March 2013 - Devel - libvirt List Archives

[libvirt] [PATCHv5 0/3] DEVICE_DELETED event
by Michael S. Tsirkin 13 Mar '13

13 Mar '13

libvirt has a long-standing bug: when removing the device, it can request removal but does not know when the removal completes. Add an event so we can fix this in a robust way. First patch only adds the event for devices with ID, second path adds it for all devices and adds a path fields. Split this way for ease of backport (stable downstreams without QOM would want to only take the first patch), and also because the 2nd/3rd patches might turn out to be more controversial - if they really turn out such I'd like just the 1st one to get applied while we discuss 2 and 3. Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Changes from v4: - Add extra triggers and extra fields as requested by Markus Changes from v3: - Document that we only emit events for devices with and ID, as suggested by Markus Changes from v2: - move event toward the end of device_unparent, so that parents are reported after their children, as suggested by Paolo Changes from v1: - move to device_unparent - address comments by Andreas and Eric -- Anthony Liguori Michael S. Tsirkin (3): qdev: DEVICE_DELETED event qom: pass original path to unparent method qmp: add path to device_deleted event QMP/qmp-events.txt | 18 ++++++++++++++++++ hw/qdev.c | 15 +++++++++++++-- include/monitor/monitor.h | 1 + include/qom/object.h | 3 ++- monitor.c | 1 + qapi-schema.json | 4 +++- qom/object.c | 4 +++- 7 files changed, 41 insertions(+), 5 deletions(-) -- MST

2 5

Re: [libvirt] if_bridge.h: include in6.h for struct in6_addr use
by Thomas Backlund 13 Mar '13

13 Mar '13

Thomas Backlund skrev 13.1.2013 20:38: > patch both inline and attached as thunderbird tends to mess up ... > ----- > > if_bridge.h uses struct in6_addr ip6; but does not include the in6.h > header. > > Found by trying to build libvirt and connman against 3.8-rc3 headers. > Ok, ignore this patch, it's not the correct fix as it introduces redefinitions... Btw, the error that I hit that made me suggest this fix was libvirt config check bailing out: config.log:/usr/include/linux/if_bridge.h:173:20: error: field 'ip6' has incomplete type > Reported-by: Colin Guthrie <colin(a)mageia.org> > Reported-by: Christiaan Welvaart <cjw(a)daneel.dyndns.org> > Signed-off-by: Thomas Backlund <tmb(a)mageia.org> > > -- > > diff -Nurp linux-3.8-rc3/include/uapi/linux/if_bridge.h > linux-3.8-rc3.fix/include/uapi/linux/if_bridge.h > --- linux-3.8-rc3/include/uapi/linux/if_bridge.h 2013-01-13 > 20:09:54.257271755 +0200 > +++ linux-3.8-rc3.fix/include/uapi/linux/if_bridge.h 2013-01-13 > 20:15:04.153676151 +0200 > @@ -14,6 +14,7 @@ > #define _UAPI_LINUX_IF_BRIDGE_H > > #include <linux/types.h> > +#include <linux/in6.h> > > #define SYSFS_BRIDGE_ATTR "bridge" > #define SYSFS_BRIDGE_FDB "brforward" > > > ----- > Thomas > -- Thomas

9 14

[libvirt] understanding managedsave
by Nicolas Sebrecht 13 Mar '13

13 Mar '13

Hi, I'm trying to understand how the managedsave and restoration features work at the libvirt/qemu level. I'd like to write a little python script to reproduce the feature to learn how it works. So, I start a kvm guest by hand with qemu-kvm [...] -qmp tcp:localhost:4444,server and connect to the QMP with telnet or the qemu qmp-shell for my tests. >From what I've read, libvirt internally does a migration to a file: qemuDomainSaveMemory() -> qemuMonitorToFile() -> qemuMonitorMigrateToFd(). Why do a migration instead of QMP stop/memsave? What whould be the whole QMP/Monitor sequence of commands to handle the process? -- Nicolas Sebrecht

2 2

[libvirt] [PATCH] Apply security label when entering LXC namespaces
by Daniel P. Berrange 13 Mar '13

13 Mar '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Add a new virDomainLxcEnterSecurityLabel() function as a counterpart to virDomainLxcEnterNamespaces(), which can change the current calling process to have a new security context. This call runs client side, not in libvirtd so we can't use the security driver infrastructure. When entering a namespace, the process spawned from virsh will default to running with the security label of virsh. The actual desired behaviour is to run with the security label of the container most of the time. So this changes virsh lxc-enter-namespace command to invoke the virDomainLxcEnterSecurityLabel method. The current behaviour is: LABEL PID TTY TIME CMD system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 1 pts/0 00:00:00 systemd system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 3 pts/1 00:00:00 sh system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 24 ? 00:00:00 systemd-journal system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 29 ? 00:00:00 dhclient staff_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 47 ? 00:00:00 ps Note the ps command is running as unconfined_t, After this patch, The new behaviour is this: virsh -c lxc:/// lxc-enter-namespace dan -- /bin/ps -eZ LABEL PID TTY TIME CMD system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 1 pts/0 00:00:00 systemd system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 3 pts/1 00:00:00 sh system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 24 ? 00:00:00 systemd-journal system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 32 ? 00:00:00 dhclient system_u:system_r:svirt_lxc_net_t:s0:c0.c1023 38 ? 00:00:00 ps The '--noseclabel' flag can be used to skip security labelling. --- include/libvirt/libvirt-lxc.h | 4 ++ python/generator.py | 1 + src/libvirt-lxc.c | 96 +++++++++++++++++++++++++++++++++++++++++++ tools/virsh-domain.c | 32 +++++++++++++++ 4 files changed, 133 insertions(+) diff --git a/include/libvirt/libvirt-lxc.h b/include/libvirt/libvirt-lxc.h index f2c87fb..5021813 100644 --- a/include/libvirt/libvirt-lxc.h +++ b/include/libvirt/libvirt-lxc.h @@ -42,6 +42,10 @@ int virDomainLxcEnterNamespace(virDomainPtr domain, unsigned int *noldfdlist, int **oldfdlist, unsigned int flags); +int virDomainLxcEnterSecurityLabel(virSecurityModelPtr model, + virSecurityLabelPtr label, + virSecurityLabelPtr oldlabel, + unsigned int flags); # ifdef __cplusplus } diff --git a/python/generator.py b/python/generator.py index 8236bd2..6a25c2d 100755 --- a/python/generator.py +++ b/python/generator.py @@ -557,6 +557,7 @@ skip_function = ( lxc_skip_function = ( "virDomainLxcEnterNamespace", + "virDomainLxcEnterSecurityLabel", ) qemu_skip_function = ( #"virDomainQemuAttach", diff --git a/src/libvirt-lxc.c b/src/libvirt-lxc.c index f580c3c..c1bf168 100644 --- a/src/libvirt-lxc.c +++ b/src/libvirt-lxc.c @@ -29,6 +29,9 @@ #include "virlog.h" #include "virprocess.h" #include "datatypes.h" +#ifdef WITH_SELINUX +#include <selinux/selinux.h> +#endif #define VIR_FROM_THIS VIR_FROM_NONE @@ -163,3 +166,96 @@ error: virDispatchError(domain->conn); return -1; } + + +/** + * virDomainLxcEnterSecurityLabel: + * @model: the security model to set + * @label: the security label to apply + * @oldlabel: filled with old security label + * @flags: currently unused, pass 0 + * + * This API is LXC specific, so it will only work with hypervisor + * connections to the LXC driver. + * + * Attaches the process to the security label specified + * by @label. @label is interpreted relative to @model + * Depending on the security driver, this may + * not take effect until the next call to exec(). + * + * If @oldlabel is not NULL, it will be filled with info + * about the current security label. This may let the + * process be moved back to the previous label if no + * exec() has yet been performed. + * + * Returns 0 on success, -1 on error + */ +int +virDomainLxcEnterSecurityLabel(virSecurityModelPtr model, + virSecurityLabelPtr label, + virSecurityLabelPtr oldlabel, + unsigned int flags) +{ + virCheckFlagsGoto(0, error); + + virCheckNonNullArgGoto(model, error); + virCheckNonNullArgGoto(label, error); + + if (oldlabel) + memset(oldlabel, 0, sizeof(*oldlabel)); + + if (STREQ(model->model, "selinux")) { +#ifdef WITH_SELINUX + if (oldlabel) { + security_context_t ctx; + + if (getcon(&ctx) < 0) { + virReportSystemError(errno, + _("unable to get PID %d security context"), + getpid()); + goto error; + } + + if (strlen((char *) ctx) >= VIR_SECURITY_LABEL_BUFLEN) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("security label exceeds " + "maximum length: %d"), + VIR_SECURITY_LABEL_BUFLEN - 1); + freecon(ctx); + goto error; + } + + strcpy(oldlabel->label, (char *) ctx); + freecon(ctx); + + if ((oldlabel->enforcing = security_getenforce()) < 0) { + virReportSystemError(errno, "%s", + _("error calling security_getenforce()")); + goto error; + } + } + + if (setexeccon(label->label) < 0) { + virReportSystemError(errno, + _("Cannot set context %s"), + label->label); + goto error; + } +#else + virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, "%s", + _("Support for SELinux is not enabled")); + goto error; +#endif + } else { + virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, + _("Security model %s cannot be entered"), + model->model); + goto error; + } + + return 0; + +error: + virDispatchError(NULL); + return -1; +} diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 7caa364..f85e74d 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -7992,6 +7992,7 @@ static const vshCmdInfo info_lxc_enter_namespace[] = { static const vshCmdOptDef opts_lxc_enter_namespace[] = { {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, N_("domain name, id or uuid")}, + {"noseclabel", VSH_OT_BOOL, 0, N_("Do not change process security label")}, {"cmd", VSH_OT_ARGV, VSH_OFLAG_REQ, N_("namespace")}, {NULL, 0, 0, NULL} }; @@ -8008,11 +8009,17 @@ cmdLxcEnterNamespace(vshControl *ctl, const vshCmd *cmd) int nfdlist; int *fdlist; size_t i; + bool setlabel = true; + virSecurityModelPtr secmodel = NULL; + virSecurityLabelPtr seclabel = NULL; dom = vshCommandOptDomain(ctl, cmd, NULL); if (dom == NULL) goto cleanup; + if (vshCommandOptBool(cmd, "noseclabel")) + setlabel = false; + while ((opt = vshCommandOptArgv(cmd, opt))) { if (VIR_EXPAND_N(cmdargv, ncmdargv, 1) < 0) { vshError(ctl, _("%s: %d: failed to allocate argv"), @@ -8029,12 +8036,35 @@ cmdLxcEnterNamespace(vshControl *ctl, const vshCmd *cmd) if ((nfdlist = virDomainLxcOpenNamespace(dom, &fdlist, 0)) < 0) goto cleanup; + if (setlabel) { + fprintf(stderr, "Getr sec\n"); + if (VIR_ALLOC(secmodel) < 0) { + vshError(ctl, "%s", _("Failed to allocate security model")); + goto cleanup; + } + if (VIR_ALLOC(seclabel) < 0) { + vshError(ctl, "%s", _("Failed to allocate security label")); + goto cleanup; + } + if (virNodeGetSecurityModel(ctl->conn, secmodel) < 0) + goto cleanup; + if (virDomainGetSecurityLabel(dom, seclabel) < 0) + goto cleanup; + } + /* Fork once because we don't want to affect * virsh's namespace itself */ if (virFork(&pid) < 0) goto cleanup; if (pid == 0) { + if (setlabel && + virDomainLxcEnterSecurityLabel(secmodel, + seclabel, + NULL, + 0) < 0) + _exit(255); + if (virDomainLxcEnterNamespace(dom, nfdlist, fdlist, @@ -8067,6 +8097,8 @@ cmdLxcEnterNamespace(vshControl *ctl, const vshCmd *cmd) ret = true; cleanup: + VIR_FREE(seclabel); + VIR_FREE(secmodel); if (dom) virDomainFree(dom); VIR_FREE(cmdargv); -- 1.8.1.4

3 3

[libvirt] [PATCH] Fix query of LXC security label
by Daniel P. Berrange 13 Mar '13

13 Mar '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> The virDomainGetSecurityLabel method is currently (mistakenly) showing the label of the libvirt_lxc process: ...snip... Security model: selinux Security DOI: 0 Security label: system_u:system_r:virtd_t:s0-s0:c0.c1023 (permissive) when it should be showing the init process label ...snip... Security model: selinux Security DOI: 0 Security label: system_u:system_r:svirt_t:s0:c724,c995 (permissive) --- src/lxc/lxc_driver.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 338b8eb..7791966 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1162,8 +1162,16 @@ static int lxcDomainGetSecurityLabel(virDomainPtr dom, virSecurityLabelPtr secla * LXC monitor hasn't seen SIGHUP/ERR on poll(). */ if (virDomainObjIsActive(vm)) { + virLXCDomainObjPrivatePtr priv = vm->privateData; + + if (!priv->initpid) { + virReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("Init pid is not yet available")); + goto cleanup; + } + if (virSecurityManagerGetProcessLabel(driver->securityManager, - vm->def, vm->pid, seclabel) < 0) { + vm->def, priv->initpid, seclabel) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to get security label")); goto cleanup; -- 1.8.1.4

2 1

[libvirt] [PATCHv5] virtio-rng: Add rate limiting options for virtio-RNG
by Peter Krempa 13 Mar '13

13 Mar '13

Qemu's implementation of virtio RNG supports rate limiting of the entropy used. This patch exposes the option to tune this functionality. This patch is based on qemu commit 904d6f588063fb5ad2b61998acdf1e73fb4 The rate limiting is exported in the XML as: <devices> ... <rng model='virtio'> <rate period='1234'>4321</rate> <backend model='random'/> </rng> ... --- Notes: Version 5: - adjust after forbiding arbitrary name passthrough - rebase to master Version 4: - Reword docs - state it is available since 1.0.4 as the tree is frozen and this was actually never acked before Version 3: - State the time unit in docs Version 2: - qemu uses bytes/period, adapt the value according to that docs/formatdomain.html.in | 14 ++++++++++++++ docs/schemas/domaincommon.rng | 18 +++++++++++++++++- src/conf/domain_conf.c | 17 +++++++++++++++++ src/conf/domain_conf.h | 2 ++ src/qemu/qemu_command.c | 9 +++++++++ .../qemuxml2argv-virtio-rng-random.args | 2 +- .../qemuxml2argv-virtio-rng-random.xml | 1 + 7 files changed, 61 insertions(+), 2 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 4cafc92..b8eb53d 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -4280,6 +4280,7 @@ qemu-kvm -net nic,model=? /dev/null ... <devices> <rng model='virtio'> + <rate period="2000">1234</rate> <backend model='random'>/dev/random</backend>  <backend model='egd' type='udp'> @@ -4302,6 +4303,19 @@ qemu-kvm -net nic,model=? /dev/null <li>'virtio' — supported by qemu and virtio-rng kernel module</li> </ul> </dd> + <dt><code>rate</code></dt> + <dd> + + The optional <code>rate</code> element allows limiting the rate at + which entropy can be consumed from the source. An optional + <code>period</code> attribute specifies the duration of a period in + milliseconds; if omitted, the period is taken as 1000 milliseconds + (1 second). The element contents specify how many bits are permitted + per period. Drivers may enforce a minimum rate, and may round the + rate down to a minimum granularity. + Since 1.0.4 + + </dd> <dt><code>backend</code></dt> <dd> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 4b60885..dc45af1 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -3500,7 +3500,12 @@ <value>virtio</value> </choice> </attribute> - <ref name="rng-backend"/> + <interleave> + <ref name="rng-backend"/> + <optional> + <ref name="rng-rate"/> + </optional> + </interleave> </element> </define> @@ -3527,6 +3532,17 @@ </element> </define> + <define name="rng-rate"> + <element name="rate"> + <optional> + <attribute name="period"> + <ref name="positiveInteger"/> + </attribute> + </optional> + <ref name="positiveInteger"/> + </element> + </define> + <define name="usbmaster"> <element name="master"> <attribute name="startport"> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 739bd72..8a02375 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -7400,6 +7400,17 @@ virDomainRNGDefParseXML(const xmlNodePtr node, ctxt->node = node; + if (virXPathUInt("string(./rate)", ctxt, &def->rate) < -1) { + virReportError(VIR_ERR_XML_ERROR, "%s", _("invalid RNG rate value")); + goto error; + } + + if (def->rate > 0 && + virXPathUInt("string(./rate/@period)", ctxt, &def->period) < -1) { + virReportError(VIR_ERR_XML_ERROR, "%s", _("invalid RNG period value")); + goto error; + } + if ((nbackends = virXPathNodeSet("./backend", ctxt, &backends)) < 0) goto error; @@ -13715,6 +13726,12 @@ virDomainRNGDefFormat(virBufferPtr buf, const char *backend = virDomainRNGBackendTypeToString(def->backend); virBufferAsprintf(buf, " <rng model='%s'>\n", model); + if (def->rate) { + virBufferAddLit(buf, " <rate"); + if (def->period) + virBufferAsprintf(buf, " period='%u'", def->period); + virBufferAsprintf(buf, ">%u</rate>\n", def->rate); + } virBufferAsprintf(buf, " <backend model='%s'", backend); switch ((enum virDomainRNGBackend) def->backend) { diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 2509193..e19b5c0 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1721,6 +1721,8 @@ enum virDomainRNGBackend { struct _virDomainRNGDef { int model; int backend; + unsigned int rate; /* bits per period */ + unsigned int period; /* milliseconds */ union { char *file; /* file name for 'random' source */ diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 201fac1..c0f8dd2 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -4255,6 +4255,15 @@ qemuBuildRNGDeviceArgs(virCommandPtr cmd, virBufferAsprintf(&buf, "virtio-rng-pci,rng=%s", dev->info.alias); + if (dev->rate > 0) { + /* qemu uses bytes */ + virBufferAsprintf(&buf, ",max-bytes=%u", dev->rate / 8); + if (dev->period) + virBufferAsprintf(&buf, ",period=%u", dev->period); + else + virBufferAddLit(&buf, ",period=1000"); + } + if (qemuBuildDeviceAddressStr(&buf, &dev->info, qemuCaps) < 0) goto cleanup; diff --git a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args index 7ab9dbc..33b5adc 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args @@ -3,4 +3,4 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu \ -monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -usb \ -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 \ -object rng-random,id=rng0,filename=/dev/hwrng \ --device virtio-rng-pci,rng=rng0,bus=pci.0,addr=0x4 +-device virtio-rng-pci,rng=rng0,max-bytes=100,period=1234,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml index 1e2c4be..813e42e 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml @@ -17,6 +17,7 @@ <controller type='usb' index='0'/> <memballoon model='virtio'/> <rng model='virtio'> + <rate period='1234'>800</rate> <backend model='random'>/dev/hwrng</backend> </rng> </devices> -- 1.8.1.1

2 2

[libvirt] [PATCH] Use separate symbol file for GNUTLS symbols
by Daniel P. Berrange 13 Mar '13

13 Mar '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> A number of symbols are only present when GNUTLS is enabled. Thus we must use a separate libvirt_gnutls.syms file for them instead of libvirt_private.syms Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/Makefile.am | 7 +++++++ src/libvirt_gnutls.syms | 46 ++++++++++++++++++++++++++++++++++++++++++++++ src/libvirt_private.syms | 23 ----------------------- 3 files changed, 53 insertions(+), 23 deletions(-) create mode 100644 src/libvirt_gnutls.syms diff --git a/src/Makefile.am b/src/Makefile.am index 60935f4..a6cc839 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -1458,6 +1458,12 @@ else SYM_FILES += $(srcdir)/libvirt_sasl.syms endif +if WITH_GNUTLS +USED_SYM_FILES += $(srcdir)/libvirt_gnutls.syms +else +SYM_FILES += $(srcdir)/libvirt_gnutls.syms +endif + if WITH_SSH2 USED_SYM_FILES += $(srcdir)/libvirt_libssh2.syms else @@ -1481,6 +1487,7 @@ EXTRA_DIST += \ libvirt_openvz.syms \ libvirt_qemu.syms \ libvirt_sasl.syms \ + libvirt_gnutls.syms \ libvirt_vmx.syms \ libvirt_xenxs.syms \ libvirt_libssh2.syms diff --git a/src/libvirt_gnutls.syms b/src/libvirt_gnutls.syms new file mode 100644 index 0000000..bd4f950 --- /dev/null +++ b/src/libvirt_gnutls.syms @@ -0,0 +1,46 @@ +# +# GNUTLS-specific symbols +# + +# rpc/virnetclient.h +virNetClientGetTLSKeySize; +virNetClientSetTLSSession; + + +# rpc/virnetserver.h +virNetServerSetTLSContext; + + +# rpc/virnetserverclient.h +virNetServerClientGetTLSKeySize; +virNetServerClientHasTLSSession; + + +# rpc/virnetserverservice.h +virNetServerServiceGetTLSContext; + + +# rpc/virnetsocket.h +virNetSocketSetTLSSession; + + +# rpc/virnettlscontext.h +virNetTLSContextCheckCertificate; +virNetTLSContextNewClient; +virNetTLSContextNewClientPath; +virNetTLSContextNewServer; +virNetTLSContextNewServerPath; +virNetTLSInit; +virNetTLSSessionGetHandshakeStatus; +virNetTLSSessionGetKeySize; +virNetTLSSessionHandshake; +virNetTLSSessionNew; +virNetTLSSessionRead; +virNetTLSSessionSetIOCallbacks; +virNetTLSSessionWrite; + + +# Let emacs know we want case-insensitive sorting +# Local Variables: +# sort-fold-case: t +# End: diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 0e8fcbf..fbd540a 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -762,7 +762,6 @@ virNetClientAddStream; virNetClientClose; virNetClientDupFD; virNetClientGetFD; -virNetClientGetTLSKeySize; virNetClientHasPassFD; virNetClientIsEncrypted; virNetClientIsOpen; @@ -784,7 +783,6 @@ virNetClientSendNoReply; virNetClientSendWithReply; virNetClientSendWithReplyStream; virNetClientSetCloseCallback; -virNetClientSetTLSSession; # rpc/virnetclientprogram.h @@ -844,7 +842,6 @@ virNetServerPreExecRestart; virNetServerQuit; virNetServerRemoveShutdownInhibition; virNetServerRun; -virNetServerSetTLSContext; virNetServerUpdateServices; @@ -856,9 +853,7 @@ virNetServerClientGetAuth; virNetServerClientGetFD; virNetServerClientGetPrivateData; virNetServerClientGetReadonly; -virNetServerClientGetTLSKeySize; virNetServerClientGetUNIXIdentity; -virNetServerClientHasTLSSession; virNetServerClientImmediateClose; virNetServerClientInit; virNetServerClientInitKeepAlive; @@ -910,7 +905,6 @@ virNetServerServiceClose; virNetServerServiceGetAuth; virNetServerServiceGetMaxRequests; virNetServerServiceGetPort; -virNetServerServiceGetTLSContext; virNetServerServiceIsReadonly; virNetServerServiceNewFD; virNetServerServiceNewPostExecRestart; @@ -952,27 +946,10 @@ virNetSocketRemoteAddrString; virNetSocketRemoveIOCallback; virNetSocketSendFD; virNetSocketSetBlocking; -virNetSocketSetTLSSession; virNetSocketUpdateIOCallback; virNetSocketWrite; -# rpc/virnettlscontext.h -virNetTLSContextCheckCertificate; -virNetTLSContextNewClient; -virNetTLSContextNewClientPath; -virNetTLSContextNewServer; -virNetTLSContextNewServerPath; -virNetTLSInit; -virNetTLSSessionGetHandshakeStatus; -virNetTLSSessionGetKeySize; -virNetTLSSessionHandshake; -virNetTLSSessionNew; -virNetTLSSessionRead; -virNetTLSSessionSetIOCallbacks; -virNetTLSSessionWrite; - - # security/security_driver.h virSecurityDriverLookup; -- 1.8.1.4

2 1

[libvirt] [PATCH] qemu_driver: Try KVM_CAP_MAX_VCPUS only if defined
by Michal Privoznik 13 Mar '13

13 Mar '13

With our recent patch (1715c83b5f) we thrive to get the correct number of maximal VCPUs. However, we are using a constant from linux/kvm.h which may be not defined in every distro. Hence, we should guard usage of the constant with ifdef preprocessor directive. --- src/qemu/qemu_driver.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index de53a1b..c3a8f24 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1122,9 +1122,11 @@ kvmGetMaxVCPUs(void) { return -1; } +#ifdef KVM_CAP_MAX_VCPUS /* at first try KVM_CAP_MAX_VCPUS to determine the maximum count */ if ((ret = ioctl(fd, KVM_CHECK_EXTENSION, KVM_CAP_MAX_VCPUS)) > 0) goto cleanup; +#endif /* KVM_CAP_MAX_VCPUS */ /* as a fallback get KVM_CAP_NR_VCPUS (the recommended maximum number of * vcpus). Note that on most machines this is set to 160. */ -- 1.8.1.5

2 2

[libvirt] [PATCHv3 00/27] Driver XML conf adjustment callbacks [READ FIRST]
by Peter Krempa 13 Mar '13

13 Mar '13

This series implements the driver XML parsing callbacks as Dan suggested. This series may appear to be monstrous but I chose to split some patches to separate ones although the code does not compile afterwards to ease review. !! READ THIS: !! !! The patches starting with the word "fix" will need to be squashed into the !! non "fix" patches predceeding them otherwise it will not compile cleanly. This new version fixes the review comments by Dan, Michal and Laine. There's still outstanding work to do to get rid of all the irrelevant data from virCaps, but I'd like to stabilize the design before doing so. In this series the callback now supports passing opaque data to the callback functions and it's demonstrated in "virCaps: get rid of defaultDiskDriverName". There's also a patch demonstrating doing check that is not covered by the XML schema and thus shouldn't be part of the parser, but is docummented for all drivers and thus should be enforced. This demonstrates the use of the verification function. Peter Krempa (27): virCaps: conf: start splitting out irrelevat data fix fallout in src/conf/ fix fallout in src/esx fix fallout in src/lxc fix fallout in src/openvz fix fallout in src/parallels fix fallout in phyp driver fix fallout in src/qemu/ fix fallout in tests/ fix fallout in src/xen fix fallout in src/vmware fix fallout in src/uml fix fallout in test driver fix fallout in vbox driver fix fallout in aahelper code fix fallout in src/libxl fix fallout in src/xenapi fix fallout in selinux test conf: Add separate defaults addition and validation for XML parsing fix fallout from adding new parameter to internal APIs conf: Add argument to support use of the driver adjust callbacks qemu: Record the default NIC model in the domain XML fix fallout of the change of the network device default virCaps: get rid of "defaultInitPath" value in the virCaps struct virCaps: get rid of defaultDiskDriverName virCaps: get rid of emulatorRequired conf: Enforce ranges on cputune variables src/Makefile.am | 1 + src/conf/capabilities.c | 10 - src/conf/capabilities.h | 17 +- src/conf/domain_conf.c | 387 +++++++++++++++++---- src/conf/domain_conf.h | 71 +++- src/conf/snapshot_conf.c | 3 +- src/conf/snapshot_conf.h | 1 + src/esx/esx_driver.c | 14 +- src/esx/esx_private.h | 2 + src/libvirt_private.syms | 5 +- src/libxl/libxl_conf.h | 1 + src/libxl/libxl_driver.c | 57 +-- src/lxc/lxc_conf.c | 13 +- src/lxc/lxc_conf.h | 2 + src/lxc/lxc_controller.c | 7 +- src/lxc/lxc_domain.c | 26 +- src/lxc/lxc_domain.h | 3 +- src/lxc/lxc_driver.c | 49 +-- src/lxc/lxc_process.c | 9 +- src/openvz/openvz_conf.c | 5 +- src/openvz/openvz_conf.h | 1 + src/openvz/openvz_driver.c | 45 ++- src/parallels/parallels_driver.c | 12 +- src/parallels/parallels_utils.h | 1 + src/phyp/phyp_driver.c | 12 +- src/phyp/phyp_driver.h | 1 + src/qemu/qemu_capabilities.c | 3 - src/qemu/qemu_command.c | 9 +- src/qemu/qemu_command.h | 3 + src/qemu/qemu_conf.c | 14 +- src/qemu/qemu_conf.h | 5 + src/qemu/qemu_domain.c | 108 ++++-- src/qemu/qemu_domain.h | 7 +- src/qemu/qemu_driver.c | 134 ++++--- src/qemu/qemu_migration.c | 22 +- src/qemu/qemu_process.c | 109 +----- src/security/virt-aa-helper.c | 10 +- src/test/test_driver.c | 61 +++- src/uml/uml_conf.h | 1 + src/uml/uml_driver.c | 32 +- src/vbox/vbox_tmpl.c | 27 +- src/vmware/vmware_conf.c | 3 +- src/vmware/vmware_conf.h | 1 + src/vmware/vmware_driver.c | 25 +- src/xen/xen_driver.c | 8 +- src/xen/xen_driver.h | 1 + src/xen/xend_internal.c | 14 +- src/xen/xm_internal.c | 4 +- src/xenapi/xenapi_driver.c | 34 +- src/xenapi/xenapi_driver_private.h | 1 + tests/domainsnapshotxml2xmltest.c | 5 + tests/lxcxml2xmldata/lxc-hostdev.xml | 1 + tests/lxcxml2xmldata/lxc-systemd.xml | 1 + tests/lxcxml2xmltest.c | 7 +- tests/qemuargv2xmltest.c | 8 +- tests/qemumonitorjsontest.c | 29 +- tests/qemumonitortestutils.c | 4 +- tests/qemumonitortestutils.h | 5 +- .../qemuxml2argv-net-bandwidth.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-client.args | 4 +- .../qemuxml2argv-net-eth-ifname.args | 4 +- .../qemuxml2argv-net-eth-ifname.xml | 1 + .../qemuxml2argv-net-eth-names.args | 8 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.args | 4 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-hostdev.xml | 1 + tests/qemuxml2argvdata/qemuxml2argv-net-mcast.args | 4 +- .../qemuxml2argv-net-openvswitch.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-server.args | 4 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.args | 3 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.xml | 1 + .../qemuxml2argv-net-virtio-network-portgroup.xml | 2 + tests/qemuxml2argvtest.c | 5 +- .../qemuxml2xmlout-graphics-spice-timeout.xml | 1 + tests/qemuxml2xmltest.c | 6 +- tests/qemuxmlnstest.c | 5 +- tests/securityselinuxlabeltest.c | 6 +- tests/testutilsqemu.c | 3 +- tests/testutilsqemu.h | 2 + tests/testutilsxen.c | 6 + tests/testutilsxen.h | 2 + tests/xmconfigtest.c | 8 +- tests/xml2sexprtest.c | 8 +- tests/xml2vmxtest.c | 8 +- 84 files changed, 1034 insertions(+), 516 deletions(-) -- 1.8.1.5

2 56

[libvirt] failed to mount cgroup
by Yin Olivia-R63875 13 Mar '13

13 Mar '13

Hi, I tried to run libvirt-1.0.2 with LXC as below, but it failed to mount cgroup. 1). mount /cgroup # mkdir /cgroup # mount -t cgroup cgroup /cgroup # vi vm1.xml <domain type='lxc'> <name>vm1</name> <memory>32768</memory> <os> <type>exe</type> <init>/bin/sh</init> </os> <vcpu>1</vcpu> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/libexec/libvirt_lxc</emulator> <console type='pty' tty='/dev/pts/3'> <source path='/dev/pts/3'/> <target port='0'/> </console> </devices> </domain> # virsh -c lxc:/// define vm1.xml # virsh -c lxc:/// start vm1 error: Failed to start domain vm1 error: internal error guest failed to start: 2013-03-05 02:13:03.639+0000: 2403: info : libvirt version: 1.0.2 2013-03-05 02:13:03.639+0000: 2403: warning : lxcCapsInit:73 : Failed to get host power management capabilities PATH=/bin:/sbin TERM=linux container=lxc-libvirt container_uuid=d50aec75-4566-62b3-ed81-adf3e8705efe LIBVIRT_LXC_UUID=d50aec75-4566-62b3-ed81-adf3e8705efe LIBVIRT_LXC_NAME=vm1 /bin/sh 2013-03-05 02:13:03.779+0000: 1: error : lxcContainerIdentifyCGroups:1775 : Unable to read directory : No such file or directory 2013-03-05 02:13:03.779+0000: 2404: error : virLXCControllerRun:1468 : error receiving signal from container: Input/output error # ps aux | grep libvirtd root 2314 0.1 0.4 108320 4876 ? Sl 02:47 0:00 libvirtd -d root 2383 0.0 0.0 4116 452 ttyS0 S+ 02:52 0:00 grep libvirtd # cat /proc/2314/status Name: libvirtd State: S (sleeping) Tgid: 2314 Pid: 2314 PPid: 1 TracerPid: 0 Uid: 0 0 0 0 Gid: 0 0 0 0 FDSize: 32 Groups: 0 VmPeak: 112420 kB VmSize: 108320 kB VmLck: 0 kB VmHWM: 4884 kB VmRSS: 4876 kB VmData: 91540 kB VmStk: 136 kB VmExe: 348 kB VmLib: 13092 kB VmPTE: 96 kB VmSwap: 0 kB Threads: 11 SigQ: 0/7760 SigPnd: 0000000000000000 ShdPnd: 0000000000000000 SigBlk: 0000000000000000 SigIgn: 0000000000001000 SigCgt: 0000000180004cef CapInh: 0000000000000000 CapPrm: ffffffffffffffff CapEff: ffffffffffffffff CapBnd: ffffffffffffffff Cpus_allowed: 3 Cpus_allowed_list: 0-1 Mems_allowed: 1 Mems_allowed_list: 0 voluntary_ctxt_switches: 59 nonvoluntary_ctxt_switches: 24 # cat /proc/2314/cgroup 1:freezer,devices,memory,cpuacct,cpuset:/ # mount | grep cgroup cgroup on /cgroup type cgroup (rw,relatime,freezer,devices,memory,cpuacct,cpuset) # vi /var/log/libvirt/lxc/vm1.log 2013-03-05 02:48:39.218+0000: starting up PATH=/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin LIBVIRT_DEBUG=3 LIBVIRT_LOG_OUTPUTS=3:stderr /usr/libexec/libvirt_lxc --name vm1 --console 17 --security=none --handshake 20 --background 2013-03-05 02:48:39.229+0000: 2365: info : libvirt version: 1.0.2 2013-03-05 02:48:39.229+0000: 2365: warning : lxcCapsInit:73 : Failed to get host power management capabilities PATH=/bin:/sbin TERM=linux container=lxc-libvirt container_uuid=523492fb-36c8-3662-c61e-91697d5ad0ab LIBVIRT_LXC_UUID=523492fb-36c8-3662-c61e-91697d5ad0ab LIBVIRT_LXC_NAME=vm1 /bin/sh 2013-03-05 02:48:39.299+0000: 1: error : lxcContainerIdentifyCGroups:1775 : Unable to read directory : No such file or directory 2013-03-05 02:48:39.300+0000: 2366: error : virLXCControllerRun:1468 : error receiving signal from container: Input/output error But if mount cgroup on another poinit, the error message is different. 2). mount /dev/cgroups # mkdir /dev/cgroups # mount -t cgroup cgroup /dev/cgroups # virsh -c lxc:/// start vm1 # vi /var/log/libvirt/lxc/vm1.log 2013-03-05 03:15:49.660+0000: starting up PATH=/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin LIBVIRT_DEBUG=3 LIBVIRT_LOG_OUTPUTS=3:stderr /usr/libexec/libvirt_lxc --name vm1 --console 2013-03-05 03:15:49.666+0000: 2430: info : libvirt version: 1.0.2 2013-03-05 03:15:49.666+0000: 2430: warning : lxcCapsInit:73 : Failed to get host power management capabilities PATH=/bin:/sbin TERM=linux container=lxc-libvirt container_uuid=518d6735-42d5-90b0-3780-d41a651385fc LIBVIRT_LXC_UUID=518d6735-42d5-90b0-3780-d41a651385fc LI 2013-03-05 03:15:49.749+0000: 1: error : lxcContainerMountCGroups:1892 : Failed to mount cgroup on '/dev/cgroups': No such file or directory 2013-03-05 03:15:49.749+0000: 2431: error : virLXCControllerRun:1468 : error receiving signal from container: Input/output error Best Regards, Olivia

4 8