- Devel - libvirt List Archives

[libvirt] [PATCH v2] docs: Improve the schema for fs device
by Osier Yang 17 Apr '12

17 Apr '12

<filesystemtgt> is redundant, as every group uses it; <address> shouldn't be in <filesystemtgt> in case of the meaning could be "filesystemtarget"; The elements <address>, <alias>, <target>, ... should be interleaved. --- v1 ~ v2: No changes --- docs/schemas/domaincommon.rng | 58 ++++++++++++++++++---------------------- 1 files changed, 26 insertions(+), 32 deletions(-) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 0cc04af..f116710 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1210,7 +1210,6 @@ </attribute> <empty/> </element> - <ref name="filesystemtgt"/> </interleave> </group> <group> @@ -1224,7 +1223,6 @@ </attribute> <empty/> </element> - <ref name="filesystemtgt"/> </interleave> </group> <group> @@ -1257,7 +1255,6 @@ <empty/> </element> </optional> - <ref name="filesystemtgt"/> </interleave> </group> <group> @@ -1271,42 +1268,39 @@ </attribute> <empty/> </element> - <ref name="filesystemtgt"/> </interleave> </group> </choice> - <optional> - <attribute name="accessmode"> - <choice> - <value>passthrough</value> - <value>mapped</value> - <value>squash</value> - </choice> - </attribute> - </optional> - <optional> - <element name='readonly'> + <interleave> + <element name="target"> + <attribute name="dir"> + <ref name="absDirPath"/> + </attribute> <empty/> </element> - </optional> + <optional> + <attribute name="accessmode"> + <choice> + <value>passthrough</value> + <value>mapped</value> + <value>squash</value> + </choice> + </attribute> + </optional> + <optional> + <element name='readonly'> + <empty/> + </element> + </optional> + <optional> + <ref name="alias"/> + </optional> + <optional> + <ref name="address"/> + </optional> + </interleave> </element> </define> - <define name="filesystemtgt"> - <interleave> - <optional> - <ref name="alias"/> - </optional> - <optional> - <ref name="address"/> - </optional> - <element name="target"> - <attribute name="dir"> - <ref name="absDirPath"/> - </attribute> - <empty/> - </element> - </interleave> - </define> <!-- An interface description can either be of type bridge in which case it will use a bridging source, or of type ethernet which uses a device -- 1.7.7.3

2 2

[libvirt] [PATCH] libvirt: xen: do not use ioemu type for any emulated NIC
by Stefan Bader 17 Apr '12

17 Apr '12

This is a re-send as there was some positive feedback but the patch itself made it into the repo. -Stefan >From a3198c5c1ae8908818f6c0f0df4237dbe5ddeec7 Mon Sep 17 00:00:00 2001 From: Stefan Bader <stefan.bader(a)canonical.com> Date: Thu, 12 Apr 2012 15:32:41 +0200 Subject: [PATCH] libvirt: xen: do not use ioemu type for any emulated NIC When using the xm/xend stack to manage instances there is a bug that causes the emulated interfaces to be unusable when the vif config contains type=ioemu (MAC address all zero). The current code already has a special quirk to not use this keyword if no specific model is given for the emulated NIC (defaulting to rtl8139). Essentially it works because regardless of the type argument, the Xen stack always creates emulated and paravirt interfaces and lets the guest decide which one to use. So neither xl nor xm stack actually require the type keyword for emulated NICs. Signed-off-by: Stefan Bader <stefan.bader(a)canonical.com> --- src/xenxs/xen_sxpr.c | 4 +++- src/xenxs/xen_xm.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/src/xenxs/xen_sxpr.c b/src/xenxs/xen_sxpr.c index e1bbd76..71602fa 100644 --- a/src/xenxs/xen_sxpr.c +++ b/src/xenxs/xen_sxpr.c @@ -2012,7 +2012,9 @@ xenFormatSxprNet(virConnectPtr conn, } else { virBufferEscapeSexpr(buf, "(model '%s')", def->model); - virBufferAddLit(buf, "(type ioemu)"); + /* See above. Also needed when model is specified. */ + if (xendConfigVersion <= XEND_CONFIG_MAX_VERS_NET_TYPE_IOEMU) + virBufferAddLit(buf, "(type ioemu)"); } if (!isAttach) diff --git a/src/xenxs/xen_xm.c b/src/xenxs/xen_xm.c index d65e97a..e072599 100644 --- a/src/xenxs/xen_xm.c +++ b/src/xenxs/xen_xm.c @@ -1394,7 +1394,9 @@ static int xenFormatXMNet(virConnectPtr conn, } else { virBufferAsprintf(&buf, ",model=%s", net->model); - virBufferAddLit(&buf, ",type=ioemu"); + /* See above. Also needed if model is specified. */ + if (xendConfigVersion <= XEND_CONFIG_MAX_VERS_NET_TYPE_IOEMU) + virBufferAddLit(&buf, ",type=ioemu"); } if (net->ifname) -- 1.7.9.5

4 19

[libvirt] "virsh list" hangs
by Qian Zhang 17 Apr '12

17 Apr '12

Hi, I am using RHEL 6.2 64bit, and the libvirt shipped in it is "libvirt-0.9.4-23.el6.x86_64". I found sometimes the command "virsh list" hangs forever, and same issue for virt-manager which is always in "Connecting" status. But after restarting the libvirtd service, this issue is gone. Is this a bug of libvirt? Any help will be appreciated. Thanks, Qian

4 6

Re: [libvirt] [PATCH 7/8] block: add witness argument to drive-reopen
by Eric Blake 17 Apr '12

17 Apr '12

[adding libvir-list for some interesting discussions on potential libvirt design issues] On 04/17/2012 12:47 AM, Paolo Bonzini wrote: > Il 16/04/2012 23:55, Eric Blake ha scritto: >>> Do transient guests have persistent storage for them in /var while they >>> are running? >> >> Yes - that's how libvirt tracks the pid of the qemu process that it >> should be re-attaching to on a libvirtd restart, as well as several >> other aspects (for example, the capabilities of the qemu binary running >> that pid, in case qemu has been upgraded in the meantime). It's just >> that right now, the files that libvirt stores in /var are intended to be >> internal to libvirt, so management apps shouldn't be poking around in >> there, so much as having an interface to ask libvirt what state things >> are in. > > Can libvirtd report the block-job-completed event to vdsm before the > transient-domain-died event? Yes, it should be possible to organize the code so that on libvirtd restart, libvirt is careful to emit all other events, such as block-job completed/failed, prior to the final domain died event as libvirt cleans out its records of the dead domain. But it probably involves quite a bit of code cleanup on libvirt's side to reach that state. There's also the question of whether vdsm will be listening to the events soon enough; after all, when libvirtd restarts, libvirt is doing quite a bit of bookkeeping before even accepting connections, and if vdsm isn't connected to the restarted libvirt at the time libvirt queues up an event to be sent, then the event is lost. Maybe that implies that libvirt needs to have a configurable-depth array of prior events that can be played back to the user on request. -- Eric Blake eblake(a)redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

2 1

[libvirt] [PATCHv4 00/18] live block migration via virDomainBlockCopy
by Eric Blake 17 Apr '12

17 Apr '12

v3 was here: https://www.redhat.com/archives/libvir-list/2012-April/msg00288.html changes from v3: - split v3 patch 8 into 1/18 and 9/18 - incorporate Laine's review comments on v3 patches 1 and 2 (here 2 and 3) - fix more bugs found during testing - upgrade to Paolo's latest build (which dropped 'mode':'no-backing-file' and replaced it with 'full':'true') - added a new patch 14/18 that lets blockcopy work with SELinux enforcing (it still leaks files rather than clean up context on job abort or pivot, but that's a harder problem to solve) repo.or.cz is acting up on me at the moment, so I won't have a repo to point to until I can investigate why tomorrow. I'll send an interdiff in reply to this message. Adam Litke (2): blockjob: add API for async virDomainBlockJobAbort blockjob: wire up qemu async virDomainBlockJobAbort Eric Blake (16): blockjob: add qemu capabilities related to block jobs blockjob: allow for fast-finishing job blockjob: add new API flags blockjob: add 'blockcopy' to virsh blockjob: enhance xml to track mirrors across libvirtd restart blockjob: react to active block copy blockjob: expose qemu commands for mirrored storage migration blockjob: return appropriate event and info blockjob: support pivot operation on cancel blockjob: implement block copy for qemu blockjob: allow for existing files blockjob: allow mirroring under SELinux blockjob: accommodate RHEL backport names blockjob: add virDomainBlockCopy blockjob: enhance virsh 'blockcopy' blockjob: wire up qemu and RPC for block copy docs/apibuild.py | 1 + docs/formatdomain.html.in | 11 + docs/schemas/domaincommon.rng | 19 ++- include/libvirt/libvirt.h.in | 50 +++++- include/libvirt/virterror.h | 1 + src/conf/domain_conf.c | 54 ++++++ src/conf/domain_conf.h | 6 + src/driver.h | 5 + src/libvirt.c | 223 +++++++++++++++++++++- src/libvirt_private.syms | 1 + src/libvirt_public.syms | 5 + src/qemu/qemu_capabilities.c | 3 + src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_driver.c | 417 +++++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_hotplug.c | 7 + src/qemu/qemu_monitor.c | 50 +++++ src/qemu/qemu_monitor.h | 18 ++- src/qemu/qemu_monitor_json.c | 167 ++++++++++++++--- src/qemu/qemu_monitor_json.h | 19 ++- src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 12 +- src/remote_protocol-structs | 9 + src/rpc/gendispatch.pl | 1 + src/util/virterror.c | 6 + tools/virsh.c | 141 +++++++++++--- tools/virsh.pod | 65 ++++++- 26 files changed, 1195 insertions(+), 99 deletions(-) -- 1.7.7.6

4 58

[libvirt] cgroup cpuset limited by default?
by Marc-André Lureau 17 Apr '12

17 Apr '12

Hi, I noticed recently that my system VMs feel much slower, and that seems to be because of cpuset cgroup which constrain the VM on the first cpu. This doesn't happen with session VMs, since they are not ruled by cgroup. I run f17 with libvirt git, the VMs are created with virt-manager, using mostly default settings <vcpu>4</vcpu> <os> <type arch='x86_64' machine='pc-0.15'>hvm</type> <boot dev='hd'/> </os> <features> <acpi/> <apic/> <pae/> </features> <clock offset='localtime'/> Is that intended or is it a bug? -- Marc-André Lureau

3 6

[libvirt] [PATCH v2] qemu: Split ide-drive into ide-cd and ide-hd
by Osier Yang 17 Apr '12

17 Apr '12

A "ide-drive" device can be either a hard disk or a CD-ROM, if there is ",media=cdrom" specified for the backend, it's a CD-ROM, otherwise it's a hard disk. Upstream qemu splitted "ide-drive" into "ide-hd" and "ide-cd" since commit 1f56e32, and ",media=cdrom" is not required for ide-cd anymore. "ide-drive" is still supported for backwards compatibility, but no doubt we should go foward. --- No changes between v1 and v2, just rebasing. --- src/qemu/qemu_capabilities.c | 3 ++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 25 ++++++++++++++- tests/qemuhelptest.c | 3 +- .../qemuxml2argv-disk-ide-drive-split.args | 8 +++++ .../qemuxml2argv-disk-ide-drive-split.xml | 32 ++++++++++++++++++++ tests/qemuxml2argvtest.c | 3 ++ 7 files changed, 72 insertions(+), 3 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.xml diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 4e22f3e8..3d1fb43 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -160,6 +160,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST, "block-job-sync", /* 90 */ "block-job-async", "scsi-cd", + "ide-cd", ); struct qemu_feature_flags { @@ -1454,6 +1455,8 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr flags) qemuCapsSet(flags, QEMU_CAPS_SCSI_BLOCK); if (strstr(str, "scsi-cd")) qemuCapsSet(flags, QEMU_CAPS_SCSI_CD); + if (strstr(str, "ide-cd")) + qemuCapsSet(flags, QEMU_CAPS_IDE_CD); return 0; } diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 2aba502..7279cdb 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -127,6 +127,7 @@ enum qemuCapsFlags { QEMU_CAPS_BLOCKJOB_SYNC = 90, /* RHEL 6.2 block_job_cancel */ QEMU_CAPS_BLOCKJOB_ASYNC = 91, /* qemu 1.1 block-job-cancel */ QEMU_CAPS_SCSI_CD = 92, /* -device scsi-cd */ + QEMU_CAPS_IDE_CD = 93, /* -device ide-cd */ QEMU_CAPS_LAST, /* this must always be the last item */ }; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ff82b26..26f691f 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1982,6 +1982,9 @@ qemuBuildDriveStr(virConnectPtr conn ATTRIBUTE_UNUSED, if ((disk->bus == VIR_DOMAIN_DISK_BUS_SCSI)) { if (!qemuCapsGet(qemuCaps, QEMU_CAPS_SCSI_CD)) virBufferAddLit(&opt, ",media=cdrom"); + } else if (disk->bus == VIR_DOMAIN_DISK_BUS_IDE) { + if (!qemuCapsGet(qemuCaps, QEMU_CAPS_IDE_CD)) + virBufferAddLit(&opt, ",media=cdrom"); } else { virBufferAddLit(&opt, ",media=cdrom"); } @@ -2205,7 +2208,16 @@ qemuBuildDriveDevStr(virDomainDefPtr def, _("target must be 0 for ide controller")); goto error; } - virBufferAddLit(&opt, "ide-drive"); + + if (qemuCapsGet(qemuCaps, QEMU_CAPS_IDE_CD)) { + if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) + virBufferAddLit(&opt, "ide-cd"); + else + virBufferAddLit(&opt, "ide-hd"); + } else { + virBufferAddLit(&opt, "ide-drive"); + } + virBufferAsprintf(&opt, ",bus=ide.%d,unit=%d", disk->info.addr.drive.bus, disk->info.addr.drive.unit); @@ -2301,7 +2313,16 @@ qemuBuildDriveDevStr(virDomainDefPtr def, _("target must be 0 for ide controller")); goto error; } - virBufferAddLit(&opt, "ide-drive"); + + if (qemuCapsGet(qemuCaps, QEMU_CAPS_IDE_CD)) { + if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) + virBufferAddLit(&opt, "ide-cd"); + else + virBufferAddLit(&opt, "ide-hd"); + } else { + virBufferAddLit(&opt, "ide-drive"); + } + virBufferAsprintf(&opt, ",bus=ahci%d.%d", disk->info.addr.drive.controller, disk->info.addr.drive.unit); diff --git a/tests/qemuhelptest.c b/tests/qemuhelptest.c index a01c389..d23b35a 100644 --- a/tests/qemuhelptest.c +++ b/tests/qemuhelptest.c @@ -676,7 +676,8 @@ mymain(void) QEMU_CAPS_CPU_HOST, QEMU_CAPS_FSDEV_WRITEOUT, QEMU_CAPS_SCSI_BLOCK, - QEMU_CAPS_SCSI_CD); + QEMU_CAPS_SCSI_CD, + QEMU_CAPS_IDE_CD); return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.args b/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.args new file mode 100644 index 0000000..e24490d --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.args @@ -0,0 +1,8 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test \ +/usr/bin/qemu -S -M pc -m 214 -smp 1 -nographic -nodefconfig -nodefaults \ +-monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c \ +-drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-ide0-0-1 \ +-device ide-cd,bus=ide.0,unit=1,drive=drive-ide0-0-1,id=ide0-0-1 \ +-drive file=/tmp/idedisk.img,if=none,id=drive-ide0-0-2 \ +-device ide-hd,bus=ide.0,unit=2,drive=drive-ide0-0-2,id=ide0-0-2 \ +-usb -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.xml b/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.xml new file mode 100644 index 0000000..b965677 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-ide-drive-split.xml @@ -0,0 +1,32 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219136</memory> + <currentMemory unit='KiB'>219136</currentMemory> + <vcpu>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu</emulator> + <disk type='block' device='cdrom'> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='1'/> + </disk> + <disk type='file' device='disk'> + <source file='/tmp/idedisk.img'/> + <target dev='hdc' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='2'/> + </disk> + <controller type='usb' index='0'/> + <controller type='ide' index='0'/> + <controller type='ide' index='1'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index cc1ac44..ee7a8e1 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -758,6 +758,9 @@ mymain(void) QEMU_CAPS_CHARDEV, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); DO_TEST("pseries-vio-address-clash", true, QEMU_CAPS_DRIVE, QEMU_CAPS_CHARDEV, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); + DO_TEST("disk-ide-drive-split", false, + QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, + QEMU_CAPS_IDE_CD); VIR_FREE(driver.stateDir); virCapabilitiesFree(driver.caps); -- 1.7.7.3

2 2

[libvirt] [PATCH v2] qemu: Split scsi-disk into into scsi-hd and scsi-cd
by Osier Yang 17 Apr '12

17 Apr '12

A "scsi-disk" device can be either a hard disk or a CD-ROM, if there is ",media=cdrom" specified for the backend, it's a CD-ROM, otherwise it's a hard disk. But upstream qemu splitted "scsi-disk" into "scsi-hd" and "scsi-cd" since commit b443ae, and ",media=cdrom" is not required for scsi-cd anymore. "scsi-disk" is still supported for backwards compatibility, but no doubt we should go foward. --- No changes since v1 --- src/qemu/qemu_capabilities.c | 3 +++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 39 +++++++++++++++++++++++++++++++-------- tests/qemuhelptest.c | 6 ++++-- tests/qemuxml2argvtest.c | 3 +++ 5 files changed, 42 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index fd4ca4d..4e22f3e8 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -159,6 +159,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST, "block-job-sync", /* 90 */ "block-job-async", + "scsi-cd", ); struct qemu_feature_flags { @@ -1451,6 +1452,8 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr flags) qemuCapsSet(flags, QEMU_CAPS_SCSI_DISK_CHANNEL); if (strstr(str, "scsi-block")) qemuCapsSet(flags, QEMU_CAPS_SCSI_BLOCK); + if (strstr(str, "scsi-cd")) + qemuCapsSet(flags, QEMU_CAPS_SCSI_CD); return 0; } diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 6550d59..2aba502 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -126,6 +126,7 @@ enum qemuCapsFlags { QEMU_CAPS_TRANSACTION = 89, /* transaction monitor command */ QEMU_CAPS_BLOCKJOB_SYNC = 90, /* RHEL 6.2 block_job_cancel */ QEMU_CAPS_BLOCKJOB_ASYNC = 91, /* qemu 1.1 block-job-cancel */ + QEMU_CAPS_SCSI_CD = 92, /* -device scsi-cd */ QEMU_CAPS_LAST, /* this must always be the last item */ }; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index c82f5bc..ff82b26 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1978,8 +1978,14 @@ qemuBuildDriveStr(virConnectPtr conn ATTRIBUTE_UNUSED, else virBufferAsprintf(&opt, "if=%s", bus); - if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) - virBufferAddLit(&opt, ",media=cdrom"); + if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) { + if ((disk->bus == VIR_DOMAIN_DISK_BUS_SCSI)) { + if (!qemuCapsGet(qemuCaps, QEMU_CAPS_SCSI_CD)) + virBufferAddLit(&opt, ",media=cdrom"); + } else { + virBufferAddLit(&opt, ",media=cdrom"); + } + } if (qemuCapsGet(qemuCaps, QEMU_CAPS_DEVICE)) { virBufferAsprintf(&opt, ",id=%s%s", QEMU_DRIVE_HOST_PREFIX, disk->info.alias); @@ -2229,10 +2235,19 @@ qemuBuildDriveDevStr(virDomainDefPtr def, goto error; } - if (disk->device == VIR_DOMAIN_DISK_DEVICE_LUN) + if (disk->device == VIR_DOMAIN_DISK_DEVICE_LUN) { virBufferAddLit(&opt, "scsi-block"); - else - virBufferAddLit(&opt, "scsi-disk"); + } else { + if (qemuCapsGet(qemuCaps, QEMU_CAPS_SCSI_CD)) { + if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) + virBufferAddLit(&opt, "scsi-cd"); + else + virBufferAddLit(&opt, "scsi-hd"); + } else { + virBufferAddLit(&opt, "scsi-disk"); + } + } + virBufferAsprintf(&opt, ",bus=scsi%d.%d,scsi-id=%d", disk->info.addr.drive.controller, disk->info.addr.drive.bus, @@ -2255,10 +2270,18 @@ qemuBuildDriveDevStr(virDomainDefPtr def, } } - if (disk->device != VIR_DOMAIN_DISK_DEVICE_LUN) - virBufferAddLit(&opt, "scsi-disk"); - else + if (disk->device != VIR_DOMAIN_DISK_DEVICE_LUN) { + if (qemuCapsGet(qemuCaps, QEMU_CAPS_SCSI_CD)) { + if (disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM) + virBufferAddLit(&opt, "scsi-cd"); + else + virBufferAddLit(&opt, "scsi-hd"); + } else { + virBufferAddLit(&opt, "scsi-disk"); + } + } else { virBufferAddLit(&opt, "scsi-block"); + } virBufferAsprintf(&opt, ",bus=scsi%d.0,channel=%d,scsi-id=%d,lun=%d", disk->info.addr.drive.controller, diff --git a/tests/qemuhelptest.c b/tests/qemuhelptest.c index 05fdc4d..a01c389 100644 --- a/tests/qemuhelptest.c +++ b/tests/qemuhelptest.c @@ -605,7 +605,8 @@ mymain(void) QEMU_CAPS_VIRTIO_BLK_SCSI, QEMU_CAPS_VIRTIO_BLK_SG_IO, QEMU_CAPS_DRIVE_COPY_ON_READ, - QEMU_CAPS_CPU_HOST); + QEMU_CAPS_CPU_HOST, + QEMU_CAPS_SCSI_CD); DO_TEST("qemu-1.0", 1000000, 0, 0, QEMU_CAPS_VNC_COLON, QEMU_CAPS_NO_REBOOT, @@ -674,7 +675,8 @@ mymain(void) QEMU_CAPS_VIRTIO_BLK_SG_IO, QEMU_CAPS_CPU_HOST, QEMU_CAPS_FSDEV_WRITEOUT, - QEMU_CAPS_SCSI_BLOCK); + QEMU_CAPS_SCSI_BLOCK, + QEMU_CAPS_SCSI_CD); return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; } diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index fdbe95a..cc1ac44 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -463,6 +463,9 @@ mymain(void) QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); DO_TEST("disk-scsi-device-auto", false, QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); + DO_TEST("disk-scsi-disk-split", false, + QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, + QEMU_CAPS_SCSI_CD); DO_TEST("disk-scsi-vscsi", false, QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); DO_TEST("disk-scsi-virtio-scsi", false, -- 1.7.7.3

2 4

[libvirt] [PATCH] Do not enforce source type of console[0]
by Jan Kiszka 17 Apr '12

17 Apr '12

If console[0] is an alias for serial[0], do not enforce the former to have a PTY source type. This breaks serial consoles on stdio and makes no sense. Signed-off-by: Jan Kiszka <jan.kiszka(a)siemens.com> --- src/qemu/qemu_process.c | 1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index ef311d1..216b594 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -1376,7 +1376,6 @@ qemuProcessFindCharDevicePTYs(virDomainObjPtr vm, if ((ret = virDomainChrSourceDefCopy(&chr->source, &((vm->def->serials[0])->source))) != 0) return ret; - chr->source.type = VIR_DOMAIN_CHR_TYPE_PTY; } else { if (chr->source.type == VIR_DOMAIN_CHR_TYPE_PTY && chr->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_VIRTIO) { -- 1.7.3.4

3 3

[libvirt] [PATCH] openvz: wire up more node information functions
by Guido Günther 17 Apr '12

17 Apr '12

in detail nodeGetCPUStats, nodeGetMemoryStats, nodeGetCellsFreeMemory and nodeGetFreeMemory --- src/openvz/openvz_driver.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/src/openvz/openvz_driver.c b/src/openvz/openvz_driver.c index 3b2ffea..41fd8e4 100644 --- a/src/openvz/openvz_driver.c +++ b/src/openvz/openvz_driver.c @@ -1704,6 +1704,10 @@ static virDriver openvzDriver = { .version = openvzGetVersion, /* 0.5.0 */ .getMaxVcpus = openvzGetMaxVCPUs, /* 0.4.6 */ .nodeGetInfo = nodeGetInfo, /* 0.3.2 */ + .nodeGetCPUStats = nodeGetCPUStats, /* 0.9.11 */ + .nodeGetMemoryStats = nodeGetMemoryStats, /* 0.9.11 */ + .nodeGetCellsFreeMemory = nodeGetCellsFreeMemory, /* 0.9.11 */ + .nodeGetFreeMemory = nodeGetFreeMemory, /* 0.9.11 */ .getCapabilities = openvzGetCapabilities, /* 0.4.6 */ .listDomains = openvzListDomains, /* 0.3.1 */ .numOfDomains = openvzNumDomains, /* 0.3.1 */ -- 1.7.9.5

3 4

[libvirt] 2 trivial doc patches
by Christophe Fergeau 16 Apr '12

16 Apr '12

Hey, I've just pushed these 2 documentation patches under the trivial rule. Christophe

1 2

[libvirt] OpenSSH authentification in Java libvirt API
by Alexandre Garnier 16 Apr '12

16 Apr '12

Hello, It seems that today you can’t authenticate on a qemu+ssh connection via OpenSSH without having to manually type the password in CLI or a pop-up window, using the Java API for libvirt. Being able to push a String in OpenSSH user input would be an easier and faster solution, for the end-user. Sincerely, Alexandre

1 0

[libvirt] [PATCH v2] qemuProcessStart: Switch to flags instead of bunch booleans
by Michal Privoznik 16 Apr '12

16 Apr '12

Currently, we have 3 boolean arguments we have to pass to qemuProcessStart(). As libvirt grows it is harder and harder to remember them and their position. Therefore we should switch to flags instead. --- diff to v1: -fix a test for START_PAUSED flag src/qemu/qemu_driver.c | 45 +++++++++++++++++++++++++++++---------------- src/qemu/qemu_migration.c | 7 ++++--- src/qemu/qemu_process.c | 21 +++++++++++++-------- src/qemu/qemu_process.h | 12 ++++++++---- 4 files changed, 54 insertions(+), 31 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 65ed290..436ef37 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1351,10 +1351,16 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, const char *xml, virDomainPtr dom = NULL; virDomainEventPtr event = NULL; virDomainEventPtr event2 = NULL; + unsigned int start_flags = VIR_QEMU_PROCESS_START_COLD; virCheckFlags(VIR_DOMAIN_START_PAUSED | VIR_DOMAIN_START_AUTODESTROY, NULL); + if (flags & VIR_DOMAIN_START_PAUSED) + start_flags |= VIR_QEMU_PROCESS_START_PAUSED; + if (flags & VIR_DOMAIN_START_AUTODESTROY) + start_flags |= VIR_QEMU_PROCESS_START_AUTODESROY; + qemuDriverLock(driver); if (!(def = virDomainDefParseString(driver->caps, xml, QEMU_EXPECTED_VIRT_TYPES, @@ -1383,10 +1389,9 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, const char *xml, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; /* XXXX free the 'vm' we created ? */ - if (qemuProcessStart(conn, driver, vm, NULL, true, - (flags & VIR_DOMAIN_START_PAUSED) != 0, - (flags & VIR_DOMAIN_START_AUTODESTROY) != 0, - -1, NULL, NULL, VIR_NETDEV_VPORT_PROFILE_OP_CREATE) < 0) { + if (qemuProcessStart(conn, driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + start_flags) < 0) { virDomainAuditStart(vm, "booted", false); if (qemuDomainObjEndJob(driver, vm) > 0) qemuDomainRemoveInactive(driver, vm); @@ -4138,9 +4143,9 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, } /* Set the migration source and start it up. */ - ret = qemuProcessStart(conn, driver, vm, "stdio", false, true, - false, *fd, path, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_RESTORE); + ret = qemuProcessStart(conn, driver, vm, "stdio", *fd, path, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_RESTORE, + VIR_QEMU_PROCESS_START_PAUSED); if (intermediatefd != -1) { if (ret < 0) { @@ -4710,6 +4715,10 @@ qemuDomainObjStart(virConnectPtr conn, bool autodestroy = (flags & VIR_DOMAIN_START_AUTODESTROY) != 0; bool bypass_cache = (flags & VIR_DOMAIN_START_BYPASS_CACHE) != 0; bool force_boot = (flags & VIR_DOMAIN_START_FORCE_BOOT) != 0; + unsigned int start_flags = VIR_QEMU_PROCESS_START_COLD; + + start_flags |= start_paused ? VIR_QEMU_PROCESS_START_PAUSED : 0; + start_flags |= autodestroy ? VIR_QEMU_PROCESS_START_AUTODESROY : 0; /* * If there is a managed saved state restore it instead of starting @@ -4741,9 +4750,8 @@ qemuDomainObjStart(virConnectPtr conn, } } - ret = qemuProcessStart(conn, driver, vm, NULL, true, start_paused, - autodestroy, -1, NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + ret = qemuProcessStart(conn, driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, start_flags); virDomainAuditStart(vm, "booted", ret >= 0); if (ret >= 0) { virDomainEventPtr event = @@ -11027,9 +11035,10 @@ static int qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, if (config) virDomainObjAssignDef(vm, config, false); - rc = qemuProcessStart(snapshot->domain->conn, driver, vm, NULL, - false, true, false, -1, NULL, snap, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + rc = qemuProcessStart(snapshot->domain->conn, + driver, vm, NULL, -1, NULL, snap, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + VIR_QEMU_PROCESS_START_PAUSED); virDomainAuditStart(vm, "from-snapshot", rc >= 0); detail = VIR_DOMAIN_EVENT_STARTED_FROM_SNAPSHOT; event = virDomainEventNewFromObj(vm, @@ -11114,12 +11123,16 @@ static int qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, VIR_DOMAIN_SNAPSHOT_REVERT_PAUSED)) { /* Flush first event, now do transition 2 or 3 */ bool paused = (flags & VIR_DOMAIN_SNAPSHOT_REVERT_PAUSED) != 0; + unsigned int start_flags = 0; + + start_flags |= paused ? VIR_QEMU_PROCESS_START_PAUSED : 0; if (event) qemuDomainEventQueue(driver, event); - rc = qemuProcessStart(snapshot->domain->conn, driver, vm, NULL, - false, paused, false, -1, NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + rc = qemuProcessStart(snapshot->domain->conn, + driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + start_flags); virDomainAuditStart(vm, "from-snapshot", rc >= 0); if (rc < 0) { if (!vm->persistent) { diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index dc4d616..fc83805 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1303,9 +1303,10 @@ qemuMigrationPrepareAny(struct qemud_driver *driver, /* Start the QEMU daemon, with the same command-line arguments plus * -incoming $migrateFrom */ - if (qemuProcessStart(dconn, driver, vm, migrateFrom, false, true, - true, dataFD[0], NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START) < 0) { + if (qemuProcessStart(dconn, driver, vm, migrateFrom, dataFD[0], NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START, + VIR_QEMU_PROCESS_START_PAUSED | + VIR_QEMU_PROCESS_START_AUTODESROY) < 0) { virDomainAuditStart(vm, "migrated", false); /* Note that we don't set an error here because qemuProcessStart * should have already done that. diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 692fc32..ecbf039 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -3271,13 +3271,11 @@ int qemuProcessStart(virConnectPtr conn, struct qemud_driver *driver, virDomainObjPtr vm, const char *migrateFrom, - bool cold_boot, - bool start_paused, - bool autodestroy, int stdin_fd, const char *stdin_path, virDomainSnapshotObjPtr snapshot, - enum virNetDevVPortProfileOp vmop) + enum virNetDevVPortProfileOp vmop, + unsigned int flags) { int ret; off_t pos = -1; @@ -3290,6 +3288,12 @@ int qemuProcessStart(virConnectPtr conn, unsigned long cur_balloon; int i; + /* Okay, these are just internal flags, + * but doesn't hurt to check */ + virCheckFlags(VIR_QEMU_PROCESS_START_COLD | + VIR_QEMU_PROCESS_START_PAUSED | + VIR_QEMU_PROCESS_START_AUTODESROY, -1); + hookData.conn = conn; hookData.vm = vm; hookData.driver = driver; @@ -3434,7 +3438,8 @@ int qemuProcessStart(virConnectPtr conn, goto cleanup; VIR_DEBUG("Checking for CDROM and floppy presence"); - if (qemuDomainCheckDiskPresence(driver, vm, cold_boot) < 0) + if (qemuDomainCheckDiskPresence(driver, vm, + flags & VIR_QEMU_PROCESS_START_COLD) < 0) goto cleanup; VIR_DEBUG("Setting up domain cgroup (if required)"); @@ -3633,7 +3638,7 @@ int qemuProcessStart(virConnectPtr conn, VIR_DEBUG("Handshake complete, child running"); if (migrateFrom) - start_paused = true; + flags |= VIR_QEMU_PROCESS_START_PAUSED; if (ret == -1) /* The VM failed to start; tear filters before taps */ virDomainConfVMNWFilterTeardown(vm); @@ -3708,7 +3713,7 @@ int qemuProcessStart(virConnectPtr conn, } qemuDomainObjExitMonitorWithDriver(driver, vm); - if (!start_paused) { + if (!(flags & VIR_QEMU_PROCESS_START_PAUSED)) { VIR_DEBUG("Starting domain CPUs"); /* Allow the CPUS to start executing */ if (qemuProcessStartCPUs(driver, vm, conn, @@ -3726,7 +3731,7 @@ int qemuProcessStart(virConnectPtr conn, VIR_DOMAIN_PAUSED_USER); } - if (autodestroy && + if (flags & VIR_QEMU_PROCESS_START_AUTODESROY && qemuProcessAutoDestroyAdd(driver, vm, conn) < 0) goto cleanup; diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index 5cb5ffc..2d75b005 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -44,17 +44,21 @@ void qemuProcessReconnectAll(virConnectPtr conn, struct qemud_driver *driver); int qemuProcessAssignPCIAddresses(virDomainDefPtr def); +typedef enum { + VIR_QEMU_PROCESS_START_COLD = 1 << 0, + VIR_QEMU_PROCESS_START_PAUSED = 1 << 1, + VIR_QEMU_PROCESS_START_AUTODESROY = 1 << 2, +} qemuProcessStartFlags; + int qemuProcessStart(virConnectPtr conn, struct qemud_driver *driver, virDomainObjPtr vm, const char *migrateFrom, - bool cold_boot, - bool start_paused, - bool autodestroy, int stdin_fd, const char *stdin_path, virDomainSnapshotObjPtr snapshot, - enum virNetDevVPortProfileOp vmop); + enum virNetDevVPortProfileOp vmop, + unsigned int flags); void qemuProcessStop(struct qemud_driver *driver, virDomainObjPtr vm, -- 1.7.8.5

2 2

[libvirt] [libvirt-test-API PATCH 0/2] Documentation update
by Martin Kletzander 16 Apr '12

16 Apr '12

I updated part of the documentation that was changed in last few weeks. Screenshot test was updated again a little bit to suit the needs of a documentation and replaced the source codes of previous two tests mentioned in the documentation as it describe just what is needed and has no overhead on the reader. I'm not sure how the numbering goes in this case, but it can be updated later or I can send a v2 in case it needs to be fixed right away. Martin Kletzander (2): domain:screenshot: fixes and cleanup Documentation: Update to current information .../en-US/Understanding_libvirt-test-API.xml | 6 - .../en-US/Using_libvirt-test-API.xml | 26 +- .../en-US/Writing_a_test_case.xml | 325 ++++++++------------ .../libvirt-test-API_Guide/en-US/extras/log.txt | 53 ++-- repos/domain/screenshot.py | 45 ++- 5 files changed, 199 insertions(+), 256 deletions(-) -- 1.7.8.5

1 2

[libvirt] [ANNOUNCE] Perl binding Sys-Virt release 0.9.11
by Daniel P. Berrange 16 Apr '12

16 Apr '12

I am pleased to announce that release 0.9.11 of Sys-Virt, the libvirt Perl API binding is now available for download: http://search.cpan.org/CPAN/authors/id/D/DA/DANBERR/Sys-Virt-0.9.11.tar.gz Changed in this release: - Add all new APIs in libvirt 0.9.11 - Add test case to validate API coverage - Fix misc POD docs bugs - Fix reference handling in block stats - Add handling of VIR_TYPED_PARAM_STRING Further information, including online API documentation & previous release archives, is available from the CPAN home page http://search.cpan.org/dist/Sys-Virt/ Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

1 0

[libvirt] [PATCH] docs: fix path to openvz network configuration file
by Guido Günther 16 Apr '12

16 Apr '12

It's vznet.conf not vznetctl.conf, see e.g.: http://git.openvz.org/?p=vzctl;a=blob;f=bin/vznetcfg.in;h=e91f5c4a0744c1ea1… --- docs/drvopenvz.html.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/drvopenvz.html.in b/docs/drvopenvz.html.in index bb437d3..a8f9f28 100644 --- a/docs/drvopenvz.html.in +++ b/docs/drvopenvz.html.in @@ -67,7 +67,7 @@ openvz+ssh://root@example.com/system (remote access, SSH tunnelled) script must be created manually by the host OS administrator. The simplest way is to just download the latest version of this script from a newer OpenVZ release, or upstream source repository. Then - a generic configuration file <code>/etc/vz/vznetctl.conf</code> + a generic configuration file <code>/etc/vz/vznet.conf</code> must be created containing </p> -- 1.7.10

3 3

[libvirt] [PATCH RESEND] qemuProcessStart: Switch to flags instead of bunch booleans
by Michal Privoznik 16 Apr '12

16 Apr '12

Currently, we have 3 boolean arguments we have to pass to qemuProcessStart(). As libvirt grows it is harder and harder to remember them and their position. Therefore we should switch to flags instead. --- This is just rebased version of: http://www.redhat.com/archives/libvir-list/2012-March/msg00331.html src/qemu/qemu_driver.c | 45 +++++++++++++++++++++++++++++---------------- src/qemu/qemu_migration.c | 7 ++++--- src/qemu/qemu_process.c | 21 +++++++++++++-------- src/qemu/qemu_process.h | 12 ++++++++---- 4 files changed, 54 insertions(+), 31 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 65ed290..436ef37 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1351,10 +1351,16 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, const char *xml, virDomainPtr dom = NULL; virDomainEventPtr event = NULL; virDomainEventPtr event2 = NULL; + unsigned int start_flags = VIR_QEMU_PROCESS_START_COLD; virCheckFlags(VIR_DOMAIN_START_PAUSED | VIR_DOMAIN_START_AUTODESTROY, NULL); + if (flags & VIR_DOMAIN_START_PAUSED) + start_flags |= VIR_QEMU_PROCESS_START_PAUSED; + if (flags & VIR_DOMAIN_START_AUTODESTROY) + start_flags |= VIR_QEMU_PROCESS_START_AUTODESROY; + qemuDriverLock(driver); if (!(def = virDomainDefParseString(driver->caps, xml, QEMU_EXPECTED_VIRT_TYPES, @@ -1383,10 +1389,9 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, const char *xml, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; /* XXXX free the 'vm' we created ? */ - if (qemuProcessStart(conn, driver, vm, NULL, true, - (flags & VIR_DOMAIN_START_PAUSED) != 0, - (flags & VIR_DOMAIN_START_AUTODESTROY) != 0, - -1, NULL, NULL, VIR_NETDEV_VPORT_PROFILE_OP_CREATE) < 0) { + if (qemuProcessStart(conn, driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + start_flags) < 0) { virDomainAuditStart(vm, "booted", false); if (qemuDomainObjEndJob(driver, vm) > 0) qemuDomainRemoveInactive(driver, vm); @@ -4138,9 +4143,9 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, } /* Set the migration source and start it up. */ - ret = qemuProcessStart(conn, driver, vm, "stdio", false, true, - false, *fd, path, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_RESTORE); + ret = qemuProcessStart(conn, driver, vm, "stdio", *fd, path, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_RESTORE, + VIR_QEMU_PROCESS_START_PAUSED); if (intermediatefd != -1) { if (ret < 0) { @@ -4710,6 +4715,10 @@ qemuDomainObjStart(virConnectPtr conn, bool autodestroy = (flags & VIR_DOMAIN_START_AUTODESTROY) != 0; bool bypass_cache = (flags & VIR_DOMAIN_START_BYPASS_CACHE) != 0; bool force_boot = (flags & VIR_DOMAIN_START_FORCE_BOOT) != 0; + unsigned int start_flags = VIR_QEMU_PROCESS_START_COLD; + + start_flags |= start_paused ? VIR_QEMU_PROCESS_START_PAUSED : 0; + start_flags |= autodestroy ? VIR_QEMU_PROCESS_START_AUTODESROY : 0; /* * If there is a managed saved state restore it instead of starting @@ -4741,9 +4750,8 @@ qemuDomainObjStart(virConnectPtr conn, } } - ret = qemuProcessStart(conn, driver, vm, NULL, true, start_paused, - autodestroy, -1, NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + ret = qemuProcessStart(conn, driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, start_flags); virDomainAuditStart(vm, "booted", ret >= 0); if (ret >= 0) { virDomainEventPtr event = @@ -11027,9 +11035,10 @@ static int qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, if (config) virDomainObjAssignDef(vm, config, false); - rc = qemuProcessStart(snapshot->domain->conn, driver, vm, NULL, - false, true, false, -1, NULL, snap, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + rc = qemuProcessStart(snapshot->domain->conn, + driver, vm, NULL, -1, NULL, snap, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + VIR_QEMU_PROCESS_START_PAUSED); virDomainAuditStart(vm, "from-snapshot", rc >= 0); detail = VIR_DOMAIN_EVENT_STARTED_FROM_SNAPSHOT; event = virDomainEventNewFromObj(vm, @@ -11114,12 +11123,16 @@ static int qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, VIR_DOMAIN_SNAPSHOT_REVERT_PAUSED)) { /* Flush first event, now do transition 2 or 3 */ bool paused = (flags & VIR_DOMAIN_SNAPSHOT_REVERT_PAUSED) != 0; + unsigned int start_flags = 0; + + start_flags |= paused ? VIR_QEMU_PROCESS_START_PAUSED : 0; if (event) qemuDomainEventQueue(driver, event); - rc = qemuProcessStart(snapshot->domain->conn, driver, vm, NULL, - false, paused, false, -1, NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_CREATE); + rc = qemuProcessStart(snapshot->domain->conn, + driver, vm, NULL, -1, NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_CREATE, + start_flags); virDomainAuditStart(vm, "from-snapshot", rc >= 0); if (rc < 0) { if (!vm->persistent) { diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index dc4d616..fc83805 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1303,9 +1303,10 @@ qemuMigrationPrepareAny(struct qemud_driver *driver, /* Start the QEMU daemon, with the same command-line arguments plus * -incoming $migrateFrom */ - if (qemuProcessStart(dconn, driver, vm, migrateFrom, false, true, - true, dataFD[0], NULL, NULL, - VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START) < 0) { + if (qemuProcessStart(dconn, driver, vm, migrateFrom, dataFD[0], NULL, NULL, + VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START, + VIR_QEMU_PROCESS_START_PAUSED | + VIR_QEMU_PROCESS_START_AUTODESROY) < 0) { virDomainAuditStart(vm, "migrated", false); /* Note that we don't set an error here because qemuProcessStart * should have already done that. diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 692fc32..d18d721 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -3271,13 +3271,11 @@ int qemuProcessStart(virConnectPtr conn, struct qemud_driver *driver, virDomainObjPtr vm, const char *migrateFrom, - bool cold_boot, - bool start_paused, - bool autodestroy, int stdin_fd, const char *stdin_path, virDomainSnapshotObjPtr snapshot, - enum virNetDevVPortProfileOp vmop) + enum virNetDevVPortProfileOp vmop, + unsigned int flags) { int ret; off_t pos = -1; @@ -3290,6 +3288,12 @@ int qemuProcessStart(virConnectPtr conn, unsigned long cur_balloon; int i; + /* Okay, these are just internal flags, + * but doesn't hurt to check */ + virCheckFlags(VIR_QEMU_PROCESS_START_COLD | + VIR_QEMU_PROCESS_START_PAUSED | + VIR_QEMU_PROCESS_START_AUTODESROY, -1); + hookData.conn = conn; hookData.vm = vm; hookData.driver = driver; @@ -3434,7 +3438,8 @@ int qemuProcessStart(virConnectPtr conn, goto cleanup; VIR_DEBUG("Checking for CDROM and floppy presence"); - if (qemuDomainCheckDiskPresence(driver, vm, cold_boot) < 0) + if (qemuDomainCheckDiskPresence(driver, vm, + flags & VIR_QEMU_PROCESS_START_COLD) < 0) goto cleanup; VIR_DEBUG("Setting up domain cgroup (if required)"); @@ -3633,7 +3638,7 @@ int qemuProcessStart(virConnectPtr conn, VIR_DEBUG("Handshake complete, child running"); if (migrateFrom) - start_paused = true; + flags |= VIR_QEMU_PROCESS_START_PAUSED; if (ret == -1) /* The VM failed to start; tear filters before taps */ virDomainConfVMNWFilterTeardown(vm); @@ -3708,7 +3713,7 @@ int qemuProcessStart(virConnectPtr conn, } qemuDomainObjExitMonitorWithDriver(driver, vm); - if (!start_paused) { + if (flags & ~VIR_QEMU_PROCESS_START_PAUSED) { VIR_DEBUG("Starting domain CPUs"); /* Allow the CPUS to start executing */ if (qemuProcessStartCPUs(driver, vm, conn, @@ -3726,7 +3731,7 @@ int qemuProcessStart(virConnectPtr conn, VIR_DOMAIN_PAUSED_USER); } - if (autodestroy && + if (flags & VIR_QEMU_PROCESS_START_AUTODESROY && qemuProcessAutoDestroyAdd(driver, vm, conn) < 0) goto cleanup; diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index 5cb5ffc..2d75b005 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -44,17 +44,21 @@ void qemuProcessReconnectAll(virConnectPtr conn, struct qemud_driver *driver); int qemuProcessAssignPCIAddresses(virDomainDefPtr def); +typedef enum { + VIR_QEMU_PROCESS_START_COLD = 1 << 0, + VIR_QEMU_PROCESS_START_PAUSED = 1 << 1, + VIR_QEMU_PROCESS_START_AUTODESROY = 1 << 2, +} qemuProcessStartFlags; + int qemuProcessStart(virConnectPtr conn, struct qemud_driver *driver, virDomainObjPtr vm, const char *migrateFrom, - bool cold_boot, - bool start_paused, - bool autodestroy, int stdin_fd, const char *stdin_path, virDomainSnapshotObjPtr snapshot, - enum virNetDevVPortProfileOp vmop); + enum virNetDevVPortProfileOp vmop, + unsigned int flags); void qemuProcessStop(struct qemud_driver *driver, virDomainObjPtr vm, -- 1.7.8.5

2 1

[libvirt] [PATCH] storage: lvm: use correct lv* command parameters
by Cole Robinson 16 Apr '12

16 Apr '12

lvcreate want's the parent pool's name, not the pool path lvchange and lvremove want lv specified as $vgname/$lvname This largely worked before because these commands strip off a starting /dev. But https://bugzilla.redhat.com/show_bug.cgi?id=714986 is from a user using a 'nested VG' that was having problems. I couldn't find any info on nested LVM and the reporter never responded, but I reproduced with XML that specified a valid source name, and set target path to a symlink. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/storage/storage_backend_logical.c | 21 +++++++++++---------- 1 files changed, 11 insertions(+), 10 deletions(-) diff --git a/src/storage/storage_backend_logical.c b/src/storage/storage_backend_logical.c index 6a235f6..9a91dd9 100644 --- a/src/storage/storage_backend_logical.c +++ b/src/storage/storage_backend_logical.c @@ -672,7 +672,7 @@ virStorageBackendLogicalCreateVol(virConnectPtr conn, char size[100]; const char *cmdargvnew[] = { LVCREATE, "--name", vol->name, "-L", size, - pool->def->target.path, NULL + pool->def->source.name, NULL }; const char *cmdargvsnap[] = { LVCREATE, "--name", vol->name, "-L", size, @@ -778,23 +778,23 @@ virStorageBackendLogicalDeleteVol(virConnectPtr conn ATTRIBUTE_UNUSED, unsigned int flags) { int ret = -1; + char *volpath = NULL; virCommandPtr lvchange_cmd = NULL; virCommandPtr lvremove_cmd = NULL; virCheckFlags(0, -1); - virFileWaitForDevices(); + if (virAsprintf(&volpath, "%s/%s", + pool->def->source.name, vol->name) < 0) { + virReportOOMError(); + goto cleanup; + } - lvchange_cmd = virCommandNewArgList(LVCHANGE, - "-aln", - vol->target.path, - NULL); + virFileWaitForDevices(); - lvremove_cmd = virCommandNewArgList(LVREMOVE, - "-f", - vol->target.path, - NULL); + lvchange_cmd = virCommandNewArgList(LVCHANGE, "-aln", volpath, NULL); + lvremove_cmd = virCommandNewArgList(LVREMOVE, "-f", volpath, NULL); if (virCommandRun(lvremove_cmd, NULL) < 0) { if (virCommandRun(lvchange_cmd, NULL) < 0) { @@ -807,6 +807,7 @@ virStorageBackendLogicalDeleteVol(virConnectPtr conn ATTRIBUTE_UNUSED, ret = 0; cleanup: + VIR_FREE(volpath); virCommandFree(lvchange_cmd); virCommandFree(lvremove_cmd); return ret; -- 1.7.7.6

2 4

[libvirt] [PATCH] domain:screenshot: Added cleanup function
by Martin Kletzander 16 Apr '12

16 Apr '12

Added cleanup function to the screeshot testcase. This makes use of the new "sharedmod" module. --- WARNING: don't push this before the patch with sharedmod is pushed in the repo, otherwise this will not work. Thanks. repos/domain/screenshot.py | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/repos/domain/screenshot.py b/repos/domain/screenshot.py index 3e727a7..5a12c4b 100644 --- a/repos/domain/screenshot.py +++ b/repos/domain/screenshot.py @@ -6,6 +6,7 @@ import os import mimetypes import libvirt +import sharedmod required_params = ('guestname', 'filename',) optional_params = ('screen',) @@ -36,6 +37,7 @@ def screenshot(params): logger.debug('Mimetype of the file is %s' % mime) ret = st.finish() + sharedmod.dict['screenshot_filename'] = filename finally: # Some error occurred, cleanup @@ -43,3 +45,7 @@ def screenshot(params): conn.close() return ret + +def cleanup(params): + if sharedmod.has_key('screenshot_filename'): + os.remove(sharedmod['screenshot_filename']) -- 1.7.8.5

2 7

[libvirt] numad: Convert node list to cpumap before affinity setting
by Osier Yang 16 Apr '12

16 Apr '12

Previous patches on numad made mistake with thinking numad will return CPUs list (numad document says CPUs list one place, and node list another place. :( Please see the patches' commit message for the details. Osier Yang (3) numad: Convert node list to cpumap before setting affinity numad: Ignore cpuset if placement is auto qemu: Avoid the memory allocation and freeing docs/formatdomain.html.in | 10 ++++---- src/conf/domain_conf.c | 28 ++++++++++++++------------ src/qemu/qemu_process.c | 47 +++++++++++++++++++++++--------------------- 3 files changed, 45 insertions(+), 40 deletions(-) Regards, Osier

2 7

[libvirt] [test-API PATCH 1/2] add a new option -t to print template of testcase config file
by Guannan Ren 16 Apr '12

16 Apr '12

If a given testcase has clean function, the clean flag will be used automatically # python libvirt-test-api.py -t repos/domain/attach_disk.py \ repos/storage/create_netfs_pool.py \ repos/domain/save.py output: domain:attach_disk guestname GUESTNAME guesttype GUESTTYPE imagename IMAGENAME imagesize IMAGESIZE hdmodel HDMODEL storage:create_netfs_pool poolname POOLNAME sourcename SOURCENAME sourcepath SOURCEPATH pooltype POOLTYPE [targetpath] TARGETPATH domain:save guestname GUESTNAME filepath FILEPATH clean --- libvirt-test-api.py | 49 +++++++++++++++++++++++++++++++++++++++++++++++-- proxy.py | 17 +++++++++++++++++ 2 files changed, 64 insertions(+), 2 deletions(-) diff --git a/libvirt-test-api.py b/libvirt-test-api.py index c171276..385b52d 100644 --- a/libvirt-test-api.py +++ b/libvirt-test-api.py @@ -35,6 +35,7 @@ def usage(): print "Usage: libvirt_test_api.py <OPTIONS> <ARGUS>" print "\noptions: -h, --help : Display usage information \ \n -c, --casefile: Specify configuration file \ + \n -t, --template: Print testcase config file template \ \n -f, --logxml: Specify log file with type xml \ \n -l, --log-level: 0 or 1 currently \ \n -d, --delete-log: Delete log items \ @@ -45,6 +46,7 @@ def usage(): print "example: \ \n python libvirt-test-api.py -l 0|1 -c TEST.CONF \ \n python libvirt-test-api.py -c TEST.CONF -f TEST.XML \ + \n python libvirt-test-api.py -t repos/domain/start.py ... \ \n python libvirt-test-api.py -m TESTONE.XML TESTTWO.XML \ \n python libvirt-test-api.py -d TEST.XML TESTRUNID TESTID \ \n python libvirt-test-api.py -d TEST.XML TESTRUNID \ @@ -215,6 +217,41 @@ class Main(object): return 1 return 0 + def print_casefile(self, testcases): + """print testcase file template""" + modcasename = [] + for case in testcases: + if not os.path.isfile(case) or not case.endswith('.py'): + print "testcase %s couldn't be recognized" % case + return 1 + + paths = case.split('/') + modcasename.append(paths[1] + ':' + paths[2][:-3]) + + proxy_obj = proxy.Proxy(modcasename) + case_params = proxy_obj.get_params_variables() + + string = "# the file is generated automatically, please\n" \ + "# make some modifications before the use of it\n" \ + "# params in [] are optional to its testcase\n" + for key in modcasename: + string += "%s\n" % key + required_params, optional_params = case_params[key] + for p in required_params: + string += " " * 4 + p + "\n" + string += " " * 8 + p.upper() + "\n" + for p in optional_params: + string += " " * 4 + "[" + p + "]\n" + string += " " * 8 + p.upper() + "\n" + + if proxy_obj.has_clean_function(key): + string += "clean\n" + + string += "\n" + + print string + return 0 + def remove_log(self, testrunid, testid = None): """ to remove log item in the log xmlfile """ log_xml_parser = LogXMLParser(self.logxml) @@ -274,8 +311,8 @@ if __name__ == "__main__": loglevel = 0 try: - opts, args = getopt.getopt(sys.argv[1:], "hc:l:dmr", - ["help", "casefile=", "logxml=", + opts, args = getopt.getopt(sys.argv[1:], "hc:tl:dmr", + ["help", "casefile=", "template", "logxml=", "delete-log=", "merge=", "rerun="]) except getopt.GetoptError, err: print str(err) @@ -288,6 +325,14 @@ if __name__ == "__main__": sys.exit(0) if o == "-c" or o == "--casefile": casefile = v + if o == "-t" or o == "--template": + if len(args) <= 0: + usage() + sys.exit(1) + main = Main('', '', '', '') + if main.print_casefile(args): + sys.exit(1) + sys.exit(0) if o == "-f" or o == "--logxml": logxml = v if o == "-l" or o == "--log-level": diff --git a/proxy.py b/proxy.py index fdbffd9..bc82a84 100644 --- a/proxy.py +++ b/proxy.py @@ -120,6 +120,23 @@ class Proxy(object): ("required_params or optional_params not found in %s" % testcase_name) return case_params + def has_clean_function(self, testcase_name): + """ Return true if the testcase have clean function + """ + if testcase_name not in self.testcases_names: + return False + + elements = testcase_name.split(':') + casename = elements[1] + func = casename + '_clean' + + casemod_ref = self.testcase_ref_dict[testcase_name] + var_func_names = dir(casemod_ref) + + if func in var_func_names: + return True + return False + def get_call_dict(self, module, casename, func = None): """ Return testing function reference dictionary """ case_abs_path = '%s.%s.%s' % ('repos', module, casename) -- 1.7.7.5

2 3

[libvirt] [test-API PATCH 0/3] make clean function optional
by Guannan Ren 16 Apr '12

16 Apr '12

The set of patches is to make the writing of clean function optional. If a testcase which doesn't dirty the testing environment, then the TESTCASE_clean function could be optional or omitted. Because loggers in python are static objects managed by the module itself. When you create one, it won't be removed until the shell quit. In test-API there are two loggers which are configured with two handlers for each. we need to find a way to destruct these two loggers in order no to affect other application. The idea is to clear the handlers of each logger in destructor of CaseLog and EnvLog. def __del__(self): self.logger.handlers = [] rename some unclear function and variables. for example: repos/domain/start.py domain -> mod start.py -> case start() -> func

2 9

[libvirt] [test-API PATCH 0/5] add TESTCASE_check optional function support
by Guannan Ren 16 Apr '12

16 Apr '12

In some cases, we need to check if the testing environment is satisfied to run a certain testcase. The testcase only will be executed If some specific hardware is present on box. The patches add a optional check function support. For example: a testcase named testa.py with testa_check() defined in this file. The framework will run test_check(params) first, return 0 means pass, 1 means failure. If the check function failed. the testa(params) will not be run. "Skip" is marked in output. # testa.py required_params = () optional_params = ('options') def testa_check(params): logger = params['logger'] logger.info("I am from testa_check") return 1 def testa(params): logger = params['logger'] logger.info("I am from testa") return 1 The Output: Checking Testing Environment... Linux localhost.localdomain 3.2.5-3.fc16.x86_64 #1 SMP Thu Feb 9 01:24:38 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux Virsh command line tool of libvirt: 0.9.10 libvirtd (libvirt) 0.9.10 Default URI: qemu:///system QEMU emulator version 0.15.0 (qemu-kvm-0.15.0), Copyright (c) 2003-2008 Fabrice Bellard Start Testing: Case Count: 1 Log File: log/20120416142755/libvirt_test001 test:testa 14:27:56|INFO |I am from testa_check 14:27:56|INFO |Failed to meet testing requirement Result: Skip Summary: Total:1 [Pass:0 Fail:0 Skip:1] The testcase.conf file: test:testa options value1

1 5

[libvirt] [PATCH] Web placeholder for pending patches on the web site
by Daniel Veillard 16 Apr '12

16 Apr '12

This is based on recent developments on patch checker and the goal is to keep a list of pending patches needing review on the project web site. The page template in git just hold a pointer to the web page. diff --git a/docs/pending.html.in b/docs/pending.html.in new file mode 100644 index 0000000..dfbe647 --- /dev/null +++ b/docs/pending.html.in @@ -0,0 +1,13 @@ +<html> + <head> + <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /> + <title>Pending patches needing review</title> + </head> + <body> + <h1>Pending patches needing review</h1> + <p> A list of pending patches needing review upstream is available + on <a href="http://libvirt.org/pending.html">the project pending + patches page</a>.</p> + </body> +</html> + diff --git a/docs/sitemap.html.in b/docs/sitemap.html.in index 1de2b20..8f58d46 100644 --- a/docs/sitemap.html.in +++ b/docs/sitemap.html.in @@ -332,6 +332,10 @@ <a href="todo.html">Todo list</a> <span>Main feature request list</span> </li> + <li> + <a href="pending.html">Pending patches</a> + <span>Pending patches awaiting reviews and integration</span> + </li> </ul> </li> <li> -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

3 5

[libvirt] [PATCH] domain:screenshot: fixed screen parameter
by Martin Kletzander 16 Apr '12

16 Apr '12

The screen parameter must be an integer. I also chnaged it to variable parameter with the reasonable default. --- repos/domain/screenshot.py | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/repos/domain/screenshot.py b/repos/domain/screenshot.py index d73e980..3e727a7 100644 --- a/repos/domain/screenshot.py +++ b/repos/domain/screenshot.py @@ -7,8 +7,8 @@ import mimetypes import libvirt -required_params = ('guestname', 'screen', 'filename',) -optional_params = () +required_params = ('guestname', 'filename',) +optional_params = ('screen',) def saver(stream, data, file_): return file_.write(data) @@ -24,7 +24,8 @@ def screenshot(params): dom = conn.lookupByName(params['guestname']) st = conn.newStream(0) - mime = dom.screenshot(st, params['screen'], 0) + screen = params.get('screen', 0) + mime = dom.screenshot(st, int('screen'), 0) ext = mimetypes.guess_extension(mime) or '.ppm' filename = params['filename'] + ext -- 1.7.8.5

2 2