- Devel - libvirt List Archives

Re: [libvirt] [PATCH 2/9] Attach encryption information to virStorageVolDef.
by Daniel P. Berrange 24 Jul '09

24 Jul '09

On Fri, Jul 24, 2009 at 12:19:22AM -0400, Miloslav Trmac wrote: > ----- "Daniel P. Berrange" <berrange(a)redhat.com> wrote: > > > On Tue, Jul 21, 2009 at 01:11:58PM +0200, Miloslav Trma?? wrote: > > > Note that partial encryption information (e.g. specifying an encryption > > > format, but not the key/passphrase) is valid: > > > * virStorageVolGetXMLDesc() will never reveal the key/passphrase, even > > > if known by libvirt. > > > > I don't think that restriction really adds anything in the scenario > > that we're using domain XML files for persistent storage of keys. > > The patches leave this decision to the client: the client might decide > not to use persistent domains. In any case, allowing the node to store > the secrets locally does not imply that anyone that can connect to the > node read-write can access the secrets - or does it? In general all libvirt clients have the same privileges, the only restriction being local read-only clients. Basically if a client authenticates read-write then they can access / modify all objects managed on that host. In the future we will introduce fine grained access control down to a level of (user, object, operation). > > eg, if domain XML lets you see passphrases, then vol XML should > > too (given a suitable VIR_STORAGE_VOL_SECURE flag). > > Domain XML should not allow returning the secrets either, because domain > migration lets other nodes connect read-write - see [0/9] for more details. We have to allow returning the secrets, because clients need to be able to round-trip the XML without loosing data, eg xml = virDomainGetXMLDesc(dom, VIR_DOMAIN_XML_INACTIVE | VIR_DOMAIN_XML_SECURE) ...modify xml... virDomainDefineXML(conn, xml) must not loose any data, including the secrets. > > > * Future mechanisms could be set up to allow a libvirt user to specify > > > during volume creation that a volume should be encrypted, leaving > > > libvirt to choose suitable parameters and key and return them: > > > this would allow the libvirt user to automatically support any > > > encryption parameters (and perhaps encryption formats) supported in > > > libvirt, as long as the user can send the same information back when > > > using the volume in the future. > > > > We could allow this now without extra APIs, if we let virStorageVolGetXML > > show the ke/passphrase given a new VIR_STORAGE_VOL_SECURE flag. > > That is racy with respect to pool refresh, and gives other clients than > the node creator (again, other nodes) access to the secrets - again, > see [0/9]. Allowing other clients access to the secrets is fine - all authenticated clients are considered equal. We shouldn't try to define restrictions on this specifically for disk encryption, as fine grained per-client access control will be introduced across the whole of libvirt at a later date. There is a small race wrt to pool refresh, however, in the scenario of not using a keystore, this is an acceptable limitation IMHO. If using a keystore, then libvirt can associate a key ID with a volume and maintain that mapping long term, and reliably pass the actual secrets about internally. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

1 0

[libvirt] [PATCH] docs: say that the old repository is deprecated...
by Jim Meyering 24 Jul '09

24 Jul '09

I've just officially deprecated the original cvs-to-git-mirror libvirt repository on et.redhat.com, by git-removing all files and adding one with a pointer to the new repository: http://git.et.redhat.com/?p=libvirt.git;a=commitdiff;h=2940bc0fa9f2a3 Here's the corresponding change to update libvirt's web documentation: >From 521ac5174827400804d33e219ce02e6bd3da1089 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 24 Jul 2009 10:57:39 +0200 Subject: [PATCH] docs: say that the old repository is deprecated... * docs/downloads.html.in: but will remain for sake of old links. --- docs/downloads.html.in | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/docs/downloads.html.in b/docs/downloads.html.in index 0d2adb4..a34ddd1 100644 --- a/docs/downloads.html.in +++ b/docs/downloads.html.in @@ -74,9 +74,10 @@ <p> Jim Mereying was maintaining a CVS to git mirror on <a href="http://git.et.redhat.com/?p=libvirt.git">git.et.redhat.com</a>. - Existing users should migrate to the new git server, as the - old one will be deprecated and turned into a mirror of the - libvirt.org one. It's available as: + Existing users should migrate to the new libvirt.org git server, as the + old one is now deprecated. For the sake of old links including now- + rewritten SHA1s, we'll leave the old repository on-line for some time. + It is available as: </p> <pre> -- 1.6.4.rc2.182.g24de1

2 2

[libvirt] OpenVZ : cannot assign bridge to any interface
by Yuji NISHIDA 24 Jul '09

24 Jul '09

Hi, I'm bit confuing between 2 problems with libvirt-0.6.5 for OpenVZ. [1] bridge was created but no interface assigned to it [2] virsh dumpxml returns xml in which type is "qemu" [1] ------------------------------------------------- I tried libvirt-0.6.5 on Fedora8 to create OpenVZ container ( OpenVZ kernel is linux-2.6.18 ). A bridge was seemed to successfully create but no interface could not be assigned to the bridge. Due to that, ping could not reach the guest in container(10.0.1.2). ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ [root@node13 test]# brctl show bridge name bridge id STP enabled interfaces br101 8000.000000000000 no [root@node13 test]# ifconfig -a br101 Link encap:Ethernet HWaddr 00:00:00:00:00:00 inet addr:10.0.1.1 Bcast:10.0.1.15 Mask:255.255.255.240 inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:30 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:5031 (4.9 KiB) eth0 Link encap:Ethernet HWaddr 00:1C:C4:xx:xx:xx inet addr:133.xx.xx.xx Bcast:133.xx.xx.xx Mask:255.255.255.0 inet6 addr: fe80::21c:c4ff:xxxx:xxxx/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:196949 errors:0 dropped:0 overruns:0 frame:0 TX packets:64300 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:228373526 (217.7 MiB) TX bytes:5652593 (5.3 MiB) Interrupt:169 eth1 Link encap:Ethernet HWaddr 00:1C:C4:xx:xx:xx BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) Interrupt:50 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:323 errors:0 dropped:0 overruns:0 frame:0 TX packets:323 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:29949 (29.2 KiB) TX bytes:29949 (29.2 KiB) sit0 Link encap:IPv6-in-IPv4 NOARP MTU:1480 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) venet0 Link encap:UNSPEC HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00 UP BROADCAST POINTOPOINT RUNNING NOARP MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) veth101 Link encap:Ethernet HWaddr 52:54:00:52:C3:FF inet6 addr: fe80::5054:ff:fe52:c3ff/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ The followings are python script and xml that I tested with. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ [root@node13 test]# cat test.py import libvirt conn = libvirt.open('openvz:///system') conn.networkDefineXML( file('net.xml').read() ) conn.defineXML( file('test.xml').read() ) net = conn.networkLookupByName('guestnet101') net.create() guest = conn.lookupByName('101') guest.create() [root@node13 test]# cat test.xml <domain type='openvz' id='101'> <name>101</name> <os> <type>exe</type> <init>/sbin/fakeinit.sh</init> </os> <memory>524288</memory> <devices> <filesystem type='template'> <source name='fedora-10-x86_64'/> <target dir='/'/> </filesystem> <console type='pty'/> <interface type='bridge'> <mac address="52:54:00:00:01:01" /> <source bridge='guestnet101'/> <target dev='veth101'/> </interface> </devices> </domain> [root@node13 test]# cat net.xml <network> <name>guestnet101</name> <bridge name="br101" stp="off" /> <forward mode="nat"/> <ip address="10.0.1.1" netmask="255.255.255.240"> <dhcp> <range start="10.0.1.2" end="10.0.1.2" /> </dhcp> </ip> </network> ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ [2] ------------------------------------------------- I guess "type" in domain must be 'openvz' but virsh dumpxml returns it as 'qemu'. ( The configuration of VE 101 remains in [1] ) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ [root@node13 test]# virsh dumpxml 101 <domain type='qemu' id='101'> <name>101</name> <uuid>dedee02b-b74f-6e43-f5ec-27050986efa8</uuid> <memory>0</memory> <currentMemory>0</currentMemory> <vcpu>1</vcpu> <os> <type>exe</type> <init>/sbin/init</init> </os> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>destroy</on_reboot> <on_crash>destroy</on_crash> <devices> <filesystem type='template'> <source name='fedora-10-x86_64'/> <target dir='/'/> </filesystem> <interface type='bridge'> <mac address='52:54:00:00:01:01'/> <source bridge='guestnet101'/> <target dev='veth101'/> </interface> </devices> </domain> [root@node13 test]# virsh net-dumpxml guestnet101 <network> <name>guestnet101</name> <uuid>5462e7a5-2048-3067-af9d-b134381a3f8a</uuid> <forward mode='nat'/> <bridge name='br101' stp='off' forwardDelay='0' /> <ip address='10.0.1.1' netmask='255.255.255.240'> <dhcp> <range start='10.0.1.2' end='10.0.1.2' /> </dhcp> </ip> </network> ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Regards. ----- Yuji Nishida nishidy(a)nict.go.jp

3 3

[libvirt] How to call QEMU with -snapshot ?
by Jun Koi 24 Jul '09

24 Jul '09

Hi, I want to execute qemu (actually is KVM, qemu-system_x86-64) with -snapshot. Is there any way to do that? I tried following ways, and none works: - Put -snapshot in <emulator> in configuration file. - Create a script calling "qemu -snapshot", and rename it to qemu-system_x86-64 (and of course i rename the original one to smt else, and let my scrip call that) In all these cases, I got error about parsing for QEMU version. Thanks, J

1 0

Re: [libvirt] [PATCH 9/9] Add support for qcow encrypted volumes to qemu.
by Miloslav Trmac 24 Jul '09

24 Jul '09

----- "Daniel P. Berrange" <berrange(a)redhat.com> wrote: > On Tue, Jul 21, 2009 at 01:12:05PM +0200, Miloslav Trma?? wrote: > > --- > > src/qemu_driver.c | 51 > ++++++++++++++++++++++++++++++++++++++++++++++++++- > > 1 files changed, 50 insertions(+), 1 deletions(-) > > IIUC this should only be needed when starting a QEMU guest. > > The 'cont' method is called from quite a few different contexts. > So it might be nice to pass a flag down from qemudStartVMDaemon > such that the passphrase handling is guarenteed to only be run > during startup (qemudStartVMDaemon is also used for migrate > and restore, so covers those scenarios OK) qemudStartVMDaemon is called during migration (and restore), but it does not sent "cont". The only place where "cont" is sent where passphrase handling is not necessary is qemudDomainResume. In any case, the passphrase handling only does something if qemu prompts for a passphrase, it does not hurt anything in other cases - on the contrary, should something go wrong and qemu would prompt for a passphrase, this ensures libvirt either supplies a passphrase, or reports an error, instead of hanging. Mirek

1 0

[libvirt] [PATCH] Add bare %s format string to printf-derivatives called with no format string
by Laine Stump 23 Jul '09

23 Jul '09

--- src/datatypes.c | 24 ++++++++++++------------ src/domain_conf.c | 9 +++++---- src/interface_conf.c | 4 ++-- src/lxc_driver.c | 2 +- src/qemu_driver.c | 4 ++-- src/storage_backend.c | 6 +++--- src/virsh.c | 4 ++-- 7 files changed, 27 insertions(+), 26 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index 1e33ccb..8b3f8c3 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -289,7 +289,7 @@ virGetDomain(virConnectPtr conn, const char *name, const unsigned char *uuid) { if (virHashAddEntry(conn->domains, name, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add domain to connection hash table")); + "%s", _("failed to add domain to connection hash table")); goto error; } conn->refs++; @@ -330,7 +330,7 @@ virReleaseDomain(virDomainPtr domain) { if (virHashRemoveEntry(conn->domains, domain->name, NULL) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("domain missing from connection hash table")); + "%s", _("domain missing from connection hash table")); conn = NULL; } @@ -430,7 +430,7 @@ virGetNetwork(virConnectPtr conn, const char *name, const unsigned char *uuid) { if (virHashAddEntry(conn->networks, name, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add network to connection hash table")); + "%s", _("failed to add network to connection hash table")); goto error; } conn->refs++; @@ -468,7 +468,7 @@ virReleaseNetwork(virNetworkPtr network) { if (virHashRemoveEntry(conn->networks, network->name, NULL) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("network missing from connection hash table")); + "%s", _("network missing from connection hash table")); conn = NULL; } @@ -603,7 +603,7 @@ _("Failed to change interface mac address from %s to %s due to differing lengths if (virHashAddEntry(conn->interfaces, name, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add interface to connection hash table")); + "%s", _("failed to add interface to connection hash table")); goto error; } conn->refs++; @@ -642,7 +642,7 @@ virReleaseInterface(virInterfacePtr iface) { /* unlock before reporting error because error report grabs lock */ virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("interface missing from connection hash table")); + "%s", _("interface missing from connection hash table")); /* don't decr the conn refct if we weren't connected to it */ conn = NULL; } @@ -744,7 +744,7 @@ virGetStoragePool(virConnectPtr conn, const char *name, const unsigned char *uui if (virHashAddEntry(conn->storagePools, name, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add storage pool to connection hash table")); + "%s", _("failed to add storage pool to connection hash table")); goto error; } conn->refs++; @@ -783,7 +783,7 @@ virReleaseStoragePool(virStoragePoolPtr pool) { if (virHashRemoveEntry(conn->storagePools, pool->name, NULL) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("pool missing from connection hash table")); + "%s", _("pool missing from connection hash table")); conn = NULL; } @@ -887,7 +887,7 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c if (virHashAddEntry(conn->storageVols, key, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add storage vol to connection hash table")); + "%s", _("failed to add storage vol to connection hash table")); goto error; } conn->refs++; @@ -927,7 +927,7 @@ virReleaseStorageVol(virStorageVolPtr vol) { if (virHashRemoveEntry(conn->storageVols, vol->key, NULL) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("vol missing from connection hash table")); + "%s", _("vol missing from connection hash table")); conn = NULL; } @@ -1023,7 +1023,7 @@ virGetNodeDevice(virConnectPtr conn, const char *name) if (virHashAddEntry(conn->nodeDevices, name, ret) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("failed to add node dev to conn hash table")); + "%s", _("failed to add node dev to conn hash table")); goto error; } conn->refs++; @@ -1061,7 +1061,7 @@ virReleaseNodeDevice(virNodeDevicePtr dev) { if (virHashRemoveEntry(conn->nodeDevices, dev->name, NULL) < 0) { virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, - _("dev missing from connection hash table")); + "%s", _("dev missing from connection hash table")); conn = NULL; } diff --git a/src/domain_conf.c b/src/domain_conf.c index fbe9d78..96cadd5 100644 --- a/src/domain_conf.c +++ b/src/domain_conf.c @@ -2189,7 +2189,7 @@ virSecurityLabelDefParseXML(virConnectPtr conn, VIR_FREE(p); if (def->seclabel.type < 0) { virDomainReportError(conn, VIR_ERR_XML_ERROR, - _("invalid security type")); + "%s", _("invalid security type")); goto error; } @@ -2211,7 +2211,7 @@ virSecurityLabelDefParseXML(virConnectPtr conn, VIR_SECURITY_LABEL_BUFLEN-1, ctxt); if (p == NULL) { virDomainReportError(conn, VIR_ERR_XML_ERROR, - _("security label is missing")); + "%s", _("security label is missing")); goto error; } @@ -2224,8 +2224,9 @@ virSecurityLabelDefParseXML(virConnectPtr conn, p = virXPathStringLimit(conn, "string(./seclabel/imagelabel[1])", VIR_SECURITY_LABEL_BUFLEN-1, ctxt); if (p == NULL) { - virDomainReportError(conn, VIR_ERR_XML_ERROR, - _("security imagelabel is missing")); + +virDomainReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("security imagelabel is missing")); goto error; } def->seclabel.imagelabel = p; diff --git a/src/interface_conf.c b/src/interface_conf.c index 7da1a2f..a74d17e 100644 --- a/src/interface_conf.c +++ b/src/interface_conf.c @@ -601,7 +601,7 @@ virInterfaceDefParseXML(virConnectPtr conn, xmlXPathContextPtr ctxt) { tmp = virXPathString(conn, "string(./@type)", ctxt); if (tmp == NULL) { virInterfaceReportError(conn, VIR_ERR_XML_ERROR, - _("interface misses the type attribute")); + "%s", _("interface misses the type attribute")); return(NULL); } type = virInterfaceTypeFromString(tmp); @@ -643,7 +643,7 @@ virInterfaceDefParseXML(virConnectPtr conn, xmlXPathContextPtr ctxt) { bridge = virXPathNode(conn, "./bridge[1]", ctxt); if (bridge == NULL) { virInterfaceReportError(conn, VIR_ERR_XML_ERROR, - _("bridge interface misses the bridge element")); + "%s", _("bridge interface misses the bridge element")); goto error; } tmp = virXMLPropString(bridge, "stp"); diff --git a/src/lxc_driver.c b/src/lxc_driver.c index 3503573..66cfa8e 100644 --- a/src/lxc_driver.c +++ b/src/lxc_driver.c @@ -102,7 +102,7 @@ static virDrvOpenStatus lxcOpen(virConnectPtr conn, /* URI was good, but driver isn't active */ if (lxc_driver == NULL) { lxcError(conn, NULL, VIR_ERR_INTERNAL_ERROR, - _("lxc state driver is not active")); + "%s", _("lxc state driver is not active")); return VIR_DRV_OPEN_ERROR; } } diff --git a/src/qemu_driver.c b/src/qemu_driver.c index c4683ae..3b56092 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -1599,7 +1599,7 @@ static int qemudSecurityHook(void *data) { if (qemudDomainSetSecurityLabel(h->conn, h->driver, h->vm) < 0) { qemudReportError(h->conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Failed to set security label")); + "%s", _("Failed to set security label")); return -1; } @@ -3592,7 +3592,7 @@ static int qemudDomainGetSecurityLabel(virDomainPtr dom, virSecurityLabelPtr sec if (driver->securityDriver && driver->securityDriver->domainGetSecurityLabel) { if (driver->securityDriver->domainGetSecurityLabel(dom->conn, vm, seclabel) == -1) { qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, - _("Failed to get security label")); + "%s", _("Failed to get security label")); goto cleanup; } } diff --git a/src/storage_backend.c b/src/storage_backend.c index 67815d7..9f2dcca 100644 --- a/src/storage_backend.c +++ b/src/storage_backend.c @@ -423,7 +423,7 @@ virStorageBackendCreateQemuImg(virConnectPtr conn, use_kvmimg = 0; else { virStorageReportError(conn, VIR_ERR_INTERNAL_ERROR, - _("unable to find kvm-img or qemu-img")); + "%s", _("unable to find kvm-img or qemu-img")); return -1; } @@ -485,9 +485,9 @@ virStorageBackendCreateQcowCreate(virConnectPtr conn, return -1; } if (vol->backingStore.path != NULL) { - virStorageReportError(conn, VIR_ERR_NO_SUPPORT, + virStorageReportError(conn, VIR_ERR_NO_SUPPORT, "%s", _("copy-on-write image not supported with " - "qcow-create")); + "qcow-create")); return -1; } diff --git a/src/virsh.c b/src/virsh.c index fff73a1..4499aca 100644 --- a/src/virsh.c +++ b/src/virsh.c @@ -4869,7 +4869,7 @@ cmdVolClone(vshControl *ctl, const vshCmd *cmd) origpool = virStoragePoolLookupByVolume(origvol); if (!origpool) { - vshError(ctl, FALSE, _("failed to get parent pool")); + vshError(ctl, FALSE, "%s", _("failed to get parent pool")); goto cleanup; } @@ -6583,7 +6583,7 @@ cmdCd(vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) int found; if (!ctl->imode) { - vshError(ctl, FALSE, _("cd: command valid only in interactive mode")); + vshError(ctl, FALSE, "%s", _("cd: command valid only in interactive mode")); return -1; } -- 1.6.0.6

3 3

[libvirt] [PATCH 0/8] Support cgroups in QEMU driver (v2)
by Daniel P. Berrange 23 Jul '09

23 Jul '09

This is an updated series of the v1 patches here: http://www.redhat.com/archives/libvir-list/2009-July/msg00435.html I have re-ordered the patches a little to put the generic refactoring all first. I identified some bugs in the existing LXC driver schedular parameter handling which I fixed. Also fixed virsh to use the correct data types when setting schedular parameters instead of hardcoding a bogus type. Finally, I've made QEMU's use of cgroups configurable. Daniel P. Berrange (8): Use enums for cgroup controller types / labels Use virFileReadAll/virFileWriteStr for key cgroup read/write helpers Make cgroups a little more efficient Refactor cgroups to allow a group per driver to be managed directly Place every QEMU guest in a private cgroup Implement schedular tunables API using cgroups Use cgroups for block device whitelisting in QEMU guests Make QEMU cgroups use configurable qemud/libvirtd_qemu.aug | 2 + qemud/test_libvirtd_qemu.aug | 21 +- src/Makefile.am | 3 +- src/cgroup.c | 867 +++++++++++++++++++++++------------------- src/cgroup.h | 42 ++- src/libvirt_private.syms | 2 + src/lxc_conf.h | 2 + src/lxc_controller.c | 19 +- src/lxc_driver.c | 34 ++- src/qemu.conf | 34 ++ src/qemu_conf.c | 61 +++ src/qemu_conf.h | 5 + src/qemu_driver.c | 399 +++++++++++++++++++- src/util.c | 47 +++- src/util.h | 5 + src/virsh.c | 232 +++++++----- 16 files changed, 1237 insertions(+), 538 deletions(-)

2 18

[libvirt] clone memory leak and libvirt connection selection
by Johnny Luong 23 Jul '09

23 Jul '09

Hi, I'm trying to use libvirt-0.6.5 within a multithreaded application without libvirtd and from what it looks like, it appears to have two things that are somewhat undesirable: 1) It tries to connect to libvirtd which isn't running. But it does eventually connect to the hypervisor. 2) It appears to leak some memory within a readonly connect and close. This is not too bad but it would be an issue in my long lived daemon application. I've attached an example program illustrating the issue with the linker log and the log from valgrind showing the memory leak. The code itself is based off the one on the website. Is there a URI I can use to avoid the connection to the libvirtd daemon while still connecting to the xendstore and is the memory leak an issue with the application environment or the library? Let me know if you need any more information. Thanks, Johnny

2 1

[libvirt] [PATCH 0/2] Add some more features to LXC driver
by Daniel P. Berrange 23 Jul '09

23 Jul '09

This patch series simply adds a couple of easy features to the LXC driver. Daniel P. Berrange (2): Add domain events support to LXC driver Add domain autostart for LXC driver src/lxc_conf.h | 7 + src/lxc_driver.c | 365 +++++++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 367 insertions(+), 5 deletions(-)

2 6

[libvirt] MemoryPeek() is slow
by Jun Koi 23 Jul '09

23 Jul '09

Hi, I play around with MemoryPeek() API on QEMU. While it works well, I found that it is too slow. That is expected because of the way it works: we always need to save memory to a file, and read it in again, and that is too inefficient. I am trying to figure out a better way to do this. To do that, clearly we need to re-architecture QEMU for this: We must have another way to read memory from outside the QEMU process. Anybody could suggest a solution for this problem? I am willing to spend time on this feature to improve the situation. Thanks, J

3 6

[libvirt] PATCH: Fix many many win32 bugs
by Daniel P. Berrange 23 Jul '09

23 Jul '09

Latest GCC changed the meaning of 'printf' attribute annotation to refer to the host platform's native format specifiers. On Win32/mingw32 this means it is now validating against Microsoft's useless format specifiers. We replace all the printf like functions on Win32 thanks to GNULIB and thus should always validate against the GNU format specifiers. This means we have to use the 'gnu_printf' annotation everywhere, but only for GCC >= 4.4. This also disables the 'pwd' and 'cd' commands in virsh because neither of those build successfully on Win32. Finally, disables the OpenNebula driver on Win32 Daniel diff --git a/HACKING b/HACKING index ca39d61..da28e98 100644 --- a/HACKING +++ b/HACKING @@ -312,7 +312,7 @@ gcc's printf attribute directive in the prototype. For example, here's the one for virAsprintf, in util.h: int virAsprintf(char **strp, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 2, 3); + ATTRIBUTE_FMT_PRINTF(2, 3); This makes it so gcc's -Wformat and -Wformat-security options can do their jobs and cross-check format strings with the number and types diff --git a/autobuild.sh b/autobuild.sh index 54abede..d090e74 100755 --- a/autobuild.sh +++ b/autobuild.sh @@ -76,6 +76,7 @@ if [ -x /usr/bin/i686-pc-mingw32-gcc ]; then --without-uml \ --without-vbox \ --without-openvz \ + --without-one \ --without-libvirtd make diff --git a/mingw32-libvirt.spec.in b/mingw32-libvirt.spec.in index 271c6ca..9514032 100644 --- a/mingw32-libvirt.spec.in +++ b/mingw32-libvirt.spec.in @@ -52,6 +52,7 @@ MinGW Windows libvirt virtualization library. --without-uml \ --without-vbox \ --without-openvz \ + --without-one \ --without-libvirtd make @@ -90,6 +91,7 @@ rm -rf $RPM_BUILD_ROOT %{_mingw32_datadir}/libvirt/schemas/storagevol.rng %{_mingw32_datadir}/libvirt/schemas/nodedev.rng %{_mingw32_datadir}/libvirt/schemas/capability.rng +%{_mingw32_datadir}/libvirt/schemas/interface.rng %{_mingw32_datadir}/locale/*/LC_MESSAGES/libvirt.mo diff --git a/qemud/dispatch.h b/qemud/dispatch.h index ab45b19..1d85df9 100644 --- a/qemud/dispatch.h +++ b/qemud/dispatch.h @@ -41,7 +41,7 @@ remoteDispatchClientRequest (struct qemud_server *server, void remoteDispatchFormatError (remote_error *rerr, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 2, 3); + ATTRIBUTE_FMT_PRINTF(2, 3); void remoteDispatchAuthError (remote_error *rerr); void remoteDispatchGenericError (remote_error *rerr); diff --git a/qemud/qemud.h b/qemud/qemud.h index abacbbb..254db44 100644 --- a/qemud/qemud.h +++ b/qemud/qemud.h @@ -52,15 +52,61 @@ #ifdef HAVE_ANSIDECL_H #include <ansidecl.h> #endif + +#ifndef __GNUC_PREREQ +#if defined __GNUC__ && defined __GNUC_MINOR__ +# define __GNUC_PREREQ(maj, min) \ + ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min)) +#else +#define __GNUC_PREREQ(maj,min) 0 +#endif +#endif + +/** + * ATTRIBUTE_UNUSED: + * + * Macro to flag conciously unused parameters to functions + */ #ifndef ATTRIBUTE_UNUSED #define ATTRIBUTE_UNUSED __attribute__((__unused__)) #endif -#ifndef ATTRIBUTE_FORMAT -#define ATTRIBUTE_FORMAT(args...) __attribute__((__format__ (args))) + +/** + * ATTRIBUTE_FMT_PRINTF + * + * Macro used to check printf like functions, if compiling + * with gcc. + * + * We use gnulib which guarentees we always have GNU style + * printf format specifiers even on broken Win32 platforms + * hence we have to force 'gnu_printf' for new GCC + */ +#ifndef ATTRIBUTE_FMT_PRINTF +#if __GNUC_PREREQ (4, 4) +#define ATTRIBUTE_FMT_PRINTF(fmtpos,argpos) __attribute__((__format__ (gnu_printf, fmtpos,argpos))) +#else +#define ATTRIBUTE_FMT_PRINTF(fmtpos,argpos) __attribute__((__format__ (printf, fmtpos,argpos))) +#endif #endif + +#ifndef ATTRIBUTE_RETURN_CHECK +#if __GNUC_PREREQ (3, 4) +#define ATTRIBUTE_RETURN_CHECK __attribute__((__warn_unused_result__)) #else +#define ATTRIBUTE_RETURN_CHECK +#endif +#endif + +#else +#ifndef ATTRIBUTE_UNUSED #define ATTRIBUTE_UNUSED -#define ATTRIBUTE_FORMAT(...) +#endif +#ifndef ATTRIBUTE_FMT_PRINTF +#define ATTRIBUTE_FMT_PRINTF(...) +#endif +#ifndef ATTRIBUTE_RETURN_CHECK +#define ATTRIBUTE_RETURN_CHECK +#endif #endif #define qemudDebug DEBUG @@ -213,7 +259,7 @@ struct qemud_server { }; void qemudLog(int priority, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf,2,3); + ATTRIBUTE_FMT_PRINTF(2,3); diff --git a/src/buf.h b/src/buf.h index 0b78d17..7d31cb2 100644 --- a/src/buf.h +++ b/src/buf.h @@ -40,7 +40,7 @@ unsigned int virBufferUse(const virBufferPtr buf); void virBufferAdd(const virBufferPtr buf, const char *str, int len); void virBufferAddChar(const virBufferPtr buf, char c); void virBufferVSprintf(const virBufferPtr buf, const char *format, ...) - ATTRIBUTE_FORMAT(printf, 2, 3); + ATTRIBUTE_FMT_PRINTF(2, 3); void virBufferStrcat(const virBufferPtr buf, ...); void virBufferEscapeString(const virBufferPtr buf, const char *format, const char *str); void virBufferURIEncodeString (const virBufferPtr buf, const char *str); diff --git a/src/internal.h b/src/internal.h index adb6507..cf7e196 100644 --- a/src/internal.h +++ b/src/internal.h @@ -67,8 +67,13 @@ #ifdef __GNUC__ #ifndef __GNUC_PREREQ +#if defined __GNUC__ && defined __GNUC_MINOR__ +# define __GNUC_PREREQ(maj, min) \ + ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min)) +#else #define __GNUC_PREREQ(maj,min) 0 #endif +#endif /** * ATTRIBUTE_UNUSED: @@ -80,13 +85,21 @@ #endif /** - * ATTRIBUTE_FORMAT + * ATTRIBUTE_FMT_PRINTF * - * Macro used to check printf/scanf-like functions, if compiling + * Macro used to check printf like functions, if compiling * with gcc. + * + * We use gnulib which guarentees we always have GNU style + * printf format specifiers even on broken Win32 platforms + * hence we have to force 'gnu_printf' for new GCC */ -#ifndef ATTRIBUTE_FORMAT -#define ATTRIBUTE_FORMAT(args...) __attribute__((__format__ (args))) +#ifndef ATTRIBUTE_FMT_PRINTF +#if __GNUC_PREREQ (4, 4) +#define ATTRIBUTE_FMT_PRINTF(fmtpos,argpos) __attribute__((__format__ (gnu_printf, fmtpos,argpos))) +#else +#define ATTRIBUTE_FMT_PRINTF(fmtpos,argpos) __attribute__((__format__ (printf, fmtpos,argpos))) +#endif #endif #ifndef ATTRIBUTE_RETURN_CHECK @@ -98,9 +111,15 @@ #endif #else +#ifndef ATTRIBUTE_UNUSED #define ATTRIBUTE_UNUSED -#define ATTRIBUTE_FORMAT(...) +#endif +#ifndef ATTRIBUTE_FMT_PRINTF +#define ATTRIBUTE_FMT_PRINTF(...) +#endif +#ifndef ATTRIBUTE_RETURN_CHECK #define ATTRIBUTE_RETURN_CHECK +#endif #endif /* __GNUC__ */ /* diff --git a/src/logging.h b/src/logging.h index 0b9ae79..f1e2525 100644 --- a/src/logging.h +++ b/src/logging.h @@ -121,6 +121,6 @@ extern int virLogParseFilters(const char *filters); extern int virLogParseOutputs(const char *output); extern void virLogMessage(const char *category, int priority, const char *funcname, long long linenr, int flags, - const char *fmt, ...) ATTRIBUTE_FORMAT(printf, 6, 7); + const char *fmt, ...) ATTRIBUTE_FMT_PRINTF(6, 7); #endif diff --git a/src/security.h b/src/security.h index 585caa0..5fc3086 100644 --- a/src/security.h +++ b/src/security.h @@ -82,7 +82,7 @@ virSecurityDriverVerify(virConnectPtr conn, virDomainDefPtr def); void virSecurityReportError(virConnectPtr conn, int code, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 3, 4); + ATTRIBUTE_FMT_PRINTF(3, 4); /* Helpers */ void virSecurityDriverInit(virSecurityDriverPtr drv); diff --git a/src/sexpr.h b/src/sexpr.h index c3d038e..1d9adaa 100644 --- a/src/sexpr.h +++ b/src/sexpr.h @@ -49,7 +49,7 @@ void sexpr_free(struct sexpr *sexpr); const char *sexpr_node(const struct sexpr *sexpr, const char *node); int sexpr_node_copy(const struct sexpr *sexpr, const char *node, char **dst); const char *sexpr_fmt_node(const struct sexpr *sexpr, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf,2,3); + ATTRIBUTE_FMT_PRINTF(2,3); struct sexpr *sexpr_lookup(const struct sexpr *sexpr, const char *node); int sexpr_has(const struct sexpr *sexpr, const char *node); #endif diff --git a/src/threads-win32.c b/src/threads-win32.c index 10233b0..ef3f2ef 100644 --- a/src/threads-win32.c +++ b/src/threads-win32.c @@ -41,8 +41,10 @@ void virCondEventCleanup(void *data); int virThreadInitialize(void) { - virMutexInit(&virThreadLocalLock); - virThreadLocalInit(&virCondEvent, virCondEventCleanup); + if (virMutexInit(&virThreadLocalLock) < 0) + return -1; + if (virThreadLocalInit(&virCondEvent, virCondEventCleanup) < 0) + return -1; return 0; } diff --git a/src/util.h b/src/util.h index 1a7286c..43a5bbc 100644 --- a/src/util.h +++ b/src/util.h @@ -160,7 +160,7 @@ int virMacAddrCompare (const char *mac1, const char *mac2); void virSkipSpaces(const char **str); int virParseNumber(const char **str); int virAsprintf(char **strp, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 2, 3); + ATTRIBUTE_FMT_PRINTF(2, 3); #define VIR_MAC_BUFLEN 6 #define VIR_MAC_PREFIX_BUFLEN 3 diff --git a/src/virsh.c b/src/virsh.c index fff73a1..6919292 100644 --- a/src/virsh.c +++ b/src/virsh.c @@ -202,7 +202,7 @@ typedef struct __vshControl { static const vshCmdDef commands[]; static void vshError(vshControl *ctl, int doexit, const char *format, ...) - ATTRIBUTE_FORMAT(printf, 3, 4); + ATTRIBUTE_FMT_PRINTF(3, 4); static int vshInit(vshControl *ctl); static int vshDeinit(vshControl *ctl); static void vshUsage(void); @@ -272,9 +272,9 @@ static virStorageVolPtr vshCommandOptVolBy(vshControl *ctl, const vshCmd *cmd, VSH_BYUUID|VSH_BYNAME) static void vshPrintExtra(vshControl *ctl, const char *format, ...) - ATTRIBUTE_FORMAT(printf, 2, 3); + ATTRIBUTE_FMT_PRINTF(2, 3); static void vshDebug(vshControl *ctl, int level, const char *format, ...) - ATTRIBUTE_FORMAT(printf, 3, 4); + ATTRIBUTE_FMT_PRINTF(3, 4); /* XXX: add batch support */ #define vshPrint(_ctl, ...) fprintf(stdout, __VA_ARGS__) @@ -495,6 +495,8 @@ cmdConnect(vshControl *ctl, const vshCmd *cmd) return ctl->conn ? TRUE : FALSE; } +#ifndef WIN32 + /* * "console" command */ @@ -510,8 +512,6 @@ static const vshCmdOptDef opts_console[] = { {NULL, 0, 0, NULL} }; -#ifndef __MINGW32__ - static int cmdRunConsole(vshControl *ctl, virDomainPtr dom) { @@ -574,17 +574,6 @@ cmdRunConsole(vshControl *ctl, virDomainPtr dom) return ret; } -#else /* __MINGW32__ */ - -static int -cmdRunConsole(vshControl *ctl, virDomainPtr dom ATTRIBUTE_UNUSED) -{ - vshError (ctl, FALSE, "%s", _("console not implemented on this platform")); - return FALSE; -} - -#endif /* __MINGW32__ */ - static int cmdConsole(vshControl *ctl, const vshCmd *cmd) { @@ -603,6 +592,9 @@ cmdConsole(vshControl *ctl, const vshCmd *cmd) return ret; } +#endif /* WIN32 */ + + /* * "list" command */ @@ -931,7 +923,9 @@ static const vshCmdInfo info_create[] = { static const vshCmdOptDef opts_create[] = { {"file", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("file containing an XML domain description")}, +#ifndef WIN32 {"console", VSH_OT_BOOL, 0, gettext_noop("attach to console after creation")}, +#endif {NULL, 0, 0, NULL} }; @@ -943,7 +937,9 @@ cmdCreate(vshControl *ctl, const vshCmd *cmd) int found; int ret = TRUE; char *buffer; +#ifndef WIN32 int console = vshCommandOptBool(cmd, "console"); +#endif if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -961,8 +957,10 @@ cmdCreate(vshControl *ctl, const vshCmd *cmd) if (dom != NULL) { vshPrint(ctl, _("Domain %s created from %s\n"), virDomainGetName(dom), from); +#ifndef WIN32 if (console) cmdRunConsole(ctl, dom); +#endif virDomainFree(dom); } else { vshError(ctl, FALSE, _("Failed to create domain from %s"), from); @@ -1083,7 +1081,9 @@ static const vshCmdInfo info_start[] = { static const vshCmdOptDef opts_start[] = { {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive domain")}, +#ifndef WIN32 {"console", VSH_OT_BOOL, 0, gettext_noop("attach to console after creation")}, +#endif {NULL, 0, 0, NULL} }; @@ -1092,7 +1092,9 @@ cmdStart(vshControl *ctl, const vshCmd *cmd) { virDomainPtr dom; int ret = TRUE; +#ifndef WIN32 int console = vshCommandOptBool(cmd, "console"); +#endif if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -1109,8 +1111,10 @@ cmdStart(vshControl *ctl, const vshCmd *cmd) if (virDomainCreate(dom) == 0) { vshPrint(ctl, _("Domain %s started\n"), virDomainGetName(dom)); +#ifndef WIN32 if (console) cmdRunConsole(ctl, dom); +#endif } else { vshError(ctl, FALSE, _("Failed to start domain %s"), virDomainGetName(dom)); @@ -6562,6 +6566,8 @@ editReadBackFile (vshControl *ctl, const char *filename) return ret; } + +#ifndef WIN32 /* * "cd" command */ @@ -6603,6 +6609,9 @@ cmdCd(vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) return 0; } +#endif + +#ifndef WIN32 /* * "pwd" command */ @@ -6638,6 +6647,7 @@ cmdPwd(vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) free (cwd); return !err; } +#endif /* * "edit" command @@ -6802,9 +6812,13 @@ static const vshCmdDef commands[] = { {"attach-interface", cmdAttachInterface, opts_attach_interface, info_attach_interface}, {"autostart", cmdAutostart, opts_autostart, info_autostart}, {"capabilities", cmdCapabilities, NULL, info_capabilities}, +#ifndef WIN32 {"cd", cmdCd, opts_cd, info_cd}, +#endif {"connect", cmdConnect, opts_connect, info_connect}, +#ifndef WIN32 {"console", cmdConsole, opts_console, info_console}, +#endif {"create", cmdCreate, opts_create, info_create}, {"start", cmdStart, opts_start, info_start}, {"destroy", cmdDestroy, opts_destroy, info_destroy}, @@ -6882,7 +6896,9 @@ static const vshCmdDef commands[] = { {"pool-undefine", cmdPoolUndefine, opts_pool_undefine, info_pool_undefine}, {"pool-uuid", cmdPoolUuid, opts_pool_uuid, info_pool_uuid}, +#ifndef WIN32 {"pwd", cmdPwd, NULL, info_pwd}, +#endif {"quit", cmdQuit, NULL, info_quit}, {"reboot", cmdReboot, opts_reboot, info_reboot}, {"restore", cmdRestore, opts_restore, info_restore}, diff --git a/src/virterror_internal.h b/src/virterror_internal.h index fe9a96f..da89de7 100644 --- a/src/virterror_internal.h +++ b/src/virterror_internal.h @@ -46,7 +46,7 @@ void virRaiseErrorFull(virConnectPtr conn, int int1, int int2, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 13, 14); + ATTRIBUTE_FMT_PRINTF(13, 14); /* Includes 'dom' and 'net' for compatbility, but they're ignored */ #define virRaiseError(conn, dom, net, domain, code, level, \ @@ -61,7 +61,7 @@ void virReportErrorHelper(virConnectPtr conn, int domcode, int errcode, const char *funcname ATTRIBUTE_UNUSED, size_t linenr ATTRIBUTE_UNUSED, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 7, 8); + ATTRIBUTE_FMT_PRINTF(7, 8); void virReportSystemErrorFull(virConnectPtr conn, int domcode, @@ -70,7 +70,7 @@ void virReportSystemErrorFull(virConnectPtr conn, const char *funcname, size_t linenr, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf, 7, 8); + ATTRIBUTE_FMT_PRINTF(7, 8); #define virReportSystemError(conn, theerrno, fmt,...) \ virReportSystemErrorFull((conn), \ diff --git a/src/xend_internal.c b/src/xend_internal.c index 55ffd3f..e23ae2b 100644 --- a/src/xend_internal.c +++ b/src/xend_internal.c @@ -582,7 +582,7 @@ xend_op(virConnectPtr xend, const char *name, const char *key, ...) * Returns a parsed S-Expression in case of success, NULL in case of failure */ static struct sexpr *sexpr_get(virConnectPtr xend, const char *fmt, ...) - ATTRIBUTE_FORMAT(printf,2,3); + ATTRIBUTE_FMT_PRINTF(2,3); static struct sexpr * sexpr_get(virConnectPtr xend, const char *fmt, ...) -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] How to config my VM to use KVM with libvirt?
by Jun Koi 22 Jul '09

22 Jul '09

Hi, I have a VM running under libvirt, and it is currently run with -no-kvm option (I saw that in "ps" output). Now I want to run this VM with KVM. How can I reconfigure it for that? I looked into its configuration file under /etc/libvirt/qemu, but didnt see any option to turn KVM on. Thanks, J

4 7

[libvirt] [PATCH 0/4] Fixes to some bugs encountered testing virInterface*
by Laine Stump 22 Jul '09

22 Jul '09

This series of patches fixes some problems I found while testing the virInterface implementation. Note that the 3rd patch in the series only fixes virInterface-related instances of the problem (releasing the conn lock before reporting errors). If that patch is approved, a similar thing must be done for several other functions in datatypes.c.

3 18

[libvirt] [PATCH] Release conn lock before reporting errors
by Laine Stump 22 Jul '09

22 Jul '09

This is a follow-on to 528d37. It's fixing several other cases in datatypes.c where we try to report an error while holding the conn's lock, which can't work because the error reporting also tries to lock the conn. --- src/datatypes.c | 133 ++++++++++++++++++++++++++++++++++--------------------- 1 files changed, 82 insertions(+), 51 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index ba5401d..c2030dd 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -270,11 +270,13 @@ virGetDomain(virConnectPtr conn, const char *name, const unsigned char *uuid) { /* TODO check the UUID */ if (ret == NULL) { if (VIR_ALLOC(ret) < 0) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->name = strdup(name); if (ret->name == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -285,6 +287,7 @@ virGetDomain(virConnectPtr conn, const char *name, const unsigned char *uuid) { memcpy(&(ret->uuid[0]), uuid, VIR_UUID_BUFLEN); if (virHashAddEntry(conn->domains, name, ret) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("failed to add domain to connection hash table")); goto error; @@ -299,7 +302,6 @@ virGetDomain(virConnectPtr conn, const char *name, const unsigned char *uuid) { return(ret); error: - virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret); @@ -325,24 +327,28 @@ virReleaseDomain(virDomainPtr domain) { DEBUG("release domain %p %s", domain, domain->name); /* TODO search by UUID first as they are better differenciators */ - if (virHashRemoveEntry(conn->domains, domain->name, NULL) < 0) + if (virHashRemoveEntry(conn->domains, domain->name, NULL) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("domain missing from connection hash table")); + conn = NULL; + } domain->magic = -1; domain->id = -1; VIR_FREE(domain->name); VIR_FREE(domain); - DEBUG("unref connection %p %d", conn, conn->refs); - conn->refs--; - if (conn->refs == 0) { - virReleaseConnect(conn); - /* Already unlocked mutex */ - return; + if (conn) { + DEBUG("unref connection %p %d", conn, conn->refs); + conn->refs--; + if (conn->refs == 0) { + virReleaseConnect(conn); + /* Already unlocked mutex */ + return; + } + virMutexUnlock(&conn->lock); } - - virMutexUnlock(&conn->lock); } @@ -406,11 +412,13 @@ virGetNetwork(virConnectPtr conn, const char *name, const unsigned char *uuid) { /* TODO check the UUID */ if (ret == NULL) { if (VIR_ALLOC(ret) < 0) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->name = strdup(name); if (ret->name == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -420,6 +428,7 @@ virGetNetwork(virConnectPtr conn, const char *name, const unsigned char *uuid) { memcpy(&(ret->uuid[0]), uuid, VIR_UUID_BUFLEN); if (virHashAddEntry(conn->networks, name, ret) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("failed to add network to connection hash table")); goto error; @@ -431,7 +440,6 @@ virGetNetwork(virConnectPtr conn, const char *name, const unsigned char *uuid) { return(ret); error: - virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret); @@ -457,23 +465,27 @@ virReleaseNetwork(virNetworkPtr network) { DEBUG("release network %p %s", network, network->name); /* TODO search by UUID first as they are better differenciators */ - if (virHashRemoveEntry(conn->networks, network->name, NULL) < 0) + if (virHashRemoveEntry(conn->networks, network->name, NULL) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("network missing from connection hash table")); + conn = NULL; + } network->magic = -1; VIR_FREE(network->name); VIR_FREE(network); - DEBUG("unref connection %p %d", conn, conn->refs); - conn->refs--; - if (conn->refs == 0) { - virReleaseConnect(conn); - /* Already unlocked mutex */ - return; + if (conn) { + DEBUG("unref connection %p %d", conn, conn->refs); + conn->refs--; + if (conn->refs == 0) { + virReleaseConnect(conn); + /* Already unlocked mutex */ + return; + } + virMutexUnlock(&conn->lock); } - - virMutexUnlock(&conn->lock); } @@ -714,11 +726,13 @@ virGetStoragePool(virConnectPtr conn, const char *name, const unsigned char *uui /* TODO check the UUID */ if (ret == NULL) { if (VIR_ALLOC(ret) < 0) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->name = strdup(name); if (ret->name == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -728,6 +742,7 @@ virGetStoragePool(virConnectPtr conn, const char *name, const unsigned char *uui memcpy(&(ret->uuid[0]), uuid, VIR_UUID_BUFLEN); if (virHashAddEntry(conn->storagePools, name, ret) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("failed to add storage pool to connection hash table")); goto error; @@ -739,7 +754,6 @@ virGetStoragePool(virConnectPtr conn, const char *name, const unsigned char *uui return(ret); error: - virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret); @@ -766,23 +780,27 @@ virReleaseStoragePool(virStoragePoolPtr pool) { DEBUG("release pool %p %s", pool, pool->name); /* TODO search by UUID first as they are better differenciators */ - if (virHashRemoveEntry(conn->storagePools, pool->name, NULL) < 0) + if (virHashRemoveEntry(conn->storagePools, pool->name, NULL) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("pool missing from connection hash table")); + conn = NULL; + } pool->magic = -1; VIR_FREE(pool->name); VIR_FREE(pool); - DEBUG("unref connection %p %d", conn, conn->refs); - conn->refs--; - if (conn->refs == 0) { - virReleaseConnect(conn); - /* Already unlocked mutex */ - return; + if (conn) { + DEBUG("unref connection %p %d", conn, conn->refs); + conn->refs--; + if (conn->refs == 0) { + virReleaseConnect(conn); + /* Already unlocked mutex */ + return; + } + virMutexUnlock(&conn->lock); } - - virMutexUnlock(&conn->lock); } @@ -845,16 +863,19 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c ret = (virStorageVolPtr) virHashLookup(conn->storageVols, key); if (ret == NULL) { if (VIR_ALLOC(ret) < 0) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->pool = strdup(pool); if (ret->pool == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } ret->name = strdup(name); if (ret->name == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -864,6 +885,7 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c ret->conn = conn; if (virHashAddEntry(conn->storageVols, key, ret) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("failed to add storage vol to connection hash table")); goto error; @@ -875,7 +897,6 @@ virGetStorageVol(virConnectPtr conn, const char *pool, const char *name, const c return(ret); error: - virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret->pool); @@ -903,24 +924,28 @@ virReleaseStorageVol(virStorageVolPtr vol) { DEBUG("release vol %p %s", vol, vol->name); /* TODO search by UUID first as they are better differenciators */ - if (virHashRemoveEntry(conn->storageVols, vol->key, NULL) < 0) + if (virHashRemoveEntry(conn->storageVols, vol->key, NULL) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("vol missing from connection hash table")); + conn = NULL; + } vol->magic = -1; VIR_FREE(vol->name); VIR_FREE(vol->pool); VIR_FREE(vol); - DEBUG("unref connection %p %d", conn, conn->refs); - conn->refs--; - if (conn->refs == 0) { - virReleaseConnect(conn); - /* Already unlocked mutex */ - return; + if (conn) { + DEBUG("unref connection %p %d", conn, conn->refs); + conn->refs--; + if (conn->refs == 0) { + virReleaseConnect(conn); + /* Already unlocked mutex */ + return; + } + virMutexUnlock(&conn->lock); } - - virMutexUnlock(&conn->lock); } @@ -981,7 +1006,8 @@ virGetNodeDevice(virConnectPtr conn, const char *name) ret = (virNodeDevicePtr) virHashLookup(conn->nodeDevices, name); if (ret == NULL) { - if (VIR_ALLOC(ret) < 0) { + if (VIR_ALLOC(ret) < 0) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } @@ -989,11 +1015,13 @@ virGetNodeDevice(virConnectPtr conn, const char *name) ret->conn = conn; ret->name = strdup(name); if (ret->name == NULL) { + virMutexUnlock(&conn->lock); virReportOOMError(conn); goto error; } if (virHashAddEntry(conn->nodeDevices, name, ret) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("failed to add node dev to conn hash table")); goto error; @@ -1005,7 +1033,6 @@ virGetNodeDevice(virConnectPtr conn, const char *name) return(ret); error: - virMutexUnlock(&conn->lock); if (ret != NULL) { VIR_FREE(ret->name); VIR_FREE(ret); @@ -1031,24 +1058,28 @@ virReleaseNodeDevice(virNodeDevicePtr dev) { virConnectPtr conn = dev->conn; DEBUG("release dev %p %s", dev, dev->name); - if (virHashRemoveEntry(conn->nodeDevices, dev->name, NULL) < 0) + if (virHashRemoveEntry(conn->nodeDevices, dev->name, NULL) < 0) { + virMutexUnlock(&conn->lock); virLibConnError(conn, VIR_ERR_INTERNAL_ERROR, _("dev missing from connection hash table")); + conn = NULL; + } dev->magic = -1; VIR_FREE(dev->name); VIR_FREE(dev->parent); VIR_FREE(dev); - DEBUG("unref connection %p %d", conn, conn->refs); - conn->refs--; - if (conn->refs == 0) { - virReleaseConnect(conn); - /* Already unlocked mutex */ - return; + if (conn) { + DEBUG("unref connection %p %d", conn, conn->refs); + conn->refs--; + if (conn->refs == 0) { + virReleaseConnect(conn); + /* Already unlocked mutex */ + return; + } + virMutexUnlock(&conn->lock); } - - virMutexUnlock(&conn->lock); } -- 1.6.0.6

3 2

[libvirt] [PATCH] Rename remaining occurrences of 'interface' to avoid MINGW32 build problem
by Laine Stump 22 Jul '09

22 Jul '09

--- qemud/remote.c | 2 +- src/interface_conf.c | 34 +++++++++++++++++----------------- src/interface_conf.h | 4 ++-- src/veth.c | 6 +++--- src/veth.h | 2 +- 5 files changed, 24 insertions(+), 24 deletions(-) diff --git a/qemud/remote.c b/qemud/remote.c index 651e9fb..d32d513 100644 --- a/qemud/remote.c +++ b/qemud/remote.c @@ -61,7 +61,7 @@ static virDomainPtr get_nonnull_domain (virConnectPtr conn, remote_nonnull_domain domain); static virNetworkPtr get_nonnull_network (virConnectPtr conn, remote_nonnull_network network); -static virInterfacePtr get_nonnull_interface (virConnectPtr conn, remote_nonnull_interface interface); +static virInterfacePtr get_nonnull_interface (virConnectPtr conn, remote_nonnull_interface iface); static virStoragePoolPtr get_nonnull_storage_pool (virConnectPtr conn, remote_nonnull_storage_pool pool); static virStorageVolPtr get_nonnull_storage_vol (virConnectPtr conn, remote_nonnull_storage_vol vol); static void make_nonnull_domain (remote_nonnull_domain *dom_dst, virDomainPtr dom_src); diff --git a/src/interface_conf.c b/src/interface_conf.c index 1cc9f36..c2e8d07 100644 --- a/src/interface_conf.c +++ b/src/interface_conf.c @@ -1207,51 +1207,51 @@ virInterfaceObjPtr virInterfaceAssignDef(virConnectPtr conn, virInterfaceObjListPtr interfaces, const virInterfaceDefPtr def) { - virInterfaceObjPtr interface; + virInterfaceObjPtr iface; - if ((interface = virInterfaceFindByName(interfaces, def->name))) { - if (interface->def) - virInterfaceDefFree(interface->def); - interface->def = def; + if ((iface = virInterfaceFindByName(interfaces, def->name))) { + if (iface->def) + virInterfaceDefFree(iface->def); + iface->def = def; - return interface; + return iface; } - if (VIR_ALLOC(interface) < 0) { + if (VIR_ALLOC(iface) < 0) { virReportOOMError(conn); return NULL; } - if (virMutexInit(&interface->lock) < 0) { + if (virMutexInit(&iface->lock) < 0) { virInterfaceReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", _("cannot initialize mutex")); - VIR_FREE(interface); + VIR_FREE(iface); return NULL; } - virInterfaceObjLock(interface); - interface->def = def; + virInterfaceObjLock(iface); + iface->def = def; if (VIR_REALLOC_N(interfaces->objs, interfaces->count + 1) < 0) { virReportOOMError(conn); - VIR_FREE(interface); + VIR_FREE(iface); return NULL; } - interfaces->objs[interfaces->count] = interface; + interfaces->objs[interfaces->count] = iface; interfaces->count++; - return interface; + return iface; } void virInterfaceRemove(virInterfaceObjListPtr interfaces, - const virInterfaceObjPtr interface) + const virInterfaceObjPtr iface) { unsigned int i; - virInterfaceObjUnlock(interface); + virInterfaceObjUnlock(iface); for (i = 0 ; i < interfaces->count ; i++) { virInterfaceObjLock(interfaces->objs[i]); - if (interfaces->objs[i] == interface) { + if (interfaces->objs[i] == iface) { virInterfaceObjUnlock(interfaces->objs[i]); virInterfaceObjFree(interfaces->objs[i]); diff --git a/src/interface_conf.h b/src/interface_conf.h index aea1208..bb9dce4 100644 --- a/src/interface_conf.h +++ b/src/interface_conf.h @@ -186,14 +186,14 @@ virInterfaceObjPtr virInterfaceFindByName(const virInterfaceObjListPtr void virInterfaceDefFree(virInterfaceDefPtr def); -void virInterfaceObjFree(virInterfaceObjPtr interface); +void virInterfaceObjFree(virInterfaceObjPtr iface); void virInterfaceObjListFree(virInterfaceObjListPtr vms); virInterfaceObjPtr virInterfaceAssignDef(virConnectPtr conn, virInterfaceObjListPtr interfaces, const virInterfaceDefPtr def); void virInterfaceRemove(virInterfaceObjListPtr interfaces, - const virInterfaceObjPtr interface); + const virInterfaceObjPtr iface); virInterfaceDefPtr virInterfaceDefParseString(virConnectPtr conn, const char *xmlStr); diff --git a/src/veth.c b/src/veth.c index 04e68b2..71d7de6 100644 --- a/src/veth.c +++ b/src/veth.c @@ -191,16 +191,16 @@ error_out: * * Returns 0 on success or -1 in case of error */ -int moveInterfaceToNetNs(const char* interface, int pidInNs) +int moveInterfaceToNetNs(const char* iface, int pidInNs) { int rc = -1; char *pid = NULL; const char *argv[] = { - "ip", "link", "set", interface, "netns", NULL, NULL + "ip", "link", "set", iface, "netns", NULL, NULL }; int cmdResult; - if (NULL == interface) { + if (NULL == iface) { goto error_out; } diff --git a/src/veth.h b/src/veth.h index 43181cd..429eb3d 100644 --- a/src/veth.h +++ b/src/veth.h @@ -19,6 +19,6 @@ int vethCreate(char* veth1, int veth1MaxLen, char* veth2, int veth2MaxLen); int vethDelete(const char* veth); int vethInterfaceUpOrDown(const char* veth, int upOrDown); -int moveInterfaceToNetNs(const char *interface, int pidInNs); +int moveInterfaceToNetNs(const char *iface, int pidInNs); #endif /* VETH_H */ -- 1.6.0.6

3 2

[libvirt] managing libvirt users in a one-server context
by Juan Miscaro 22 Jul '09

22 Jul '09

I have a single KVM server on which I want to allow co-workers the ability to create vm's. Here are my criteria in order of importance: 1. I do not want to have to grant all users root privileges. 2. I would like the vm's to use the host's bridge (br0) by default. 3. I do not want one user to be able to start, stop, or (re)define the vm's created by another user by default. 4. I would like a user to be able grant access to another user. Mostly for viewing. Can I do all this with hardcoded usernames and passwords? If not, what is my best option? -- /jm

2 1

[libvirt] Bugfix for failing to reconnect to vbox:///session
by Pritesh Kothari 22 Jul '09

22 Jul '09

Hi All, While reconnecting to vbox:///session, since the VBoxCGlueInit() is not intialized, the g_pfnGetFunctions is NULL and thus reconnection fails with SEGV. Fixed this in the patch below. Regards, Pritesh

3 4

[libvirt] [PATCH 0/1] RFC Fix problems with TLS connections & threading
by Daniel P. Berrange 22 Jul '09

22 Jul '09

Even though we are not using gcrypt directly in any libvirt code, it turns out that we are responsible for initializing its thread support, since its used by gnutls. This really sucks in terms of library API encapsulation. It is also a little annoying if libvirt is used within an app that already uses gnutls/gcrypt, because there is no checking in gcrypt to see if threading ops are already set, so our call would override any pre-existing initialization, and an apps own call would override us. The only other option I see is to punt on the whole problem and declare apps are responsible for initializing gcrypt, but then this sucks too because it means apps have to presume we're built with gnutls, and not nss - not that we support nss currently. This bug only really impacts the libvirt client if using multiple virConnectPtr objects from separate threads with a TLS based URI. The server isn't impacted because all TLS I/O is done from the single I/O thread, and not any of the RPC workers. Daniel P. Berrange (1): Initialize gcrypt threading src/remote_internal.c | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 43 insertions(+), 0 deletions(-)

1 1

[libvirt] [PATCH] Support physical memory in virDomainMemoryPeek()
by Nguyen Anh Quynh 22 Jul '09

22 Jul '09

Hi, This patch provides support for physical memory in virDomainMemoryPeek(). Please consider applying. Signed-off-by: Nguyen Anh Quynh <aquynh(a)gmail.com> # diffstat pmemsave3.diff docs/libvirt-api.xml | 2 +- include/libvirt/libvirt.h.in | 1 + src/libvirt.c | 14 +++++--------- src/qemu_driver.c | 17 ++++++++--------- 4 files changed, 15 insertions(+), 19 deletions(-)

3 8

[libvirt] [PATCH 00/14] Add support for qemu NIC hotplug
by Mark McLoughlin 22 Jul '09

22 Jul '09

Hey, Here's a series of patches to (mostly) support NIC hotplug with qemu 0.10.x or later. qemu-0.10.0 introduced the host_net_add monitor command without which we cannot do NIC hotplug, so we do not try and support hotplug with older versions of KVM. With qemu-0.11.0 we will be able to pass a tap file descriptor over the monitor, but until then we cannot support 'bridge' and 'network' type interfaces. A forthcoming patch series will rectify that. Cheers, Mark.

3 60

[libvirt] fix "make distcheck" failures
by Jim Meyering 22 Jul '09

22 Jul '09

Running "make check" from a tarball would show two test failures, since some test-related files were not being distributed. [that is one of the things that "make distcheck" tests for] With these two fixes, "make distcheck" passes once again: >From 553cecaf27e86755e40bab346d4014fb0291e5cd Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 22 Jul 2009 05:07:07 -0400 Subject: [PATCH 1/2] avoid a make distcheck failure: distribute tests/interfaceschemadata/ * tests/Makefile.am (EXTRA_DIST): Add interfaceschemadata. --- tests/Makefile.am | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index efd7c6e..74e98d1 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -53,6 +53,7 @@ EXTRA_DIST = \ networkschematest \ domainschematest \ domainschemadata \ + interfaceschemadata \ storagepoolschematest \ storagepoolschemadata \ storagevolschematest \ -- 1.6.2.5 >From 76f208cc5235023c594e84783af7a33383c8f649 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 22 Jul 2009 05:10:14 -0400 Subject: [PATCH 2/2] avoid a make distcheck failure: distribute docs/schemas/interface.rng * docs/schemas/Makefile.am (schema_DATA): Add interface.rng. --- docs/schemas/Makefile.am | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/docs/schemas/Makefile.am b/docs/schemas/Makefile.am index c4ef0b1..ef41a63 100644 --- a/docs/schemas/Makefile.am +++ b/docs/schemas/Makefile.am @@ -3,6 +3,7 @@ schemadir = $(pkgdatadir)/schemas schema_DATA = \ domain.rng \ + interface.rng \ network.rng \ storagepool.rng \ storagevol.rng \ -- 1.6.2.5

2 1

[libvirt] [PATCH] libvirt-glib: import gtk instead of pygtk
by Guido Günther 22 Jul '09

22 Jul '09

Hi, the python example imports pygtk but uses gtk.main() so it fails to work. Attached patch fixes this. Cheers, -- Guido

1 0

[libvirt] c&p error in api doc
by Thomas Treutner 22 Jul '09

22 Jul '09

I think there's a c&p error: http://libvirt.org/html/libvirt-libvirt.html#virStoragePoolUndefine "Returns: a virStoragePoolPtr object, or NULL if creation failed" I assume it is "0 on success, -1 on failure" as the return is of type int. kr,t

3 4

[libvirt] Problem with libvirt and pci passthrough
by Mirko Raasch 21 Jul '09

21 Jul '09

Hi, i´m still having problem with libvirt and pci passthrough, if i use more then one pci card in the guest. If i start qemu 0.10.5 manualy, i can use all of the pci devices in the guest, here is the log: http://nopaste.info/610fa81fb5.html If i use libvirt to start the guest and passthrough three pci device, none of the device is usable in the guest: http://nopaste.info/ccfbfcf9a0.html Is there any limitation in libvirt? Greetings Mirko

2 6

[libvirt] [PATCH] Make tcp_chr_read() use recvmsg()
by Mark McLoughlin 21 Jul '09

21 Jul '09

Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> --- qemu-char.c | 12 +++++++++++- 1 files changed, 11 insertions(+), 1 deletions(-) diff --git a/qemu-char.c b/qemu-char.c index 287e0cd..e0d7220 100644 --- a/qemu-char.c +++ b/qemu-char.c @@ -1911,15 +1911,25 @@ static void tcp_chr_read(void *opaque) { CharDriverState *chr = opaque; TCPCharDriver *s = chr->opaque; + struct msghdr msg = { 0, }; + struct iovec iov[1]; uint8_t buf[1024]; int len, size; if (!s->connected || s->max_size <= 0) return; + len = sizeof(buf); if (len > s->max_size) len = s->max_size; - size = recv(s->fd, (void *)buf, len, 0); + + iov[0].iov_base = buf; + iov[0].iov_len = len; + + msg.msg_iov = iov; + msg.msg_iovlen = 1; + + size = recvmsg(s->fd, &msg, 0); if (size == 0) { /* connection closed */ s->connected = 0; -- 1.6.2.5

1 1

[libvirt] [PATCH] Add support for fd=name to tap and socket networking
by Mark McLoughlin 21 Jul '09

21 Jul '09

This allows a program to initialize a host networking device using a file descriptor passed over a unix monitor socket. The program must first pass the file descriptor using SCM_RIGHTS ancillary data with the getfd monitor command. It then may do "host_net_add tap fd=name" to use the named file descriptor. Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> --- net.c | 42 +++++++++++++++++++++++++++++++++++------- 1 files changed, 35 insertions(+), 7 deletions(-) diff --git a/net.c b/net.c index 90cf912..1838692 100644 --- a/net.c +++ b/net.c @@ -2410,6 +2410,23 @@ void qemu_check_nic_model_list(NICInfo *nd, const char * const *models, exit(exit_status); } +static int net_handle_fd_param(Monitor *mon, const char *param) +{ + if (!qemu_isdigit(param[0])) { + int fd; + + fd = monitor_get_fd(mon, param); + if (fd == -1) { + config_error(mon, "No file descriptor named %s found", param); + return -1; + } + + return fd; + } else { + return strtol(param, NULL, 0); + } +} + int net_client_init(Monitor *mon, const char *device, const char *p) { char buf[1024]; @@ -2650,14 +2667,20 @@ int net_client_init(Monitor *mon, const char *device, const char *p) static const char * const fd_params[] = { "vlan", "name", "fd", "sndbuf", NULL }; + ret = -1; if (check_params(chkbuf, sizeof(chkbuf), fd_params, p) < 0) { config_error(mon, "invalid parameter '%s' in '%s'\n", chkbuf, p); - ret = -1; goto out; } - fd = strtol(buf, NULL, 0); + fd = net_handle_fd_param(mon, buf); + if (fd == -1) { + goto out; + } fcntl(fd, F_SETFL, O_NONBLOCK); s = net_tap_fd_init(vlan, device, name, fd); + if (!s) { + close(fd); + } } else { static const char * const tap_params[] = { "vlan", "name", "ifname", "script", "downscript", "sndbuf", NULL @@ -2697,15 +2720,20 @@ int net_client_init(Monitor *mon, const char *device, const char *p) "vlan", "name", "fd", NULL }; int fd; + ret = -1; if (check_params(chkbuf, sizeof(chkbuf), fd_params, p) < 0) { config_error(mon, "invalid parameter '%s' in '%s'\n", chkbuf, p); - ret = -1; goto out; } - fd = strtol(buf, NULL, 0); - ret = -1; - if (net_socket_fd_init(vlan, device, name, fd, 1)) - ret = 0; + fd = net_handle_fd_param(mon, buf); + if (fd == -1) { + goto out; + } + if (!net_socket_fd_init(vlan, device, name, fd, 1)) { + close(fd); + goto out; + } + ret = 0; } else if (get_param_value(buf, sizeof(buf), "listen", p) > 0) { static const char * const listen_params[] = { "vlan", "name", "listen", NULL -- 1.6.2.5

1 0

[libvirt] [PATCH] Add monitor_get_fd() command for fetching named fds
by Mark McLoughlin 21 Jul '09

21 Jul '09

Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> --- monitor.c | 24 ++++++++++++++++++++++++ monitor.h | 2 ++ 2 files changed, 26 insertions(+), 0 deletions(-) diff --git a/monitor.c b/monitor.c index 817e4b7..6ad2e14 100644 --- a/monitor.c +++ b/monitor.c @@ -1774,6 +1774,30 @@ static void do_closefd(Monitor *mon, const char *fdname) fdname); } +int monitor_get_fd(Monitor *mon, const char *fdname) +{ + mon_fd_t *monfd; + + LIST_FOREACH(monfd, &mon->fds, next) { + int fd; + + if (strcmp(monfd->name, fdname) != 0) { + continue; + } + + fd = monfd->fd; + + /* caller takes ownership of fd */ + LIST_REMOVE(monfd, next); + qemu_free(monfd->name); + qemu_free(monfd); + + return fd; + } + + return -1; +} + static const mon_cmd_t mon_cmds[] = { #include "qemu-monitor.h" { NULL, NULL, }, diff --git a/monitor.h b/monitor.h index 13e8cc7..f6a43c0 100644 --- a/monitor.h +++ b/monitor.h @@ -20,6 +20,8 @@ void monitor_read_bdrv_key_start(Monitor *mon, BlockDriverState *bs, BlockDriverCompletionFunc *completion_cb, void *opaque); +int monitor_get_fd(Monitor *mon, const char *fdname); + void monitor_vprintf(Monitor *mon, const char *fmt, va_list ap); void monitor_printf(Monitor *mon, const char *fmt, ...) __attribute__ ((__format__ (__printf__, 2, 3))); -- 1.6.2.5

1 0

[libvirt] [PATCH] Add getfd and closefd monitor commands
by Mark McLoughlin 21 Jul '09

21 Jul '09

Add monitor commands to support passing file descriptors via SCM_RIGHTS. getfd assigns the passed file descriptor a name for use with other monitor commands. closefd allows passed file descriptors to be closed. If a monitor command actually uses a named file descriptor, closefd will not be required. Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> --- monitor.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ qemu-monitor.hx | 18 ++++++++++++++ 2 files changed, 87 insertions(+), 0 deletions(-) diff --git a/monitor.c b/monitor.c index e38c86e..817e4b7 100644 --- a/monitor.c +++ b/monitor.c @@ -70,6 +70,14 @@ typedef struct mon_cmd_t { const char *help; } mon_cmd_t; +/* file descriptors passed via SCM_RIGHTS */ +typedef struct mon_fd_t mon_fd_t; +struct mon_fd_t { + char *name; + int fd; + LIST_ENTRY(mon_fd_t) next; +}; + struct Monitor { CharDriverState *chr; int flags; @@ -80,6 +88,7 @@ struct Monitor { CPUState *mon_cpu; BlockDriverCompletionFunc *password_completion_cb; void *password_opaque; + LIST_HEAD(,mon_fd_t) fds; LIST_ENTRY(Monitor) entry; }; @@ -1705,6 +1714,66 @@ static void do_inject_mce(Monitor *mon, } #endif +static void do_getfd(Monitor *mon, const char *fdname) +{ + mon_fd_t *monfd; + int fd; + + fd = qemu_chr_get_msgfd(mon->chr); + if (fd == -1) { + monitor_printf(mon, "getfd: no file descriptor supplied via SCM_RIGHTS\n"); + return; + } + + if (qemu_isdigit(fdname[0])) { + monitor_printf(mon, "getfd: monitor names may not begin with a number\n"); + return; + } + + fd = dup(fd); + if (fd == -1) { + monitor_printf(mon, "Failed to dup() file descriptor: %s\n", + strerror(errno)); + return; + } + + LIST_FOREACH(monfd, &mon->fds, next) { + if (strcmp(monfd->name, fdname) != 0) { + continue; + } + + close(monfd->fd); + monfd->fd = fd; + return; + } + + monfd = qemu_mallocz(sizeof(mon_fd_t)); + monfd->name = qemu_strdup(fdname); + monfd->fd = fd; + + LIST_INSERT_HEAD(&mon->fds, monfd, next); +} + +static void do_closefd(Monitor *mon, const char *fdname) +{ + mon_fd_t *monfd; + + LIST_FOREACH(monfd, &mon->fds, next) { + if (strcmp(monfd->name, fdname) != 0) { + continue; + } + + LIST_REMOVE(monfd, next); + close(monfd->fd); + qemu_free(monfd->name); + qemu_free(monfd); + return; + } + + monitor_printf(mon, "Failed to find file descriptor named %s\n", + fdname); +} + static const mon_cmd_t mon_cmds[] = { #include "qemu-monitor.h" { NULL, NULL, }, diff --git a/qemu-monitor.hx b/qemu-monitor.hx index 70e2475..11bdb2c 100644 --- a/qemu-monitor.hx +++ b/qemu-monitor.hx @@ -628,6 +628,24 @@ STEXI Inject an MCE on the given CPU (x86 only). ETEXI + { "getfd", "s", do_getfd, "getfd name", + "receive a file descriptor via SCM rights and assign it a name" }, +STEXI +@item getfd @var{fdname} +If a file descriptor is passed alongside this command using the SCM_RIGHTS +mechanism on unix sockets, it is stored using the name @var{fdname} for +later use by other monitor commands. +ETEXI + + { "closefd", "s", do_closefd, "closefd name", + "close a file descriptor previously passed via SCM rights" }, +STEXI +@item closefd @var{fdname} +Close the file descriptor previously assigned to @var{fdname} using the +@code{getfd} command. This is only needed if the file descriptor was never +used by another monitor command. +ETEXI + STEXI @end table ETEXI -- 1.6.2.5

1 0

[libvirt] [PATCH] Add SCM_RIGHTS support to unix socket character devices
by Mark McLoughlin 21 Jul '09

21 Jul '09

If a file descriptor is passed via a message with SCM_RIGHTS ancillary data on a unix socket, store the file descriptor for use in the chr_read() handler. Close the file descriptor if it was not used. The qemu_chr_get_msgfd() API provides access to the passed descriptor. Signed-off-by: Mark McLoughlin <markmc(a)redhat.com> --- qemu-char.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ qemu-char.h | 2 ++ 2 files changed, 52 insertions(+), 0 deletions(-) diff --git a/qemu-char.c b/qemu-char.c index e0d7220..f06a614 100644 --- a/qemu-char.c +++ b/qemu-char.c @@ -168,6 +168,11 @@ void qemu_chr_read(CharDriverState *s, uint8_t *buf, int len) s->chr_read(s->handler_opaque, buf, len); } +int qemu_chr_get_msgfd(CharDriverState *s) +{ + return s->get_msgfd ? s->get_msgfd(s) : -1; +} + void qemu_chr_accept_input(CharDriverState *s) { if (s->chr_accept_input) @@ -1832,6 +1837,7 @@ typedef struct { int do_telnetopt; int do_nodelay; int is_unix; + int msgfd; } TCPCharDriver; static void tcp_chr_accept(void *opaque); @@ -1907,12 +1913,46 @@ static void tcp_chr_process_IAC_bytes(CharDriverState *chr, *size = j; } +static int tcp_get_msgfd(CharDriverState *chr) +{ + TCPCharDriver *s = chr->opaque; + + return s->msgfd; +} + +static void unix_process_msgfd(CharDriverState *chr, struct msghdr *msg) +{ + TCPCharDriver *s = chr->opaque; + struct cmsghdr *cmsg; + + for (cmsg = CMSG_FIRSTHDR(msg); cmsg; cmsg = CMSG_NXTHDR(msg, cmsg)) { + int fd; + + if (cmsg->cmsg_len != CMSG_LEN(sizeof(int)) || + cmsg->cmsg_level != SOL_SOCKET || + cmsg->cmsg_type != SCM_RIGHTS) + continue; + + fd = *((int *)CMSG_DATA(cmsg)); + if (fd < 0) + continue; + + if (s->msgfd != -1) + close(s->msgfd); + s->msgfd = fd; + } +} + static void tcp_chr_read(void *opaque) { CharDriverState *chr = opaque; TCPCharDriver *s = chr->opaque; struct msghdr msg = { 0, }; struct iovec iov[1]; + union { + struct cmsghdr cmsg; + char control[CMSG_SPACE(sizeof(int))]; + } msg_control; uint8_t buf[1024]; int len, size; @@ -1928,6 +1968,8 @@ static void tcp_chr_read(void *opaque) msg.msg_iov = iov; msg.msg_iovlen = 1; + msg.msg_control = &msg_control; + msg.msg_controllen = sizeof(msg_control); size = recvmsg(s->fd, &msg, 0); if (size == 0) { @@ -1940,10 +1982,16 @@ static void tcp_chr_read(void *opaque) closesocket(s->fd); s->fd = -1; } else if (size > 0) { + if (s->is_unix) + unix_process_msgfd(chr, &msg); if (s->do_telnetopt) tcp_chr_process_IAC_bytes(chr, s, buf, &size); if (size > 0) qemu_chr_read(chr, buf, size); + if (s->msgfd != -1) { + close(s->msgfd); + s->msgfd = -1; + } } } @@ -2105,12 +2153,14 @@ static CharDriverState *qemu_chr_open_tcp(const char *host_str, s->connected = 0; s->fd = -1; s->listen_fd = -1; + s->msgfd = -1; s->is_unix = is_unix; s->do_nodelay = do_nodelay && !is_unix; chr->opaque = s; chr->chr_write = tcp_chr_write; chr->chr_close = tcp_chr_close; + chr->get_msgfd = tcp_get_msgfd; if (is_listen) { s->listen_fd = fd; diff --git a/qemu-char.h b/qemu-char.h index e1aa8db..77d4eda 100644 --- a/qemu-char.h +++ b/qemu-char.h @@ -51,6 +51,7 @@ struct CharDriverState { int (*chr_write)(struct CharDriverState *s, const uint8_t *buf, int len); void (*chr_update_read_handler)(struct CharDriverState *s); int (*chr_ioctl)(struct CharDriverState *s, int cmd, void *arg); + int (*get_msgfd)(struct CharDriverState *s); IOEventHandler *chr_event; IOCanRWHandler *chr_can_read; IOReadHandler *chr_read; @@ -81,6 +82,7 @@ void qemu_chr_reset(CharDriverState *s); void qemu_chr_initial_reset(void); int qemu_chr_can_read(CharDriverState *s); void qemu_chr_read(CharDriverState *s, uint8_t *buf, int len); +int qemu_chr_get_msgfd(CharDriverState *s); void qemu_chr_accept_input(CharDriverState *s); void qemu_chr_info(Monitor *mon); -- 1.6.2.5

1 0

[libvirt] libvirt-java, deprecated methods?
by Thomas Treutner 21 Jul '09

21 Jul '09

Hi, I'm playing around with libvirt-0.6.5 and libvirt-java-0.2.1 these days. It looks very interesting, but I think I may have found a bug: .... conn = new Connect("xen://node02", false); .... conn.setDom0Memory(512000) gives: exception caught:org.libvirt.LibvirtException: invalid domain pointer in virDomainSetMemory level:VIR_ERR_ERROR code:VIR_ERR_INVALID_DOMAIN domain:VIR_FROM_DOM hasConn:false hasDom:false hasNet:false message:invalid domain pointer in virDomainSetMemory str1:invalid domain pointer in %s str2:virDomainSetMemory str3:null int1:0 int2:0 Furthermore, f.e. conn.GetHypervisorVersion(conn.getType()) gives: exception caught:org.libvirt.LibvirtException: this function is not supported by the hypervisor: Xen level:VIR_ERR_ERROR code:VIR_ERR_NO_SUPPORT domain:VIR_FROM_NONE hasConn:false hasDom:false hasNet:false message:this function is not supported by the hypervisor: Xen str1:this function is not supported by the hypervisor: %s str2:Xen str3:null int1:0 int2:0 But Domain dom0=conn.domainLookupByID(0); //domainLookupByName("Domain-0"); too DomainInfo dom0info = dom0.getInfo(); .... dom0.setMemory(256000); .... works without glitches and does the job (btw, I can't see any effect of setMaxMemory() on Xen?) Is it possible that there are some inconsistences between libvirt-java and libvirt? Since libvirt-java wasn't updated approx. 1y, I assume that. Thanks for feedback, kr, thomas

3 5

[libvirt] [PATCH] Extra filesystem options during backend filesystem mount.
by Harshavardhana 21 Jul '09

21 Jul '09

New option index added to support -o options for various netfs. Currently added an option for glusterfs. --- src/storage_backend_fs.c | 32 ++++++++++++++++++++++++++++++++ 1 files changed, 32 insertions(+), 0 deletions(-) diff --git a/src/storage_backend_fs.c b/src/storage_backend_fs.c index c3d66b5..606fb47 100644 --- a/src/storage_backend_fs.c +++ b/src/storage_backend_fs.c @@ -604,6 +604,7 @@ static int virStorageBackendFileSystemMount(virConnectPtr conn, virStoragePoolObjPtr pool) { char *src; + char *options; const char **mntargv; /* 'mount -t auto' doesn't seem to auto determine nfs (or cifs), @@ -611,6 +612,10 @@ virStorageBackendFileSystemMount(virConnectPtr conn, * accommodate this */ int netauto = (pool->def->type == VIR_STORAGE_POOL_NETFS && pool->def->source.format == VIR_STORAGE_POOL_NETFS_AUTO); + int glusterfs = (pool->def->type == VIR_STORAGE_POOL_NETFS && + pool->def->source.format == VIR_STORAGE_POOL_NETFS_GLUSTERFS); + + int option_index; int source_index; const char *netfs_auto_argv[] = { @@ -632,9 +637,26 @@ virStorageBackendFileSystemMount(virConnectPtr conn, NULL, }; + const char *glusterfs_argv[] = { + MOUNT, + "-t", + pool->def->type == VIR_STORAGE_POOL_FS ? + virStoragePoolFormatFileSystemTypeToString(pool->def->source.format) : + virStoragePoolFormatFileSystemNetTypeToString(pool->def->source.format), + NULL, + "-o", + NULL, + pool->def->target.path, + NULL, + }; + if (netauto) { mntargv = netfs_auto_argv; source_index = 1; + } else if (glusterfs) { + mntargv = glusterfs_argv; + source_index = 3; + option_index = 5; } else { mntargv = fs_argv; source_index = 3; @@ -670,6 +692,12 @@ virStorageBackendFileSystemMount(virConnectPtr conn, } if (pool->def->type == VIR_STORAGE_POOL_NETFS) { + if (pool->def->source.format = VIR_STORAGE_POOL_NETFS_GLUSTERFS) { + if (virAsprintf(&options, "direct-io-mode=1") == -1) { + virReportOOMError(conn); + return -1; + } + } if (virAsprintf(&src, "%s:%s", pool->def->source.host.name, pool->def->source.dir) == -1) { @@ -685,6 +713,10 @@ virStorageBackendFileSystemMount(virConnectPtr conn, } mntargv[source_index] = src; + if (glusterfs) { + mntargv[option_index] = options; + } + if (virRun(conn, mntargv, NULL) < 0) { VIR_FREE(src); return -1; -- 1.6.0.6

3 6

[libvirt] Workaround to start xen inside a buggy kvm
by Thomas Fricke 21 Jul '09

21 Jul '09

Buggy kvm crashes with kernel panic if xen started inside a kvm machine. kvm needs the -no-kvm-pit, this argument is not available with libvirt. I have tested the following workaround on ubuntu 9.04 Rename kvm cd /usr/bin mv kvm kvm.orig Use the following script instead the original kvm #!/bin/sh exec kvm.orig -no-kvm-pit $* Do not forget chmod +x kvm Have fun, Thomas

1 0

[libvirt] [PATCH 0/6] virInterface bug fixes, test driver, and netcf backend driver
by Laine Stump 21 Jul '09

21 Jul '09

This series of patches first corrects a couple of bugs/deficiencies in existing virInterface code, then adds test and netcf-based backend interface drivers.

3 25

[libvirt] [patch] initialize a local variable in qemudOpenMonitorUnix()
by Jun Koi 21 Jul '09

21 Jul '09

This patch initializes a local variable in qemudOpenMonitorUnix(), thus also eliminates a compilation warning. diff --git a/src/qemu_driver.c b/src/qemu_driver.c index 00dc6e5..d2db1a2 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -874,7 +874,7 @@ qemudOpenMonitorUnix(virConnectPtr conn, struct sockaddr_un addr; int monfd; int timeout = 3; /* In seconds */ - int ret, i; + int ret, i = 0; if ((monfd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) { virReportSystemError(conn, errno,

2 1

[libvirt] [PATCH] rpm spec cleanup and split off client only requirements
by Daniel Veillard 21 Jul '09

21 Jul '09

This early patch changes the spec file to: - remove the old Obsoletes: livir and such, that was 3 years ago - make sure the inter packages require %{version}-%{release} and not just %{version} - compresses the ChangeLog which is getting close to a megabyte - split out a separate libvirt-client binary rpm including only what is needed to access the libvirt service from a client viewpioint. The main package requires the client anyway So I moved in the client: - the shared library, I guess it's uncontroversial - the virsh/virt-xml-validate binaries, I guess it it doesn't block i386/x86_64 to be parallel installable that's fine - the man pages, obvious based on previous - the proxy when built, more controversial, it assumes that there is a service running locally so could be left on the main package - sasl and policykit data, more controversial, I guess it's better to associate them with the client - the schemas to allow checking on the client Some of this is still open, including the name of the subpackage which I initially expected to be libvirt-libs but in retrospect maybe -client will be more in line with the content, Comments welcome :-) Daniel -- 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/

2 2

[libvirt] [PATCH 0/7] Support cgroups in QEMU driver
by Daniel P. Berrange 20 Jul '09

20 Jul '09

This series of patches introduces cgroups support to the QEMU driver. At this time it uses the 'devices' controller to whitelist block devices for QEMU guests to prevent unauthorized access. It uses the 'cpu_shares' controller to allow schedular tunables on a per guest basis. In the future we should use the 'memory' controller to enforce the limit set in the balloon driver ie if the guest does not honour the balloon request, then force the guest into swap, avoiding host overcommit. There is also some refactoring of the cgroups code to remove the assumption that libvirtd is starting in the root cgroup, remove the requirement that all controllers be active, and allow for use in non-privileged drivers. Daniel P. Berrange (7): Use enums for cgroup controller types / labels Use virFileReadAll/virFileWriteStr for key cgroup read/write helpers Make cgroups a little more efficient Place every QEMU guest in a private cgroup Implement schedular tunables API using cgroups Use cgroups for block device whitelisting in QEMU guests Refactor cgroups to allow a group per driver to be managed directly src/cgroup.c | 861 ++++++++++++++++++++++++++-------------------- src/cgroup.h | 28 +- src/libvirt_private.syms | 1 + src/lxc_conf.h | 2 + src/lxc_controller.c | 19 +- src/lxc_driver.c | 25 +- src/qemu_conf.h | 2 + src/qemu_driver.c | 355 ++++++++++++++++++- src/util.c | 27 ++- src/util.h | 2 + 10 files changed, 904 insertions(+), 420 deletions(-)

2 17

[libvirt] changing files in /var/lib/libvirt/images
by Gene Czarcinski 20 Jul '09

20 Jul '09

Assume I am working with virtual disk image files in /var/lib/libvirt/images such as removing a file, defining a new file (with qemi-img create or qemu-img convert), or simply copying a file ("cp") from elsewhere such as VMware. When I do this, Guest definition or adding hardware to an existing guest under virt- manager does not "see" the file change. That is, until I terminate virt- manager and restart /etc/init.d/libvirtd. Is there some better/easier/less-disruptive means of notifying "whatever" that the list of files has changed? I noticed that /etc/init.d/libvirtd force-reload has no effect. I also notice that restarting libvirtd and/or terminating and restarting virt- manager while there are guests running really screws things up. If such a "means" does not exist, should one be created? [should this be an RFE in bugzilla?] Gene

2 2

[libvirt] Create a new VM from an existent image?
by Jun Koi 19 Jul '09

19 Jul '09

Hi, To use libvirt, I am strying to use virt-install to create a new VM from an existent (KVM) VM image. I did the following steps: 1) run libvirtd with "libvirtd -d" (I use the latest libvirt git tree) 2) Run virt-install (0.400.3) with my existent image in img/img.winxp # virt-install --import -n winxp -f img/img.xp3 --vnc --hvm --accelerate --ram 800 However, this step returns error like: "ERROR Could not find usable default libvirt connection." Actually I am not sure that is the right way to use the new --import option with virt-install. So that might be the problem? All the code are updated, and compiled from source on Ubuntu 8.04. Any help? Thanks a lot, J

2 4

[libvirt] [PATCH] qemu: Try multiple times to open unix monitor socket
by Cole Robinson 17 Jul '09

17 Jul '09

Unlike the pty monitor (which we know exists since we scrape its path from stdout), we have no way of knowing that the unix monitor socket should exist/ be initialized. As a result, some of my KVM guests randomly fail to start on F10 host. Try to open the unix socket in a 3 second timeout loop. Ignore EACCES (path does not exist if a first time run) and ECONNREFUSED (leftover socket from a previous run hasn't been removed yet). Fixes things for me. --- src/qemu_driver.c | 22 +++++++++++++++++++++- 1 files changed, 21 insertions(+), 1 deletions(-) diff --git a/src/qemu_driver.c b/src/qemu_driver.c index e2b7acb..200718b 100644 --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -874,6 +874,8 @@ qemudOpenMonitorUnix(virConnectPtr conn, { struct sockaddr_un addr; int monfd; + int timeout = 3; /* In seconds */ + int ret, i; if ((monfd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) { virReportSystemError(conn, errno, @@ -885,10 +887,28 @@ qemudOpenMonitorUnix(virConnectPtr conn, addr.sun_family = AF_UNIX; strncpy(addr.sun_path, monitor, sizeof(addr.sun_path)); - if (connect(monfd, (struct sockaddr *) &addr, sizeof(addr)) < 0) { + do { + ret = connect(monfd, (struct sockaddr *) &addr, sizeof(addr)); + + if (ret == 0) + break; + + if (errno == EACCES || errno == ECONNREFUSED) { + /* EACCES : Socket may not have shown up yet + * ECONNREFUSED : Leftover socket hasn't been removed yet */ + continue; + } + virReportSystemError(conn, errno, "%s", _("failed to connect to monitor socket")); goto error; + + } while ((++i <= timeout*5) && (usleep(.2 * 1000000) <= 0)); + + if (ret != 0) { + virReportSystemError(conn, errno, "%s", + _("monitor socket did not show up.")); + goto error; } if (qemudOpenMonitorCommon(conn, driver, vm, monfd, reconnect) < 0) -- 1.6.0.6

4 5

[libvirt] adding scsi disk failed (code=9)
by Massimo Canonico 17 Jul '09

17 Jul '09

Hi, I'm newbie and sorry if this question was already post in this ML (is there any search engine for the ML archive?). I'm trying to attach a volume in my instance. The instance is running and the volume is available, but i got this error: ---- libvirt: operation failed: adding scsi disk failed (code=9) virDomainAttachDevice() failed (err=-1) XML=<disk type='block'><driver name='phy'/><source dev='/dev/etherd/e0.8'/><target dev='sdb'/></disk> doAttachVolume() failed error=1 ---- I'm not sure where I can start to debug this error. I'm using Fedora 11, x86_64, with these packages [root@minicloud eucalyptus]# rpm -qa | grep libv libvirt-devel-0.6.2-12.fc11.x86_64 libvorbis-1.2.0-7.fc11.x86_64 libvirt-0.6.2-12.fc11.x86_64 libvirt-python-0.6.2-12.fc11.x86_64 libvolume_id-141-3.fc11.x86_64 May you help me, please? Thanks in advance, M

4 5

[libvirt] [libvirt 0.6.4] Unable to set /domain/devices/serial[@port=1] as console?
by Charles Duffy 17 Jul '09

17 Jul '09

When I do a "virsh define" with the following: <serial type='file'> <source path='/path/to/serial.log'/> <target port='0'/> </serial> <serial type='pty'> <target port='1'/> </serial> <console type='pty'> <target port='1'/> </console> The result is this: <serial type='file'> <source path='/path/to/serial.log'/> <target port='0'/> </serial> <serial type='pty'> <target port='1'/> </serial> <console type='file'> <source path='/path/to/serial.log'/> <target port='0'/> </console> ...a markedly different configuration. Is this the result of a bug, or a known behavior I've failed to catch in the docs? Thanks!

2 1

[libvirt] Tentative schedule for 0.7.0
by Daniel Veillard 17 Jul '09

17 Jul '09

Basically I would like to freeze development at the end of next week and push at the end of the month. Since I plan to be away on 31, I will probably make the release on the 30th, but I think we can still apply new features until Friday next week. To try to help w.r.t. Fedora 12 schedule I will make a 0.7.0 beta release probably on the week-end of the 25 and push it in rawhide for parallel testing, Daniel -- 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/

1 0

[libvirt] One line logging fix
by Dave Allan 17 Jul '09

17 Jul '09

I noticed that I was incorrectly logging a debug message as an error in a success path in the node device code. Here's the fix. >From 0c65b2d1d90bf0d469ffe6d8a20a5566b359a61e Mon Sep 17 00:00:00 2001 From: David Allan <dallan(a)redhat.com> Date: Fri, 17 Jul 2009 09:57:40 -0400 Subject: [PATCH 1/1] Remove spurious log message A success path in part of the node device code contained a log message with severity ERROR; this fix makes it DEBUG. --- src/node_device_hal_linux.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/node_device_hal_linux.c b/src/node_device_hal_linux.c index ddb1eef..a119404 100644 --- a/src/node_device_hal_linux.c +++ b/src/node_device_hal_linux.c @@ -50,7 +50,7 @@ static int open_wwn_file(const char *prefix, /* fd will be closed by caller */ if ((*fd = open(wwn_path, O_RDONLY)) != -1) { - VIR_ERROR(_("Opened WWN path '%s' for reading"), + VIR_DEBUG(_("Opened WWN path '%s' for reading"), wwn_path); } else { VIR_ERROR(_("Failed to open WWN path '%s' for reading"), -- 1.6.0.6

2 1

[libvirt] Guest shutdown script
by Ján ONDREJ (SAL) 17 Jul '09

17 Jul '09

Hello, when host server is shutting down, it's goot to shutdown also guests, not only kill them. Host shutdown can be touched by an admin, but also can be started by an UPS daemon or similar action. Then kill of guests is not a best solution. I have my own script, which can shutdown guest automatically: http://www.salstar.sk/pub/salpack/etc/rc.d/init.d/virt-shutdown You need to start this rc script after installation, otherwise it will not do anything. Another init script with save/restore ability is here: http://marc.info/?l=kvm&m=124775655529836&w=4 I'm not using this so I don't know, if this will work well on Fedora. A good solution can also be, if libvirt init script can shutdown guests autoamtically. It starts them, when system is booting, but does not stop them. I know, that libvirt sometimes need to be restarted without touching guests, but you also can avoid this, if you shutdown guest only if one of: - current runlevel is 0 (halt) or 6 (reboot). You can get current runlevel by running: runlevel | cut -d" " -f2 - if "service libvirt shutdown" is called, which is not very systematic, but can work well Just please do not create a new package only for guest shutdowns. :-) I am not a member of this list, so please cc me. SAL

1 0

[libvirt] How to pass not xml parameters from Libvirt to Xen in xml definition
by Alexandre CABROL PERALES 17 Jul '09

17 Jul '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi everybody, i need to define cpuid parameter for a specific Xen Virtual Machine. This parameters doesn't exist in libvirt definition. So how can i add this "unknow" parameter in virtual machine's xml definition to be able to manage it from virt-manager? regards, - -- Alexandre CABROL PERALES - -- Ingenieur Securite des Systemes d'Information Mob. : 06.98.82.03.06 Mail : alexandre.cabrol(a)artal.fr Key fingerprint = 1E6B B8DF 5001 A6A8 E057 9D31 7B3B EAB1 4AE4 8953 - -- ARTAL Technologies Rue Pierre-Gilles de Gennes Ens."La Rue", Bat. 9, BP 38138 31681 Labege cedex Tel. : 05.61.00.39.30 Fax : 05.61.00.20.43 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpgSWsACgkQezvqsUrkiVMqiwCgkigm/WZ5TPToQ7RzdY0R6UJz tEAAoKIcdZ2fO5PdqZdcrzuTj6Ev4yh4 =8rlJ -----END PGP SIGNATURE-----

2 2

[libvirt] PATCH: Fix non-raw storage volume cloning
by Daniel P. Berrange 17 Jul '09

17 Jul '09

The recent storage patches broke cloning of non-raw volumes, due to a small typo. THis was identified by the libvirt TCK storage tests! Daniel diff --git a/src/storage_backend.c b/src/storage_backend.c index 1664804..67815d7 100644 --- a/src/storage_backend.c +++ b/src/storage_backend.c @@ -567,7 +567,7 @@ virStorageBackendGetBuildVolFromFunction(virConnectPtr conn, (inputvol->type == VIR_STORAGE_VOL_FILE && inputvol->target.format != VIR_STORAGE_VOL_FILE_RAW)) { - if ((tool_type = virStorageBackendFindFSImageTool(NULL)) != -1) { + if ((tool_type = virStorageBackendFindFSImageTool(NULL)) < 0) { virStorageReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", _("creation of non-raw file images is " "not supported without qemu-img.")); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 2

[libvirt] [PATCH] build: do not emit a trailing blank line into VC'd file, NEWS
by Jim Meyering 16 Jul '09

16 Jul '09

Cole Robinson noticed that the generated NEWS file would fail the "make syntax-check" tests. This adjusts the rule that generates it not to emit any of the offending trailing blank lines. At the same time, it regenerates NEWS. >From 53af8d848d78bcc5144b2c06b2109bbdc65803cf Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 16 Jul 2009 22:46:43 +0200 Subject: [PATCH] build: do not emit a trailing blank line into VC'd file, NEWS * docs/Makefile.am ($(top_builddir)/NEWS): Adjust rule to filter out any trailing blank lines when generating this file. * NEWS: Regenerate. --- NEWS | 56 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ docs/Makefile.am | 7 ++++- 2 files changed, 61 insertions(+), 2 deletions(-) diff --git a/NEWS b/NEWS index 5733fa8..447d2b4 100644 --- a/NEWS +++ b/NEWS @@ -5,6 +5,62 @@ http://libvirt.org/news.html +0.6.5: Jul 3 2009: + - New features: create storage columes on disk backend (Henrik Persson), + drop of capabilities based on libcap-ng when possible (Daniel + Berrange) + - Portability: fix build on non-Linux targets (Daniel Berrange) + - Documentation: typo and english fixes (Runa Bhattacharjee and + Garry Dolley), Docs on extending APIs (Dave Allan), cleanup + of debug and logging documentation (Amy Griffis), add + HACKING doc to the website (Daniel Berrange), + documentation for OpenNebula driver (Abel Miguez Rodriguez) + - Bug fixes: forbid autostart on transcient networks, + xen device removal crash (Daniel Berrange), re-detection of + transient VMs after libvirtd restart(Daniel Berrange), + bug in virFindFileInPath (Daniel Berrange), handle new + availheap sysctl in Xen (Daniel Berrange), allow USB hostdev + product 0 (Cole Robinson), cleanup when creating a storage pool + fails (Henrik Persson), domain id fix on redefinition in + test driver (Cole Robinson), fix raw storage allocation (Cole + Robinson), memory reporting for inactive qemu drivers (Cole + Robinson), segfault if storage pool has no type attribute (Cole + Robinson), OpenNebula compilation issues (Javier Fontan), + dominfo command without security driver (Daniel Berrange), + domain state problems after migration or destroy (Federico + Simoncelli), leak in node device parsing (Dave Allan), + storage pool definitions reading at startup (Cole Robinson), + bogus WWN in NPIV support (David Allan), avoid a segfault with + recent Xen (Sascha), cope with libnuma failures on weird + topologies (Dan Berrange), crash in QEMU driver with bad capabilities + data (Dan Berrange), trying to re-create a pool should not destroy + it (Dave Allan), endless loop in node device XML dump (Cole Robinson), + Re-label shared and readonly images (Dan Berrange) + - Improvements: create and destroy NPIV support (David Allan), + networking in UML driver (Daniel Berrange), HAL driver restart + thread safety (Daniel Berrange), capabilities and nodeinfo + APIs for LXC (Daniel Berrange), iNUMA API for VBox (Daniel Berrange), + dynamically search and use kvm-img qemu-img or qcow-create (Doug + Goldstein), fix qemu and kvm version parsing (Mark McLoughlin), + serial number for HAL storage (Dave Allan), improve error reporting + for virConnectOpen URIs (Daniel Berrange), include OS driver name + in device XML (Daniel Berrange), fix qemu command flags fetching + (Cole Robinson), check that qemu support -drive format= (Cole + Robinson), improve emulator detection (Cole Robinson), changes + to config parser to accomodate VMX syntax (Matthias Bolte), + update network schemas and driver for missing elements (Satoru SATOH), + avoid changing file context if not needed (Tim Waugh), + skip labelling if no src path (Cole Robinson), add arm emulation + if qemu-system-arm is present (C.J. Adams-Collier) + - Cleanups: daemon check logging env variables (Daniel Berrange), + User Mode Linux start and stop cleanups (Daniel Berrange), + share the NUMA api implementations (Daniel Berrange), storage + module dependancies (Dave Allan), refactor storage XML parsing + (Cole Robinson), big cleanup of logging code (Amy Griffis), + superfluous % on format (Matthias Bolte), cleanups and updates + on OpenNebula driver (Daniel Berrange and Abel Miguez Rodriguez) + + 0.6.4: May 29 2009: - New features: new API virStorageVolCreateXMLFrom (Cole Robinson), full VBox graphic capabilities (Pritesh Kothari), Interface config diff --git a/docs/Makefile.am b/docs/Makefile.am index 4bdb827..9ac2f6d 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -127,8 +127,11 @@ libvirt-api.xml libvirt-refs.xml: apibuild.py \ -(./apibuild.py) $(top_builddir)/NEWS: $(top_srcdir)/docs/news.xsl $(top_srcdir)/docs/news.html.in - -@(if [ -x $(XSLTPROC) ] ; then \ - $(XSLTPROC) --nonet $(top_srcdir)/docs/news.xsl $(top_srcdir)/docs/news.html.in > $(top_builddir)/NEWS ; fi ); + -@(if [ -x $(XSLTPROC) ] ; then \ + $(XSLTPROC) --nonet $(top_srcdir)/docs/news.xsl \ + $(top_srcdir)/docs/news.html.in \ + | perl -0777 -pe 's/\n\n+$$/\n/' \ + > $@-t && mv $@-t $@ ; fi ); clean-local: rm -f *~ *.bak *.hierarchy *.signals *-unused.txt -- 1.6.4.rc0.127.g81400

2 2

[libvirt] [PATCH 0/2] updated/working patches for netcf backend and virsh commands for virInterface*
by Laine Stump 16 Jul '09

16 Jul '09

These are updated versions of patches I previously sent: 1 - add iface-* commands to virsh to expose virInterface* functions to the cli. 2 - add the netcf backend driver. These haven't been exhaustively tested, but appear to work in simple smoke tests. Note that they all depend on the submitted-but-not-yet-committed virConnectListDefinedInterfaces() patch.

2 4

[libvirt] [PATCH 0/4] Storage disk backend fixes
by Cole Robinson 16 Jul '09

16 Jul '09

The following series fixes some issues in the disk storage backend, most wrt the recently added ability to create disk volumes. Cole Robinson (4): storage: disk: Fix parthelper '-g' option handling. storage: disk: Fix segfault creating volume without target path storage: disk: Default to 'ext2' for new volumes. storage: disk: Use capacity, not allocation, when creating volume. src/parthelper.c | 4 ++-- src/storage_backend_disk.c | 33 +++++++++++++++++++-------------- src/storage_conf.c | 2 +- 3 files changed, 22 insertions(+), 17 deletions(-)

3 14

[libvirt] [PATCH 0/9] Storage cloning for LVM and Disk backends
by Cole Robinson 16 Jul '09

16 Jul '09

This following series implements cloning for LVM and disk backends. Most of the functionality is already here, it just needed some reorganization to be accessible for every backend. I verified the following scenarios produced a bootable image: - Clone within a disk pool - Clone within a logical pool - Clone a raw file to a disk pool - Clone a disk pool to a logical pool Cole Robinson (9): storage: Refactor FS backend 'create' function choosing. storage: Always assume we are tracking alloc progress in FS backend storage: Move most of the FS creation functions to common backend. storage: cleanup: do away with 'createFile' storage: Break out actual raw cloning to separate function. storage: Implement 'CreateBlockFrom' helper. storage: Don't try sparse detection if writing to block device. storage: Implement CreateVolFrom for logical and disk backend. storage: Fix deadlock when cloning across pools. src/storage_backend.c | 471 +++++++++++++++++++++++++++++++++++++++++ src/storage_backend.h | 15 ++ src/storage_backend_disk.c | 15 ++ src/storage_backend_fs.c | 378 ++------------------------------- src/storage_backend_logical.c | 17 ++ src/storage_driver.c | 6 +- 6 files changed, 540 insertions(+), 362 deletions(-)

3 21