) in
preseed-create. Surely you should use it ?
> +d-i clock-setup/ntp-server string ntp.uk.xensource.com
Firstly, if this is needed it should be in preseed-base. Secondly the
hardcoded xensource.com is right out!
> + target_putfilecontents_root_stash($ho, 10, preseed(),
> + "/root/guest-preseed");
These pathnames would ideally have the flight and job in them. Doing
so will make it easier to run multiple tests on the same host install.
(Yes, I know that the ts-redhat-install script doesn''t do that.)
> + target_cmd_root($ho, <<END, 60);
> + mkdir -p $emptydir
> + genisoimage -o $emptyiso @isogen_base $emptydir/.
> +END
This is straight clone-and-hack from ts-redhat-install.
> + more_prepareguest_hvm($ho,$gho, $ram_mb, $disk_mb,
> + OnReboot => ''preserve'',
> + Bios => ''ovmf'',
> + PostImageHook => sub {
> + target_cmd_root($ho, <<END, $isotimeout);
> + set -x
> + umount /mnt ||:
> + rm -rf $newiso
> + mount -o loop -r $gho->{Rimage} /mnt
> + mkdir $newiso
> + cp -a /mnt/. $newiso/.
> + umount /mnt
More clone-and-hack from ts-redhat-install.
> + md5sum `find -follow -type f` \> md5sum.txt
This should probably use find -print0 | xargs -0.
> +sub rewrite_config_nocd () {
> + guest_editconfig($ho, $gho, sub {
> + if (m/^\s*disk\s*\=/ .. /\]/) {
> + s/\Q$gho->{Rimage}\E/$emptyiso/;
> + }
> + s/^on_reboot.*/on_reboot=''restart''/;
> + });
> +}
> +
> +if (!$stage) {
> + prep();
> + start();
> +} else {
> + $gho= selectguest($gn,$gho);
> +}
> +if ($stage<2) {
> + guest_await_reboot($ho,$gho,2000);
> + guest_destroy($ho,$gho);
> +}
> +
> +rewrite_config_nocd();
> +start();
> +guest_await_dhcp_tcp($gho,300);
> +guest_check_up($gho);
More clone and hack.
Thanks,
Ian.
Wei Liu writes ("[PATCH RFC V2 6/6] sg-run-job: OVMF
job"):> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
...> +proc need-hosts/test-debianhvm {} { return host }
> +proc run-job/test-debianhvm {} {
> + run-ts . = ts-ovmf-debian-install
> + test-guest ovmf
> +}
Perhaps "ts-ovmf-debian-install" should be called
"ts-debian-hvm-install" ?
Ian.
The comment on preseed file makes sense to me. My main question is wether you''re suggesting I factor out a script call ts-hvm-install then calls it from ts-redhat-install and ts-debian-hvm-install? Wei.
On Thu, Dec 12, 2013 at 03:42:05PM +0000, Ian Jackson wrote:> Wei Liu writes ("[PATCH RFC V2 6/6] sg-run-job: OVMF job"): > > Signed-off-by: Wei Liu <wei.liu2@citrix.com> > ... > > +proc need-hosts/test-debianhvm {} { return host } > > +proc run-job/test-debianhvm {} { > > + run-ts . = ts-ovmf-debian-install > > + test-guest ovmf > > +} > > Perhaps "ts-ovmf-debian-install" should be called > "ts-debian-hvm-install" ? >That would work as well. I don''t have strong preference on naming stuffs. Wei.> Ian.
Wei Liu writes ("[PATCH RFC V2 5/6] make-flight: OVMF test
filght"):> + job_create_test
test-$xenarch$kern-$dom0arch-xl$qemuu_suffix-ovmf-amd64 \
I haven''t checked the implications of this but when I take this series
I''ll probably mess with it. I hope that''s OK :-).
Thanks,
Ian.
Ian Jackson
2013-Dec-12 16:19 UTC
Re: [PATCH RFC V2 3/6] Debian.pm: factor out preseed_base
Wei Liu writes ("[PATCH RFC V2 3/6] Debian.pm: factor out
preseed_base"):> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
It occurs to me to ask:
> -d-i time/zone string Europe/London
> -d-i clock-setup/ntp boolean true
> + my $preseed_file= preseed_base($suite);
>
> + $preseed_file .= (<<END);
> d-i partman-auto/method string lvm
> #d-i partman-auto/method string regular
Why do you not put the base at the start of the preseed file ?
Ian.
Wei Liu writes ("Re: [PATCH RFC V2 4/6] Introduce
ts-ovmf-debian-install"):> The comment on preseed file makes sense to me.
>
> My main question is wether you''re suggesting I factor out a script
call
> ts-hvm-install then calls it from ts-redhat-install and
> ts-debian-hvm-install?
No, I think you should probably make additional perl functions in
TestSupport.pm and/or Debian.pm.
Doing this as a separate script is going to be very tricky because you
will want to specify complicated things as its arguments (for
example, perhaps you will want some of the common code to take
coderefs as hook-style callbacks).
Ian.
Dario Faggioli writes ("Re: [Xen-devel] [PATCH RFC V2 4/6] Introduce
ts-ovmf-debian-install"):> On gio, 2013-12-12 at 12:16 +0000, Wei Liu wrote:
> > $xl is in fact toolstack()->{Command}.
>
> I know.
So err, I''m not sure what your point was. Wei''s code seems
correct to
me.
> > > I don''t know how important this is, though, as I
don''t think this has
> > > much chances to work with xend, has it?
> >
> > No. Xend is deprecated so I don''t bother fiddling with that.
I think it''s better to be consistent and always use
toolstack()->{Command} even in tests which aren''t going to work with
xend.
> > > Does this mean that the guest remains up after the end of the
test? This
> > > is also probably not a big deal, but, shouldn''t we tear
it down?
> >
> > I copied this from ts-redhat-install.
> >
> You''re right, sorry for this. I thought I see gests being teared
down
> somewhere, but just finished running a (similar) test myself, and the
> guest is up.
Whether the script leaves the guest up is a matter for the individual
script. ts-debian-install && ts-debian-fixup leaves it down, but most
of the other ts-*-install scripts leave it up. I don''t think an
install script should do a shutdown just to provide a uniform interface.
The corresponding knowledge of whether to run start or not is in
sg-run-job.
Thanks,
Ian.
Ian Campbell writes ("Re: [PATCH RFC V2 2/6] Build OVMF for
Xen-unstable"):> On Thu, 2013-12-12 at 15:34 +0000, Wei Liu wrote:
> > I think there''s one bit missing in the above patch, we should
disable
> > ovmf build for 4.3-testing branch as well.
> >
> > I''m quite surprised it passed the push gate...
>
> Push gate only tests unstable...
Oops.
Ian, I''m going to kill the current wheezy attempt and force-push this
into pretest. I''ll rebase the wheezy stuff and hopefully it can go in
tomorrow.
Ian.
Dario Faggioli
2013-Dec-12 16:37 UTC
Re: [PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install
On gio, 2013-12-12 at 16:26 +0000, Ian Jackson wrote:> Dario Faggioli writes ("Re: [Xen-devel] [PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install"): > > On gio, 2013-12-12 at 12:16 +0000, Wei Liu wrote: > > > $xl is in fact toolstack()->{Command}. > > > > I know. > > So err, I''m not sure what your point was. Wei''s code seems correct to > me. >My point was that, even if it''s obviously true that toolstack()->{Command}==xl, and hence Wei''s code looks correct to me too, we better use toolstack()->{Command}.> > > > I don''t know how important this is, though, as I don''t think this has > > > > much chances to work with xend, has it? > > > > > > No. Xend is deprecated so I don''t bother fiddling with that. > > I think it''s better to be consistent and always use > toolstack()->{Command} even in tests which aren''t going to work with > xend. >So, err... Exactly my point. :-) Regards, Dario -- <<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
On Thu, Dec 12, 2013 at 04:19:48PM +0000, Ian Jackson wrote:> Wei Liu writes ("[PATCH RFC V2 3/6] Debian.pm: factor out preseed_base"): > > Signed-off-by: Wei Liu <wei.liu2@citrix.com> > > It occurs to me to ask: > > > -d-i time/zone string Europe/London > > -d-i clock-setup/ntp boolean true > > + my $preseed_file= preseed_base($suite); > > > > + $preseed_file .= (<<END); > > d-i partman-auto/method string lvm > > #d-i partman-auto/method string regular > > Why do you not put the base at the start of the preseed file ? >What do you mean? Data generated by preseed_base is at the start of preseed file. Wei.> Ian.
Ian Jackson
2013-Dec-12 16:50 UTC
Re: [PATCH RFC V2 3/6] Debian.pm: factor out preseed_base
Wei Liu writes ("Re: [PATCH RFC V2 3/6] Debian.pm: factor out
preseed_base"):> On Thu, Dec 12, 2013 at 04:19:48PM +0000, Ian Jackson wrote:
> > Why do you not put the base at the start of the preseed file ?
> >
>
> What do you mean? Data generated by preseed_base is at the start of
> preseed file.
I mean "I didn''t read your diff carefully enough". Sorry!
Ian.
On Thu, 2013-12-12 at 16:34 +0000, Ian Jackson wrote:> Ian Campbell writes ("Re: [PATCH RFC V2 2/6] Build OVMF for Xen-unstable"): > > On Thu, 2013-12-12 at 15:34 +0000, Wei Liu wrote: > > > I think there''s one bit missing in the above patch, we should disable > > > ovmf build for 4.3-testing branch as well. > > > > > > I''m quite surprised it passed the push gate... > > > > Push gate only tests unstable... > > Oops. > > Ian, I''m going to kill the current wheezy attempt and force-push this > into pretest. I''ll rebase the wheezy stuff and hopefully it can go in > tomorrow.OK, I''ll expect cron spam then ;-)> > Ian.
On Thu, Dec 12, 2013 at 03:40:58PM +0000, Ian Jackson wrote:> Wei Liu writes ("[PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install"): > > This is OVMF guest test case. It resembles ts-redhat-install: > > Thanks. This is coming along but it still has too much clone-and-hack > in it. All duplicated code should be factored out into TestSupport.pm > or Debian.pm. > > > + $preseed_file .= (<<END); > > +d-i netcfg/get_hostname string debian-hvm > > This string should presumably be the guest ident or something. > > > +d-i apt-setup/use_mirror boolean false > > I think you need to use the mirror as specified for host installs, so > I think you need to move the mirror stuff into preseed_base. > > > +d-i pkgsel/include string openssh-server, ntp, ntpdate, > > This is mostly common with preseed_create, so should be there. You > will probably need to pass the xopts to preseed_base. > > > +d-i preseed/early_command string > > + > > +d-i preseed/late_command string \\ > > + in-target mkdir -p /boot/efi/EFI/boot; \\ > > + in-target cp /boot/efi/EFI/debian/grubx64.efi /boot/efi/EFI/boot/bootx64.efi ;\\ > > Yuk. > > > + in-target mkdir -p /root/.ssh; \\ > > + cp /authorized_keys /target/root/.ssh > > There''s already machinery for doing this (rather differently) in > preseed-create. Surely you should use it ? >Using the web infrastructure for this looks overkill. I''ve replaced this part with a call to ''echo -e $authkeys > authorized_keys'', elimiating the need for extra file. Wei.
On Thu, Dec 12, 2013 at 05:57:23PM +0000, Wei Liu wrote:> On Thu, Dec 12, 2013 at 03:40:58PM +0000, Ian Jackson wrote: > > Wei Liu writes ("[PATCH RFC V2 4/6] Introduce ts-ovmf-debian-install"): > > > This is OVMF guest test case. It resembles ts-redhat-install: > > > > Thanks. This is coming along but it still has too much clone-and-hack > > in it. All duplicated code should be factored out into TestSupport.pm > > or Debian.pm. > > > > > + $preseed_file .= (<<END); > > > +d-i netcfg/get_hostname string debian-hvm > > > > This string should presumably be the guest ident or something. > > > > > +d-i apt-setup/use_mirror boolean false > > > > I think you need to use the mirror as specified for host installs, so > > I think you need to move the mirror stuff into preseed_base. > > > > > +d-i pkgsel/include string openssh-server, ntp, ntpdate, > > > > This is mostly common with preseed_create, so should be there. You > > will probably need to pass the xopts to preseed_base. > > > > > +d-i preseed/early_command string > > > + > > > +d-i preseed/late_command string \\ > > > + in-target mkdir -p /boot/efi/EFI/boot; \\ > > > + in-target cp /boot/efi/EFI/debian/grubx64.efi /boot/efi/EFI/boot/bootx64.efi ;\\ > > > > Yuk. > > > > > + in-target mkdir -p /root/.ssh; \\ > > > + cp /authorized_keys /target/root/.ssh > > > > There''s already machinery for doing this (rather differently) in > > preseed-create. Surely you should use it ? > > > > Using the web infrastructure for this looks overkill. I''ve replaced this > part with a call to ''echo -e $authkeys > authorized_keys'', elimiating > the need for extra file. >Phew, it didn''t work. The rune failed to run as d-i late_command. Will need to look for another way to do this. Wei.> Wei.