- Devel - libvirt List Archives

[libvirt] [PATCH] Correct two memory leaks triggered by udev events
by Nigel Jones 09 Jun '10

09 Jun '10

Hi List, Please find below a patch that should correct two memory leaks within the udev device handling code. The issue is triggered by 'add' udev calls and will slowly cause libvirtd to consume the majority of system memory. The first part of the patch deals with udevAddOneDevice() and frees the memory allocated to 'def' if the function would return -1 (an error condition) due to the inability to look up udev properties (for example, if the udev device is already removed). The second part of the patch deals with a remaining 8kB/cycle memory leak, in which the udev device reference isn't released at the end of running udevEventHandleCallback(). I'm happy to answer any questions about the patch, there is also a bit more background at https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/571093 --- >From 6c8183e83fbfeb031b16cf9ae2d41b16e3145378 Mon Sep 17 00:00:00 2001 From: Nigel Jones <dev(a)nigelj.com> Date: Mon, 24 May 2010 15:05:53 +0000 Subject: [PATCH] Patch 2 memory leaks. 1. Ensure that memory is free'd from udevAddOneDevice() if the return value will be non-zero 2. Release udev device reference in udevEventHandleCallback() similar to the release of the reference in udevProcessDeviceListEntry() --- src/node_device/node_device_udev.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index a1ced87..4d0effa 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -1296,6 +1296,9 @@ static int udevAddOneDevice(struct udev_device *device) ret = 0; out: + if (ret != 0) { + virNodeDeviceDefFree(def); /* Free assigned memory to prevent leaks */ + } return ret; } @@ -1426,6 +1429,7 @@ static void udevEventHandleCallback(int watch ATTRIBUTE_UNUSED, } out: + udev_device_unref(device); return; } -- 1.7.0.4

5 8

[libvirt] [PATCH] virsh: add snapshot backing store support to vol-create-as
by Justin Clift 09 Jun '10

09 Jun '10

This patch adds two new parameters to the vol-create-as command: --snapshot-backing-vol <volume-name-or-key-or-path> --snapshot-backing-vol-format <format-of-backing-vol> virsh # vol-create-as guest_images_lvm snapvol1 5G --snapshot-backing-vol \ rhel6vm1lun1 Vol snapvol1 created virsh # vol-create-as image_dir qcow2snap2 5G --format qcow2 \ --snapshot-backing-vol imagevol1.qcow2 \ --snapshot-backing-vol-format qcow2 Vol qcow2snap2 created Additionally, the virsh man page update fixes incorrect snapshot parameters that were included in my prior bulk volume command patch. --- tools/virsh.c | 61 ++++++++++++++++++++++++++++++++++++++++++++++++++++++- tools/virsh.pod | 10 ++++---- 2 files changed, 65 insertions(+), 6 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 1e00114..497c7d4 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -5274,6 +5274,8 @@ static const vshCmdOptDef opts_vol_create_as[] = { {"capacity", VSH_OT_DATA, VSH_OFLAG_REQ, N_("size of the vol with optional k,M,G,T suffix")}, {"allocation", VSH_OT_STRING, 0, N_("initial allocation size with optional k,M,G,T suffix")}, {"format", VSH_OT_STRING, 0, N_("file format type raw,bochs,qcow,qcow2,vmdk")}, + {"snapshot-backing-vol", VSH_OT_STRING, 0, N_("the backing volume if taking a snapshot")}, + {"snapshot-backing-vol-format", VSH_OT_STRING, 0, N_("format of backing volume if taking a snapshot")}, {NULL, 0, 0, NULL} }; @@ -5313,6 +5315,7 @@ cmdVolCreateAs(vshControl *ctl, const vshCmd *cmd) int found; char *xml; char *name, *capacityStr, *allocationStr, *format; + char *snapshotStrVol, *snapshotStrFormat; unsigned long long capacity, allocation = 0; virBuffer buf = VIR_BUFFER_INITIALIZER; @@ -5339,6 +5342,8 @@ cmdVolCreateAs(vshControl *ctl, const vshCmd *cmd) vshError(ctl, _("Malformed size %s"), allocationStr); format = vshCommandOptString(cmd, "format", &found); + snapshotStrVol = vshCommandOptString(cmd, "snapshot-backing-vol", &found); + snapshotStrFormat = vshCommandOptString(cmd, "snapshot-backing-vol-format", &found); virBufferAddLit(&buf, "<volume>\n"); virBufferVSprintf(&buf, " <name>%s</name>\n", name); @@ -5351,8 +5356,62 @@ cmdVolCreateAs(vshControl *ctl, const vshCmd *cmd) virBufferVSprintf(&buf, " <format type='%s'/>\n",format); virBufferAddLit(&buf, " </target>\n"); } - virBufferAddLit(&buf, "</volume>\n"); + /* Convert the snapshot parameters into backingStore XML */ + if (snapshotStrVol) { + /* Lookup snapshot backing volume. Try the snapshot-backing-vol + * parameter as a name */ + vshDebug(ctl, 5, "%s: Look up backing store volume '%s' as name\n", + cmd->def->name, snapshotStrVol); + virStorageVolPtr snapVol = virStorageVolLookupByName(pool, snapshotStrVol); + if (snapVol) + vshDebug(ctl, 5, "%s: Backing store volume found using '%s' as name\n", + cmd->def->name, snapshotStrVol); + + if (snapVol == NULL) { + /* Snapshot backing volume not found by name. Try the + * snapshot-backing-vol parameter as a key */ + vshDebug(ctl, 5, "%s: Look up backing store volume '%s' as key\n", + cmd->def->name, snapshotStrVol); + snapVol = virStorageVolLookupByKey(ctl->conn, snapshotStrVol); + if (snapVol) + vshDebug(ctl, 5, "%s: Backing store volume found using '%s' as key\n", + cmd->def->name, snapshotStrVol); + } + if (snapVol == NULL) { + /* Snapshot backing volume not found by key. Try the + * snapshot-backing-vol parameter as a path */ + vshDebug(ctl, 5, "%s: Look up backing store volume '%s' as path\n", + cmd->def->name, snapshotStrVol); + snapVol = virStorageVolLookupByPath(ctl->conn, snapshotStrVol); + if (snapVol) + vshDebug(ctl, 5, "%s: Backing store volume found using '%s' as path\n", + cmd->def->name, snapshotStrVol); + } + if (snapVol == NULL) { + vshError(ctl, _("failed to get vol '%s'"), snapshotStrVol); + return FALSE; + } + + char *snapshotStrVolPath; + if ((snapshotStrVolPath = virStorageVolGetPath(snapVol)) == NULL) { + virStorageVolFree(snapVol); + return FALSE; + } + + /* Create XML for the backing store */ + virBufferAddLit(&buf, " <backingStore>\n"); + virBufferVSprintf(&buf, " <path>%s</path>\n",snapshotStrVolPath); + if (snapshotStrFormat) + virBufferVSprintf(&buf, " <format type='%s'/>\n",snapshotStrFormat); + virBufferAddLit(&buf, " </backingStore>\n"); + + /* Cleanup snapshot allocations */ + VIR_FREE(snapshotStrVolPath); + virStorageVolFree(snapVol); + } + + virBufferAddLit(&buf, "</volume>\n"); if (virBufferError(&buf)) { vshPrint(ctl, "%s", _("Failed to allocate XML buffer")); diff --git a/tools/virsh.pod b/tools/virsh.pod index 3b0cf16..7039ad2 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -782,8 +782,8 @@ source volume is in. I<vol-name-or-key-or-path> is the name or key or path of the source volume. =item B<vol-create-as> I<pool-or-uuid> I<name> I<capacity> optional -I<--allocation> I<size> I<--format> I<string> I<--snapshot-source-vol> -I<vol-name-or-key-or-path> I<--snapshot-source-format> I<string> +I<--allocation> I<size> I<--format> I<string> I<--snapshot-backing-vol> +I<vol-name-or-key-or-path> I<--snapshot-backing-vol-format> I<string> Create a volume from a set of arguments. I<pool-or-uuid> is the name or UUID of the storage pool to create the volume @@ -795,10 +795,10 @@ I<--allocation> I<size> is the initial size to be allocated in the volume, with optional k, M, G, or T suffix. I<--format> I<string> is used in file based storage pools to specify the volume file format to use; raw, bochs, qcow, qcow2, vmdk. -I<--snapshot-source-vol> I<vol-name-or-key-or-path> is the source backing +I<--snapshot-backing-vol> I<vol-name-or-key-or-path> is the source backing volume to be used if taking a snapshot of an existing volume. -I<--snapshot-source-format> I<string> is the format of the snapshot backing volume; -raw, bochs, qcow, qcow2, vmdk. +I<--snapshot-backing-vol-format> I<string> is the format of the snapshot backing volume; +raw, bochs, qcow, qcow2, vmdk, host_device. =item B<vol-clone> [optional I<--pool> I<pool-or-uuid>] I<vol-name-or-key-or-path> I<name> -- 1.7.0.1

2 3

[libvirt] [PATCH] Enable probing of VPC disk format type
by Daniel P. Berrange 09 Jun '10

09 Jun '10

A look at the QEMU source revealed the missing bits of info about the VPC file format, so we can enable this now * src/util/storage_file.c: Enable VPC format, providing version and disk size offset fields --- src/util/storage_file.c | 6 ++---- 1 files changed, 2 insertions(+), 4 deletions(-) diff --git a/src/util/storage_file.c b/src/util/storage_file.c index b3ae905..6cc8d5f 100644 --- a/src/util/storage_file.c +++ b/src/util/storage_file.c @@ -135,11 +135,9 @@ static struct FileTypeInfo const fileTypeInfo[] = { LV_LITTLE_ENDIAN, 4, 1, 4+4+4, 8, 512, -1, vmdk4GetBackingStore }, /* Connectix / VirtualPC */ - /* XXX Untested { VIR_STORAGE_FILE_VPC, "conectix", NULL, - LV_BIG_ENDIAN, -1, 0, - -1, 0, 0, -1, NULL}, - */ + LV_BIG_ENDIAN, 12, 0x10000, + 8 + 4 + 4 + 8 + 4 + 4 + 2 + 2 + 4, 8, 1, -1, NULL}, }; static int -- 1.6.6.1

2 1

[libvirt] [PATCH] virsh: add --pool support to vol-key command
by Justin Clift 09 Jun '10

09 Jun '10

--- tools/virsh.c | 19 +++++++++---------- 1 files changed, 9 insertions(+), 10 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 2e8cbec..1e00114 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -5947,13 +5947,13 @@ cmdVolList(vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) * "vol-name" command */ static const vshCmdInfo info_vol_name[] = { - {"help", N_("convert a vol UUID to vol name")}, + {"help", N_("returns the volume name for a given volume key or path")}, {"desc", ""}, {NULL, NULL} }; static const vshCmdOptDef opts_vol_name[] = { - {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("vol key or path")}, + {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("volume key or path")}, {NULL, 0, 0, NULL} }; @@ -5975,18 +5975,18 @@ cmdVolName(vshControl *ctl, const vshCmd *cmd) } - /* * "vol-key" command */ static const vshCmdInfo info_vol_key[] = { - {"help", N_("convert a vol UUID to vol key")}, + {"help", N_("returns the volume key for a given volume name or path")}, {"desc", ""}, {NULL, NULL} }; static const vshCmdOptDef opts_vol_key[] = { - {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("vol uuid")}, + {"pool", VSH_OT_STRING, 0, N_("pool name or uuid")}, + {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("volume name or path")}, {NULL, 0, 0, NULL} }; @@ -5994,12 +5994,12 @@ static int cmdVolKey(vshControl *ctl, const vshCmd *cmd) { virStorageVolPtr vol; + char *name = NULL; if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; - if (!(vol = vshCommandOptVolBy(ctl, cmd, "vol", "pool", NULL, - VSH_BYUUID))) + if (!(vol = vshCommandOptVol(ctl, cmd, "vol", "pool", &name))) return FALSE; vshPrint(ctl, "%s\n", virStorageVolGetKey(vol)); @@ -6008,19 +6008,18 @@ cmdVolKey(vshControl *ctl, const vshCmd *cmd) } - /* * "vol-path" command */ static const vshCmdInfo info_vol_path[] = { - {"help", N_("convert a vol UUID to vol path")}, + {"help", N_("returns the volume path for a given volume name or key")}, {"desc", ""}, {NULL, NULL} }; static const vshCmdOptDef opts_vol_path[] = { {"pool", VSH_OT_STRING, 0, N_("pool name or uuid")}, - {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("vol name or key")}, + {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("volume name or key")}, {NULL, 0, 0, NULL} }; -- 1.7.0.1

1 0

[libvirt] [PATCH] virsh: remove Xen reference in header comment
by Justin Clift 08 Jun '10

08 Jun '10

--- tools/virsh.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 1279f41..e444501 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1,5 +1,5 @@ /* - * virsh.c: a Xen shell used to exercise the libvirt API + * virsh.c: a shell to exercise the libvirt API * * Copyright (C) 2005, 2007-2010 Red Hat, Inc. * -- 1.7.0.1

1 0

[libvirt] PATCH: Include port number with virtio serial devices
by Daniel P. Berrange 08 Jun '10

08 Jun '10

To ensure that the device addressing scheme is stable across hotplug/unplug, all virtio serial channels needs to have an associated port number in their address. This is then specified to QEMU using the nr=NNN parameter * src/conf/domain_conf.c, src/conf/domain_conf.h: Parsing for port number in vioserial address types. * src/qemu/qemu_conf.c: Set 'nr=NNN' parameter with virtio serial port number * tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.args, tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.xml: Expand data set to ensure coverage of port addressing --- src/conf/domain_conf.c | 30 ++++++++++++++++++-- src/conf/domain_conf.h | 1 + src/qemu/qemu_conf.c | 3 ++ .../qemuxml2argv-channel-virtio.args | 2 +- .../qemuxml2argv-channel-virtio.xml | 14 ++++++++- 5 files changed, 45 insertions(+), 5 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 312a6c0..a9b01d5 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -1055,9 +1055,10 @@ static int virDomainDeviceInfoFormat(virBufferPtr buf, break; case VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_SERIAL: - virBufferVSprintf(buf, " controller='%d' bus='%d'", + virBufferVSprintf(buf, " controller='%d' bus='%d' port='%d'", info->addr.vioserial.controller, - info->addr.vioserial.bus); + info->addr.vioserial.bus, + info->addr.vioserial.port); break; default: @@ -1190,13 +1191,14 @@ virDomainDeviceVirtioSerialAddressParseXML( virDomainDeviceVirtioSerialAddressPtr addr ) { - char *controller, *bus; + char *controller, *bus, *port; int ret = -1; memset(addr, 0, sizeof(*addr)); controller = virXMLPropString(node, "controller"); bus = virXMLPropString(node, "bus"); + port = virXMLPropString(node, "port"); if (controller && virStrToLong_ui(controller, NULL, 10, &addr->controller) < 0) { @@ -1212,6 +1214,13 @@ virDomainDeviceVirtioSerialAddressParseXML( goto cleanup; } + if (port && + virStrToLong_ui(port, NULL, 10, &addr->port) < 0) { + virDomainReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Cannot parse <address> 'port' attribute")); + goto cleanup; + } + if (!virDomainDeviceVirtioSerialAddressIsValid(addr)) { virDomainReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Insufficient specification for " @@ -4375,6 +4384,21 @@ static virDomainDefPtr virDomainDefParseXML(virCapsPtr caps, goto error; def->channels[def->nchannels++] = chr; + + if (chr->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_SERIAL && + chr->info.addr.vioserial.port == 0) { + int maxport = -1; + int j; + for (j = 0 ; j < i ; j++) { + virDomainChrDefPtr thischr = def->channels[j]; + if (thischr->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_SERIAL && + thischr->info.addr.vioserial.controller == chr->info.addr.vioserial.controller && + thischr->info.addr.vioserial.bus == chr->info.addr.vioserial.bus && + (int)thischr->info.addr.vioserial.port > maxport) + maxport = thischr->info.addr.vioserial.port; + } + chr->info.addr.vioserial.port = maxport + 1; + } } VIR_FREE(nodes); diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index f87f6c8..f83de83 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -98,6 +98,7 @@ typedef virDomainDeviceVirtioSerialAddress *virDomainDeviceVirtioSerialAddressPt struct _virDomainDeviceVirtioSerialAddress { unsigned int controller; unsigned int bus; + unsigned int port; }; typedef struct _virDomainDeviceInfo virDomainDeviceInfo; diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 511b9de..5521dbe 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -3312,6 +3312,9 @@ qemuBuildVirtioSerialPortDevStr(virDomainChrDefPtr dev) ",bus=" QEMU_VIRTIO_SERIAL_PREFIX "%d.%d", dev->info.addr.vioserial.controller, dev->info.addr.vioserial.bus); + virBufferVSprintf(&buf, + ",nr=%d", + dev->info.addr.vioserial.port); } virBufferVSprintf(&buf, ",chardev=%s", dev->info.alias); diff --git a/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.args b/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.args index 4097065..8e5fbe2 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.args @@ -1 +1 @@ -LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu -S -M pc -m 214 -smp 1 -nographic -nodefaults -monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -device virtio-serial-pci,id=virtio-serial0,max_ports=16,vectors=4,bus=pci.0,addr=0x4 -device virtio-serial-pci,id=virtio-serial1,bus=pci.0,addr=0xa -hda /dev/HostVG/QEMUGuest1 -chardev pty,id=channel0 -device virtserialport,chardev=channel0,name=org.linux-kvm.port.0 -chardev pty,id=channel1 -device virtserialport,bus=virtio-serial1.0,chardev=channel1,name=org.linux-kvm.port.1 -usb -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu -S -M pc -m 214 -smp 1 -nographic -nodefaults -monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -device virtio-serial-pci,id=virtio-serial0,max_ports=16,vectors=4,bus=pci.0,addr=0x4 -device virtio-serial-pci,id=virtio-serial1,bus=pci.0,addr=0xa -hda /dev/HostVG/QEMUGuest1 -chardev pty,id=channel0 -device virtserialport,chardev=channel0,name=org.linux-kvm.port.0 -chardev pty,id=channel1 -device virtserialport,bus=virtio-serial1.0,nr=0,chardev=channel1,name=org.linux-kvm.port.foo -chardev pty,id=channel2 -device virtserialport,bus=virtio-serial1.0,nr=3,chardev=channel2,name=org.linux-kvm.port.bar -chardev pty,id=channel3 -device virtserialport,bus=virtio-serial0.0,nr=0,chardev=channel3,name=org.linux-kvm.port.wizz -chardev pty,id=channel4 -device virtserialport,bus=virtio-serial1.0,nr=4,chardev=channel4,name=org.linux-kvm.port.ooh -usb -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.xml b/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.xml index 6c3317b..04a3e1c 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-channel-virtio.xml @@ -28,7 +28,19 @@ <target type='virtio' name='org.linux-kvm.port.0'/> </channel> <channel type='pty'> - <target type='virtio' name='org.linux-kvm.port.1'/> + <target type='virtio' name='org.linux-kvm.port.foo'/> + <address type='virtio-serial' controller='1' bus='0'/> + </channel> + <channel type='pty'> + <target type='virtio' name='org.linux-kvm.port.bar'/> + <address type='virtio-serial' controller='1' bus='0' port='3'/> + </channel> + <channel type='pty'> + <target type='virtio' name='org.linux-kvm.port.wizz'/> + <address type='virtio-serial' controller='0' bus='0'/> + </channel> + <channel type='pty'> + <target type='virtio' name='org.linux-kvm.port.ooh'/> <address type='virtio-serial' controller='1' bus='0'/> </channel> </devices> -- 1.6.6.1 -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://deltacloud.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 3

[libvirt] [PATCH] Adds the missing vol-pool command to virsh.
by Justin Clift 08 Jun '10

08 Jun '10

Hi all, This patch adds a "vol-pool" command to virsh, to round out the conversion functions for vols in virsh. At the moment, if we start with a volume name and a pool id, we're all good. We can convert from that pair to either a volume key or volume path. But, if we're given just a volume key or path we have no good way to look up the storage pool for it. Kind of makes it a one way association. :/ This patch fixes that, making it possible to work from just having a volume key or path. :) Regards and best wishes, Justin Clift --- tools/virsh.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 48 insertions(+), 0 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 1279f41..95aea0f 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -5975,6 +5975,53 @@ cmdVolName(vshControl *ctl, const vshCmd *cmd) } +/* + * "vol-pool" command + */ +static const vshCmdInfo info_vol_pool[] = { + {"help", N_("returns the storage pool for a given volume key or path")}, + {"desc", ""}, + {NULL, NULL} +}; + +static const vshCmdOptDef opts_vol_pool[] = { + {"vol", VSH_OT_DATA, VSH_OFLAG_REQ, N_("volume key or path")}, + {NULL, 0, 0, NULL} +}; + +static int +cmdVolPool(vshControl *ctl, const vshCmd *cmd) +{ + virStoragePoolPtr pool; + virStorageVolPtr vol; + + // Check the connection to libvirtd daemon is still working + if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) + return FALSE; + + // Use the supplied string to locate the volume + if (!(vol = vshCommandOptVolBy(ctl, cmd, "vol", "pool", NULL, + VSH_BYUUID))) { + return FALSE; + } + + // Look up the parent storage pool for the volume + pool = virStoragePoolLookupByVolume(vol); + if (pool == NULL) { + vshError(ctl, "%s", _("failed to get parent pool")); + virStorageVolFree(vol); + return FALSE; + } + + // Return the name of the parent storage pool + vshPrint(ctl, "%s\n", virStoragePoolGetName(pool)); + + // Cleanup + virStorageVolFree(vol); + virStoragePoolFree(pool); + return TRUE; +} + /* * "vol-key" command @@ -8837,6 +8884,7 @@ static const vshCmdDef commands[] = { {"vol-dumpxml", cmdVolDumpXML, opts_vol_dumpxml, info_vol_dumpxml}, {"vol-info", cmdVolInfo, opts_vol_info, info_vol_info}, {"vol-list", cmdVolList, opts_vol_list, info_vol_list}, + {"vol-pool", cmdVolPool, opts_vol_pool, info_vol_pool}, {"vol-path", cmdVolPath, opts_vol_path, info_vol_path}, {"vol-name", cmdVolName, opts_vol_name, info_vol_name}, {"vol-key", cmdVolKey, opts_vol_key, info_vol_key}, -- 1.7.0.1

2 2

[libvirt] [PATCH] Fix minor virsh man page typos and formatting problems.
by Justin Clift 08 Jun '10

08 Jun '10

--- tools/virsh.pod | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-) diff --git a/tools/virsh.pod b/tools/virsh.pod index 422ae7f..79900e5 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -89,7 +89,7 @@ Running hypervisor: Xen 3.0.0 =back -=item B<cd> I<directory> optional +=item B<cd> optional I<directory> Will change current directory to I<directory>. The default directory for the B<cd> command is the home directory or, if there is no I<HOME> @@ -273,13 +273,13 @@ pre-existing guest. B<Example> -virsh dumpxml <domain-id> > domain.xml -edit domain.xml -virsh create < domain.xml + virsh dumpxml <domain-id> > domain.xml + edit domain.xml + virsh create domain.xml =item B<define> I<FILE> -Define a domain from an XML <file>. The domain definitions is registered +Define a domain from an XML <file>. The domain definition is registered but not started. =item B<destroy> I<domain-id> @@ -490,7 +490,7 @@ is not available the processes will provide an exit code of 1. =item B<undefine> I<domain-id> Undefine the configuration for an inactive domain. Since it's not running -the domain name or UUId must be used as the I<domain-id>. +the domain name or UUID must be used as the I<domain-id>. =item B<vcpuinfo> I<domain-id> -- 1.7.0.1

3 4

[libvirt] Fwd: Re: [virt-tools-list] Lilo succeeds with kvm but not virt-manager
by Richard W.M. Jones 08 Jun '10

08 Jun '10

Forwarding to libvir-list ... Rich. ----- Forwarded message from Daniel Janzon <daniel.janzon(a)edgeware.tv> ----- Subject: Re: [virt-tools-list] Lilo succeeds with kvm but not virt-manager Date: Mon, 07 Jun 2010 10:56:29 +0200 From: Daniel Janzon <> To: virt-tools-list On 06/05/2010 06:00 PM, virt-tools-list-request(a)redhat.com wrote: > Message: 1 > Date: Fri, 4 Jun 2010 17:55:17 +0100 > From: "Richard W.M. Jones"<rjones(a)redhat.com> > To: Daniel Janzon<daniel.janzon(a)edgeware.tv> > Cc: virt-tools-list(a)redhat.com > Subject: Re: [virt-tools-list] Lilo bootloader succeeds with kvm but > not virt-manager > Message-ID:<20100604165517.GA4612(a)amd.home.annexia.org> > Content-Type: text/plain; charset=us-ascii > > On Fri, Jun 04, 2010 at 02:03:12PM +0200, Daniel Janzon wrote: > >> I installed Linux Suse as a guest in virt-manager/KVM and had trouble to >> install grub. So I used good old Lilo instead. When running Suse in >> virt-manager Lilo fails with >> >> LILO - Keytable read/checksum error on bootup >> >> When I go >> >> kvm -m 512 -hda SLES-10.3.img >> > libvirt won't have such a simple command line. There could be other > KVM options which affect things at boot. > > I have a couple of suggestions: > > (1) Find out the full libvirt command line. Easiest is to start the > domain under virt-manager and quickly use the 'ps ax' command. (There > is a more "official" way to convert libvirt XML into a qemu command > line but I cannot recall what that is at the moment). > Smart! I did that and found out that it is the boot=on drive option that is the cause of the problem: -drive file=/root/SLES-10.3.img,if=ide,index=0,boot=on For some reason it only works when it is set to boot=off. I tried <boot>off</boot> and boot="off" in the target tag in the corresponding disk tag in the XML file, but with no luck. I also tried to remove the <boot dev='hd'/> from the os tag. I did do a "/etc/init.d/libvirt-bin reload" just in case. Is there any way to get control the boot option from the XML file? > (2) Use virt-rescue or guestfish and see if you can install grub in > the domain. In guestfish it'd be something like this: > > guestfish -i GuestName > ><fs> mkdir-p /boot/grub > ><fs> grub-install /boot /dev/sda1 > > (You might need to change those command slightly and read the man page > for virt-rescue, guestfish and/or grub-install). > Ok, thanks for the tip. I will consider it if it impossible to control the boot option as explained above. Or maybe I'll just settle with being annoyed and start the machine from the command line :) All the best, Daniel Janzon _______________________________________________ virt-tools-list mailing list virt-tools-list(a)redhat.com https://www.redhat.com/mailman/listinfo/virt-tools-list ----- End forwarded message ----- -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://et.redhat.com/~rjones/libguestfs/ See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html

2 1

[libvirt] [PATCH] 0/10AppArmor driver updates
by Jamie Strandboge 08 Jun '10

08 Jun '10

This patch series addresses bug fixes in the AppArmor driver as well as updating it to changes made in 0.7.6 and 0.7.7. All of these are self-contained within the driver except 4_qemu_driver_stdin_path.patch. This is required by 5_apparmor-fix-save-restore.patch (see below). These all pass 'make syntax-check' and 'make check' (except 'daemon-conf', which has never passed here and I didn't patch it). 1_apparmor-dont-clear-caps.patch: originally submitted on 2010/02/08 with no feedback. The calls to virExec() in security_apparmor.c when invoking virt-aa-helper use VIR_EXEC_CLEAR_CAPS. When compiled without libcap-ng, this is not a problem (it's effectively a no-op) but with libcap-ng this causes MAC_ADMIN to be cleared. MAC_ADMIN is needed by virt-aa-helper to manipulate apparmor profiles and without it VMs will not start[1]. This patch calls virExec with the default VIR_EXEC_NONE instead. 2_apparmor-remove-unloaded-profile-is-not-fatal.patch: Don't exit with error if the user unloaded the profile outside of libvirt[2] 3_apparmor-fix-vah-xml-parse.patch: add VIR_DOMAIN_XML_INACTIVE flag to virDomainDefParseString() so virDomainDefParseString() doesn't error out when seeing <seclabel...>. This was needed due to changes since 0.7.5. 4_qemu_driver_stdin_path.patch: adjust args to qemudStartVMDaemon() to also specify path to stdin_fd, so this can be passed to the AppArmor driver via *SetSecurityAllLabel(). This updates all calls to qemudStartVMDaemon() as well as setting up the non-AppArmor security driver *SetSecurityAllLabel() declarations for the above. This is required for 5_apparmor-fix-save-restore.patch since AppArmor resolves the passed file descriptor to the pathname given to open(). 5_apparmor-fix-save-restore.patch: refactoring to update AppArmor security driver to adjust profile for save/restore[3] 6_apparmor-fix-backingstore.patch: adjust virt-aa-helper to handle backing store[4] 7_apparmor-fix-hostdev.patch: adjust virt-aa-helper to handle pci devices. Update valid_path() to have an override array to check against, and add "/sys/devices/pci" to it. Then rename file_iterate_cb() to file_iterate_hostdev_cb() and create file_iterate_pci_cb() based on it. 8_apparmor-fix-xauth.patch: adjust virt-aa-helper to handle SDL graphics, specifically Xauthority[6]. Also remove a couple redundant checks 9_apparmor-examples.patch: adjustments to the example profiles 10_apparmor-vah-test.patch: update pcidev test and add SDL xauth [1] https://launchpad.net/bugs/517714 [2] https://launchpad.net/bugs/530400 [3] https://launchpad.net/bugs/457716 [4] https://launchpad.net/bugs/470636 [5] https://launchpad.net/bugs/545795 [6] https://launchpad.net/bugs/545426 -- Jamie Strandboge | http://www.canonical.com

4 24