August 2014 - Devel - libvirt List Archives

[libvirt] [PATCHv2 00/16] LXC network configuration support
by Cédric Bosdonnat 27 Aug '14

27 Aug '14

Hi all, Here is the whole series resent with a major addition: the functions used to set the IP and add a route now use libnl when possible. The idea behind this is to avoid requiring iproute2 or ifconfig installed in the container rootfs. Otherwise nothing changed since v1. Cédric Bosdonnat (16): Forgot to cleanup ifname_guest* in domain network def parsing Domain conf: allow more than one IP address for net devices LXC: set IP addresses to veth devices in the container lxc conf2xml: convert IP addresses Allow network capabilities hostdev to configure IP addresses lxc conf2xml: convert ip addresses for hostdev NICs Domain network devices can now have a <gateway> element lxc conf2xml: convert lxc.network.ipv[46].gateway LXC: use the new net devices gateway definition LXC: honour network devices link state Wrong place for virDomainNetIpsFormat virNetDevSetIPv4Address: libnl implementation Renamed virNetDevSetIPv4Address to virNetDevSetIPAddress virNetDevAddRoute: implementation using netlink virNetDevClearIPv4Address: netlink implementation Renamed virNetDevClearIPv4Address to virNetDevClearIPAddress docs/formatdomain.html.in | 39 +++ docs/schemas/domaincommon.rng | 55 +++- src/conf/domain_conf.c | 214 +++++++++++++-- src/conf/domain_conf.h | 22 +- src/libvirt_private.syms | 7 +- src/lxc/lxc_container.c | 74 ++++- src/lxc/lxc_native.c | 173 ++++++++---- src/network/bridge_driver.c | 4 +- src/openvz/openvz_conf.c | 2 +- src/openvz/openvz_driver.c | 6 +- src/qemu/qemu_driver.c | 25 +- src/qemu/qemu_hotplug.c | 6 +- src/uml/uml_conf.c | 2 +- src/util/virnetdev.c | 305 ++++++++++++++++++--- src/util/virnetdev.h | 12 +- src/util/virnetlink.c | 38 +++ src/util/virnetlink.h | 2 + src/vbox/vbox_common.c | 3 +- src/xenconfig/xen_common.c | 15 +- src/xenconfig/xen_sxpr.c | 12 +- .../lxcconf2xmldata/lxcconf2xml-physnetwork.config | 4 + tests/lxcconf2xmldata/lxcconf2xml-physnetwork.xml | 3 + tests/lxcconf2xmldata/lxcconf2xml-simple.config | 4 + tests/lxcconf2xmldata/lxcconf2xml-simple.xml | 3 + tests/lxcxml2xmldata/lxc-hostdev.xml | 3 + tests/lxcxml2xmldata/lxc-idmap.xml | 3 + 26 files changed, 880 insertions(+), 156 deletions(-) -- 1.8.4.5

4 19

[libvirt] [PATCH 0/3] Fix VirtualBox registration
by Michal Privoznik 27 Aug '14

27 Aug '14

This is supposed to fix bug report which appeared on the list [1]. The driver registration ordering is back as it was, but the subdrivers are registered before. Unfortunately, if we don't want to dig vbox network and storage code out, we have to compile the sources couple of times. Michal Privoznik (3): virdrivermoduletest: Test all the modules virDriverLoadModule: Honor libvirt func name tranlsation vbox: Register per partes daemon/libvirtd.c | 16 +++++++++++++-- src/Makefile.am | 41 ++++++++++++++++++++++++++++++++++--- src/driver.c | 19 +++++++++++++++-- src/vbox/vbox_driver.c | 50 ++++++++++++++++++++++++++++++++++++++++----- src/vbox/vbox_driver.h | 10 +++++++++ tests/virdrivermoduletest.c | 18 ++++++++++------ 6 files changed, 136 insertions(+), 18 deletions(-) -- 1.8.5.5

2 5

[libvirt] [PATCH] blockcopy: allow larger buf-size
by Eric Blake 27 Aug '14

27 Aug '14

While qemu definitely caps granularity to 64 MiB, it places no limits on buf-size. On a machine beefy enough for lots of memory, a buf-size larger than 2 GiB is feasible, so we should pass a 64-bit parameter. * include/libvirt/libvirt.h.in (VIR_DOMAIN_BLOCK_COPY_BUF_SIZE): Allow 64 bits. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Based on Peter's review of my patch series; until we have an implementation, it is just a doc bug, but I'd rather document it correctly and avoid baking in a too-small API. include/libvirt/libvirt.h.in | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index e79c9ad..e80634e 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -2663,8 +2663,8 @@ typedef enum { * VIR_DOMAIN_BLOCK_COPY_BUF_SIZE: * Macro for the virDomainBlockCopy buffer size tunable: it represents * how much data in bytes can be in flight between source and destination, - * as an unsigned int. Specifying 0 is the same as omitting this parameter, - * to request the hypervisor default. + * as an unsigned long long. Specifying 0 is the same as omitting this + * parameter, to request the hypervisor default. */ #define VIR_DOMAIN_BLOCK_COPY_BUF_SIZE "buf-size" -- 1.9.3

1 0

[libvirt] [PATCH v2 0/8] new virDomainBlockCopy API
by Eric Blake 27 Aug '14

27 Aug '14

v2: finish the series for local files, address Peter's doc suggestions on patch 1, and make several tweaks to many of the patches to fix issues that I found in testing the final product. Not included - patches to actually target a gluster destination. But those can be delayed until 1.2.9. Eric Blake (8): blockcopy: virDomainBlockCopy with XML destination, typed params blockcopy: split out virsh implementation blockcopy: expose new API in virsh blockcopy: remote implementation for new API blockcopy: tweak how rebase calls into copy blockcopy: add a way to parse disk source blockcopy: add qemu implementation of new API blockcopy: add qemu implementation of new tunables include/libvirt/libvirt.h.in | 61 +++++++++++++-- src/conf/domain_conf.c | 147 +++++++++++++++++++++++------------ src/conf/domain_conf.h | 4 + src/driver.h | 11 ++- src/libvirt.c | 120 ++++++++++++++++++++++++++++- src/libvirt_private.syms | 1 + src/libvirt_public.syms | 5 ++ src/qemu/qemu_driver.c | 177 ++++++++++++++++++++++++++++++++----------- src/qemu/qemu_migration.c | 2 +- src/qemu/qemu_monitor.c | 8 +- src/qemu/qemu_monitor.h | 2 + src/qemu/qemu_monitor_json.c | 3 + src/qemu/qemu_monitor_json.h | 2 + src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 18 ++++- src/remote_protocol-structs | 11 +++ tests/qemumonitorjsontest.c | 2 +- tools/virsh-domain.c | 153 +++++++++++++++++++++++++++++++------ tools/virsh.pod | 35 +++++---- 19 files changed, 623 insertions(+), 140 deletions(-) -- 1.9.3

2 23

[libvirt] [RFC PATCH 0/3] Introduce virDomainOpenGraphicsFD API
by Ján Tomko 27 Aug '14

27 Aug '14

Introduce virDomainOpenGraphicsFD which returns a socket fd created by the daemon, unlike virDomainOpenGraphics, where it's created by client. This should be usable by virt-viewer under SELinux, but I was not able to verify that yet. Ján Tomko (3): Introduce virDomainOpenGraphicsFD API Add RPC implementation for virDomainOpenGraphicsFd Wire up virDomainOpenGraphicsFD in QEMU driver daemon/remote.c | 42 +++++++++++++++++++++++ include/libvirt/libvirt.h.in | 5 +++ src/driver.h | 7 ++++ src/libvirt.c | 58 ++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 5 +++ src/qemu/qemu_driver.c | 80 ++++++++++++++++++++++++++++++++++++++++++++ src/remote/remote_driver.c | 39 +++++++++++++++++++++ src/remote/remote_protocol.x | 15 ++++++++- src/rpc/virnetmessage.c | 26 ++++++++++++++ src/rpc/virnetmessage.h | 3 ++ 10 files changed, 279 insertions(+), 1 deletion(-) -- 1.8.5.5

3 14

[libvirt] Plan for the next release
by Daniel Veillard 27 Aug '14

27 Aug '14

We already have a number of commits in, over 150, I can't freeze next monday but I could do this next Tuesday (the 26) for a planned release around Sep 1st, So unless there is a good reason to change this, that's the plan :) Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

4 4

[libvirt] [libvirt-python PATCH] Implement virDomainOpenGraphicsFD
by Ján Tomko 27 Aug '14

27 Aug '14

--- generator.py | 1 + libvirt-override-api.xml | 7 +++++++ libvirt-override.c | 37 +++++++++++++++++++++++++++++++++++++ 3 files changed, 45 insertions(+) diff --git a/generator.py b/generator.py index a12c52b..32ae34d 100755 --- a/generator.py +++ b/generator.py @@ -464,6 +464,7 @@ skip_impl = ( 'virConnectGetCPUModelNames', 'virNodeGetFreePages', 'virNetworkGetDHCPLeases', + 'virDomainOpenGraphicsFD', ) lxc_skip_impl = ( diff --git a/libvirt-override-api.xml b/libvirt-override-api.xml index 09bbbf8..c7b6152 100644 --- a/libvirt-override-api.xml +++ b/libvirt-override-api.xml @@ -640,5 +640,12 @@ <arg name='flags' type='unsigned int' info='unused, pass 0'/> <return type='char *' info="list of leases"/> </function> + <function name="virDomainOpenGraphicsFD" file='python'> + <info>Returns a file descriptor connected to the graphics framebuffer </info> + <arg name='domain' type='virDomainPtr' info='a domain object'/> + <arg name='idx' type='int' info='index of the graphics framebuffer'/> + <arg name='flags' type='unsigned int' info='bitwise OR of virDomainOpenGraphicsFlags'/> + <return type='char *' info="list of fds"/> + </function> </symbols> </api> diff --git a/libvirt-override.c b/libvirt-override.c index b2271ae..42a4b05 100644 --- a/libvirt-override.c +++ b/libvirt-override.c @@ -7955,6 +7955,40 @@ libvirt_virNetworkGetDHCPLeases(PyObject *self ATTRIBUTE_UNUSED, #endif /* LIBVIR_CHECK_VERSION(1, 2, 6) */ + +#if LIBVIR_CHECK_VERSION(1, 2, 8) + +static PyObject * +libvirt_virDomainOpenGraphicsFD(PyObject *self ATTRIBUTE_UNUSED, PyObject *args) +{ + PyObject *py_retval; + int c_retval; + virDomainPtr domain; + PyObject *pyobj_domain; + int idx; + int fd; + unsigned int flags; + + if (!PyArg_ParseTuple(args, (char *)"Oii:virDomainOpenGraphicsFD", + &pyobj_domain, &idx, &flags)) + return NULL; + + domain = (virDomainPtr) PyvirDomain_Get(pyobj_domain); + + LIBVIRT_BEGIN_ALLOW_THREADS; + c_retval = virDomainOpenGraphicsFD(domain, idx, &fd, flags); + LIBVIRT_END_ALLOW_THREADS; + if (c_retval < 0) + return VIR_PY_NONE; + + py_retval = PyList_New(1); + PyList_SetItem(py_retval, 0, libvirt_intWrap(fd)); + return py_retval; +} + +#endif /* LIBVIR_CHECK_VERSION(1, 2, 8) */ + + /************************************************************************ * * * The registration stuff * @@ -8140,6 +8174,9 @@ static PyMethodDef libvirtMethods[] = { {(char *) "virNodeGetFreePages", libvirt_virNodeGetFreePages, METH_VARARGS, NULL}, {(char *) "virNetworkGetDHCPLeases", libvirt_virNetworkGetDHCPLeases, METH_VARARGS, NULL}, #endif /* LIBVIR_CHECK_VERSION(1, 2, 6) */ +#if LIBVIR_CHECK_VERSION(1, 2, 8) + {(char *) "virDomainOpenGraphicsFD", libvirt_virDomainOpenGraphicsFD, METH_VARARGS, NULL}, +#endif /* LIBVIR_CHECK_VERSION(1, 2, 8) */ {NULL, NULL, 0, NULL} }; -- 1.8.5.5

3 2

[libvirt] [PATCHv2 0/5] Bulk domain stats query API
by Peter Krempa 26 Aug '14

26 Aug '14

This version adds implementation in the qemu driver and virsh and fixes the following bugs from v1: *bad type of typed param count *off-by-one when counting and verifying domain objects in libvirt.c *off-by-one when allocating returned array in remote driver *the all domain stats func not working as the return value wasn't handled properly Peter Krempa (5): lib: Introduce API for retrieving bulk domain stats remote: Implement bulk domain stats APIs in the remote driver qemu: Implement bulk stats API and one of the stats groups to return virsh: domain: Split out code to lookup domain from string virsh: Implement command to excercise the bulk stats APIs daemon/remote.c | 91 ++++++++++++++++++++++ include/libvirt/libvirt.h.in | 26 +++++++ src/driver.h | 9 +++ src/libvirt.c | 180 +++++++++++++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 7 ++ src/qemu/qemu_driver.c | 175 +++++++++++++++++++++++++++++++++++++++++ src/remote/remote_driver.c | 88 +++++++++++++++++++++ src/remote/remote_protocol.x | 26 ++++++- src/remote_protocol-structs | 23 ++++++ tools/virsh-domain-monitor.c | 140 +++++++++++++++++++++++++++++++++ tools/virsh-domain.c | 80 ++++++++++++------- tools/virsh-domain.h | 4 + 12 files changed, 819 insertions(+), 30 deletions(-) -- 2.0.2

2 19

[libvirt] [Libvirt] segfault with patch "libxl: fix framebuffer port setting for HVM domains"
by Chris 26 Aug '14

26 Aug '14

Hello, I encountered segfaults with libvirt + libxl with this specific patch : http://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=b55cc5f4e31b488c4f9c3c… My Libvirt/Xen environment (on a slackware64-14.1) : virsh # version Compiled against library: libvirt 1.2.6 Using library: libvirt 1.2.6 Using API: Xen 1.2.6 Running hypervisor: Xen 4.4.0 My Libvirt "configure line" : ./configure \ --prefix=/usr \ --libdir=/usr/lib64 \ --bindir=/usr/sbin \ --sbindir=/usr/sbin \ --sysconfdir=/etc \ --localstatedir=/var \ --mandir=/usr/man \ --docdir=/usr/doc/libvirt-1.2.6 \ --enable-static=no \ --enable-shared=yes \ --with-yajl \ --with-qemu-group=root \ --with-init-script=none \ --without-xen \ --without-xenapi \ --with-libxl The XML of my VM : <domain type='xen'> <name>vm.infra.ld</name> <uuid>0b648359-099a-4d64-ba7a-12f32fb4f3dc</uuid> <memory unit='KiB'>1048576</memory> <currentMemory unit='KiB'>1048576</currentMemory> <vcpu placement='static'>2</vcpu> <os> <type arch='x86_64' machine='xenfv'>hvm</type> <loader>/usr/lib/xen/boot/hvmloader</loader> <boot dev='network'/> <boot dev='hd'/> </os> <features> <acpi/> <apic/> <pae/> <hap/> <viridian/> </features> <clock offset='variable' adjustment='0' basis='utc'> <timer name='hpet' present='no'/> </clock> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/lib/xen/bin/qemu-dm</emulator> <disk type='block' device='disk'> <driver name='phy'/> <source dev='/dev/data/vm.infra.ld-disk'/> <target dev='hda' bus='ide'/> <address type='drive' controller='0' bus='0' target='0' unit='0'/> </disk> <controller type='ide' index='0'/> <interface type='bridge'> <mac address='00:16:3e:xx:xx:xx'/> <source bridge='xenbr0'/> <script path='/etc/xen/scripts/vif-openvswitch'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target type='serial' port='0'/> </console> <input type='mouse' bus='ps2'/> <input type='keyboard' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes' listen='0.0.0.0' keymap='fr'> <listen type='address' address='0.0.0.0'/> </graphics> <video> <model type='cirrus' vram='9216' heads='1'/> </video> </devices> </domain> The VM starts fine with "virsh start" (cf attached libxl log), the VNC process is listening as requested on 0.0.0.0 but whenever I open a new virsh cli and send a command (e.g. "list"), libvirtd segfaults. I attached a gdb trace to this mail. If I rebuild libvirt without the patch mentioned earlier, no more segfaults, but of course the VNC process is listening on 127.0.0.1 instead of requested 0.0.0.0 when I start the VM. I may have done something wrong but don't understand what. Regards, Chris

3 4

[libvirt] [PATCHv2] blkdeviotune: check for overflow when parsing XML
by Erik Skultety 26 Aug '14

26 Aug '14

According to docs/schemas/domaincommon.rng and _virDomainBlockIoTuneInfo all the iotune values are interpreted as unsigned long long, however according to qemu_monitor_json.c, qemu silently truncates numbers larger than LLONG_MAX. There's really not much of a usage for such large numbers anyway yet. This patch provides the same overflow check during a domain start as it does during setting a blkdeviotune element in qemu_driver.c and thus reports an error when a larger number than LLONG_MAX is detected. --- src/qemu/qemu_command.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 6dac9d3..9213a60 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3620,6 +3620,18 @@ qemuBuildDriveStr(virConnectPtr conn, goto error; } + if (disk->blkdeviotune.total_bytes_sec > LLONG_MAX || + disk->blkdeviotune.read_bytes_sec > LLONG_MAX || + disk->blkdeviotune.write_bytes_sec > LLONG_MAX || + disk->blkdeviotune.total_iops_sec > LLONG_MAX || + disk->blkdeviotune.read_iops_sec > LLONG_MAX || + disk->blkdeviotune.write_iops_sec > LLONG_MAX) { + virReportError(VIR_ERR_OVERFLOW, + _("block I/O throttle limit must " + "be less than %llu using QEMU"), LLONG_MAX); + goto error; + } + if (disk->blkdeviotune.total_bytes_sec) { virBufferAsprintf(&opt, ",bps=%llu", disk->blkdeviotune.total_bytes_sec); -- 1.9.3

2 1