May 2019 - Devel - libvirt List Archives

[libvirt] [PATCH] docs: Link the relevant SEV formatdomain section from formatdomaincaps
by Erik Skultety 24 May '19

24 May '19

formatdomaincaps.html provides explanation of SEV fields, but doesn't link to the domain XML docs to show how it can be actually used in libvirt. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- docs/formatdomaincaps.html.in | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/docs/formatdomaincaps.html.in b/docs/formatdomaincaps.html.in index b31b1729f4..432d7aed5e 100644 --- a/docs/formatdomaincaps.html.in +++ b/docs/formatdomaincaps.html.in @@ -514,10 +514,19 @@ encrypted with a key unique to that VM.</p> <p> - For more details on SEV feature see: - <a href="https://support.amd.com/TechDocs/55766_SEV-KM_API_Specification.pdf"> - SEV API spec</a> and <a href="http://amd-dev.wpengine.netdna-cdn.com/wordpress/media/2013/12/AMD_Memory_E…"> - SEV White Paper</a> + For more details on the SEV feature and how to use it with libvirt have + a look at the following resources: + <dl> + <dd> + <a href="https://support.amd.com/TechDocs/55766_SEV-KM_API_Specification.pdf">SEV API spec</a> + </dd> + <dd> + <a href="http://amd-dev.wpengine.netdna-cdn.com/wordpress/media/2013/12/AMD_Memory_E…">SEV White Paper</a> + </dd> + <dd> + <a href="formatdomain.html#launchSecurity">SEV in domain XML</a> + </dd> + </dl> </p> <dl> -- 2.20.1

2 1

[libvirt] [PATCH] nodesuspend: report unsupported if systemd & pm-utils aren't available
by Daniel P. Berrangé 24 May '19

24 May '19

When libvirtd is run inside a container it is normal that neither systemd nor pm-utils will be available. In this case there is no way to suspend the host, so libvirt should just report the feature unsupported instead of raising an error. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/util/virnodesuspend.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/src/util/virnodesuspend.c b/src/util/virnodesuspend.c index 0af0908e4a..3fc5b93ac7 100644 --- a/src/util/virnodesuspend.c +++ b/src/util/virnodesuspend.c @@ -330,11 +330,10 @@ virNodeSuspendSupportsTarget(unsigned int target, bool *supported) if (ret == -2) ret = virNodeSuspendSupportsTargetPMUtils(target, supported); - /* If still unavailable, then report error */ + /* If still unavailable, then report unsupported */ if (ret == -2) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Cannot probe for supported suspend types")); - ret = -1; + *supported = false; + ret = 0; } return ret; -- 2.21.0

2 1

[libvirt] [PATCH] spec: Unconditionally set ownership of /var/lib/libvirt/sanlock
by Jiri Denemark 24 May '19

24 May '19

The libvirt-lock-sanlock subpackage requires sanlock to be installed first and the sanlock package creates the sanlock group on all distros we care about in the spec file (Fedora and RHEL >= 7). Thus instead of setting the ownership and permissions in a post scriptlet only when the sanlock group exists we can just install the directory with the appropriate metadata. https://bugzilla.redhat.com/show_bug.cgi?id=1702758 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- libvirt.spec.in | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index b159c50d8f..5bd3e30509 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1479,14 +1479,6 @@ exit 0 %postun client %systemd_postun libvirt-guests.service -%if %{with_sanlock} -%post lock-sanlock -if getent group sanlock > /dev/null ; then - chmod 0770 %{_localstatedir}/lib/libvirt/sanlock - chown root:sanlock %{_localstatedir}/lib/libvirt/sanlock -fi -%endif - %if %{with_lxc} %pre login-shell getent group virtlogin >/dev/null || groupadd -r virtlogin @@ -1739,7 +1731,7 @@ exit 0 %attr(0755, root, root) %{_libdir}/libvirt/lock-driver/sanlock.so %{_datadir}/augeas/lenses/libvirt_sanlock.aug %{_datadir}/augeas/lenses/tests/test_libvirt_sanlock.aug -%dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt/sanlock +%dir %attr(0770, root, sanlock) %{_localstatedir}/lib/libvirt/sanlock %{_sbindir}/virt-sanlock-cleanup %{_mandir}/man8/virt-sanlock-cleanup.8* %attr(0755, root, root) %{_libexecdir}/libvirt_sanlock_helper -- 2.21.0

2 1

[libvirt] [PATCH] driver: test: Fix the mingw build caused by wrong printf format specifier
by Erik Skultety 24 May '19

24 May '19

Caused by commit 326c3f54. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- Travis build: https://travis-ci.org/eskultety/libvirt/builds/536736446 Pushed under the build breaker rule. src/test/test_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index b29fd12903..7fd06fcfa8 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3268,7 +3268,7 @@ testDomainInterfaceAddresses(virDomainPtr dom, iface->addrs[0].type = VIR_IP_ADDR_TYPE_IPV4; iface->addrs[0].prefix = 24; - if (virAsprintf(&iface->addrs[0].addr, "192.168.0.%ld", 1 + i) < 0) + if (virAsprintf(&iface->addrs[0].addr, "192.168.0.%zu", 1 + i) < 0) goto cleanup; iface->naddrs = 1; -- 2.20.1

1 0

[libvirt] [PATCH] spec: Fix permissions of /var/run/libvirt/qemu
by Jiri Denemark 24 May '19

24 May '19

While libvirtd creates this directory with the default 0755 mode, the spec file stores 0700 in the RPM database. Thus RPM verification always complains about this directory. Let's fix the spec file to match reality. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Notes: Alternatively, we could change libvirt to create the directory with 0700 (instead of 0755), but all other drivers use 0755 (both in reality and in the spec file) and 0700 wouldn't really enhance security anyway. libvirt.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 970d2742ac..dc69920d75 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1661,7 +1661,7 @@ exit 0 %config(noreplace) %{_sysconfdir}/libvirt/qemu.conf %config(noreplace) %{_sysconfdir}/libvirt/qemu-lockd.conf %config(noreplace) %{_sysconfdir}/logrotate.d/libvirtd.qemu -%ghost %dir %attr(0700, root, root) %{_localstatedir}/run/libvirt/qemu/ +%ghost %dir %{_localstatedir}/run/libvirt/qemu/ %dir %attr(0751, %{qemu_user}, %{qemu_group}) %{_localstatedir}/lib/libvirt/qemu/ %dir %attr(0750, %{qemu_user}, %{qemu_group}) %{_localstatedir}/cache/libvirt/qemu/ %{_datadir}/augeas/lenses/libvirtd_qemu.aug -- 2.21.0

2 1

[libvirt] [PATCH v2 0/2] test_driver: add a guest inf and implement virDomainInterfaceAddresses
by Ilias Stamatis 24 May '19

24 May '19

Ilias Stamatis (2): test_driver: implement virDomainInterfaceAddresses test_driver: add a guest interface in the default config src/test/test_driver.c | 73 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) -- 2.21.0

2 4

[libvirt] Increasing TasksMax when creating machines via systemd
by Jim Fehlig 23 May '19

23 May '19

Hi All, I recently received an internal bug report of VM "crashing" due to hitting thread limits. Seems there was an assert in pthread_create within the VM when hitting the limit enforced by pids controller on the host Apr 28 07:45:46 lpcomp02007 kernel: cgroup: fork rejected by pids controller in /machine.slice/machine-qemu\x2d90028\x2dinstance\x2d0000634b.scope The user has TasksMax set to infinity in machine.slice, but apparently that is not inherited by child scopes and appears to be hardcoded to 16384 https://github.com/systemd/systemd/blob/51aba17b88617515e037e8985d3a4ea871a… The TasksMax property can be set when creating the machine as is done in the attached proof of concept patch. Question is whether this should be a tunable? My initial thought when seeing the report was TasksMax could be calculated based on number of vcpus, iothreads, emulator threads, etc. But it appears that could be quite tricky. The following mail thread describes the basic scenario encountered by my user http://lists.ceph.com/pipermail/ceph-users-ceph.com/2016-March/008174.html As you can see, many rbd images attached to a VM can result in an awful lot of threads. 300 images could result in 720K threads! We could punt and set the limit to infinity, but it exists for a reason - fork bomb prevention. A potential compromise between a hardcoded value and per-VM tunable is a driver tunable in qemu.conf. If a per-VM tunable is preferred, suggestions on where to place it and what to call it would be much appreciated :-). Regards, Jim

2 2

[libvirt] [PATCH v5 00/24] network: refactor to decouple virt drivers from network driver
by Daniel P. Berrangé 23 May '19

23 May '19

An update to v1: https://www.redhat.com/archives/libvir-list/2018-December/msg00681.html v2: https://www.redhat.com/archives/libvir-list/2019-February/msg01581.html v3: https://www.redhat.com/archives/libvir-list/2019-March/msg01259.html v4: https://www.redhat.com/archives/libvir-list/2019-April/msg01186.html Currently the network driver registers a set of callbacks with the virt driver in order to handle allocating/releasing network ports associated with guest NICs. This series introduces a virNetworkPortPtr object and associated XML that describes a network port. The virt drivers now call public APIs associated with this new object to create/delete ports for guest NICs. Changed in v5: - Separately track ports with type=network vs type=bridge to deal with fallout after reverting patches which merged them - Add RNG schema and XML format docs - Change to use classID attr on <bandwidth> isntead of of separate <class id="xx"/> element - Drop obsolete driver deps from RPM spec - Other misc bug fixes found during testing Changed in v4: - Merged the ACKd patches which didn't depend on other un-acked parts - Improve bandwidth error messages - Ensure we set floor sum to zero when starting network - Misc fixes for previous review comments NB, I have not added missing docs for the new XML doc format yet. This is work in progress. Changed in v3: - Remove unused API symbol - Fix dist of test data files Changed in v2: - Fix many bugs related to upgrades with running VMs - Convert over bandwidth controls to the new APIs - Handle reconnecting VIFs to bridges during startup - Much much more that I can't remember Daniel P. Berrangé (24): conf: allow bandwidth parsing / formatting to include class ID conf: introduce virNetworkPortDefPtr struct and XML support network: make networkLogAllocation independent of domain conf conf: add APIs to convert virDomainNetDef to virNetworkPortDef network: convert networkAllocateActualDevice to virNetworkPortDef network: convert networkNotifyActualDevice to virNetworkPortDef network: convert networkReleaseActualDevice to virNetworkPortDef network: convert hook script to take a network port XML network: remove the virDomainNetBandwidthChangeAllowed callback network: introduce networkAllocatePort network: introduce networkNotifyPort network: introduce networkReleasePort network: introduce networkUpdatePortBandwidth network: add public APIs for network port object access: add permissions for network port objects remote: add support for new network port APIs virsh: add support for network port APIs conf: support recording ports against virNetworkObjPtr network: add implementation of network port APIs lxc, libxl: notify network driver of NICs during reconnect lxc, libxl: save domain status after reconnect conf: record a portid against the domain conf conf: switch over to use network port APIs for virt drivers rpm: remove dependancy from qemu to network/storage drivers docs/docs.html.in | 1 + docs/formatdomain.html.in | 8 + docs/formatnetworkport.html.in | 212 +++ docs/hooks.html.in | 24 +- docs/schemas/domaincommon.rng | 5 + docs/schemas/networkport.rng | 165 +++ include/libvirt/libvirt-network.h | 122 ++ include/libvirt/virterror.h | 3 + libvirt.spec.in | 3 - src/access/genpolkit.pl | 2 +- src/access/viraccessdriver.h | 6 + src/access/viraccessdrivernop.c | 11 + src/access/viraccessdriverpolkit.c | 26 + src/access/viraccessdriverstack.c | 25 + src/access/viraccessmanager.c | 16 + src/access/viraccessmanager.h | 6 + src/access/viraccessperm.c | 6 + src/access/viraccessperm.h | 44 + src/conf/Makefile.inc.am | 2 + src/conf/domain_conf.c | 531 +++++++- src/conf/domain_conf.h | 49 +- src/conf/netdev_bandwidth_conf.c | 30 +- src/conf/netdev_bandwidth_conf.h | 2 + src/conf/network_conf.c | 8 +- src/conf/virnetworkobj.c | 303 +++++ src/conf/virnetworkobj.h | 34 + src/conf/virnetworkportdef.c | 509 +++++++ src/conf/virnetworkportdef.h | 113 ++ src/datatypes.c | 60 + src/datatypes.h | 41 + src/driver-network.h | 41 + src/libvirt-network.c | 444 +++++++ src/libvirt_private.syms | 23 +- src/libvirt_public.syms | 15 + src/libxl/libxl_driver.c | 33 + src/lxc/lxc_process.c | 35 + src/network/bridge_driver.c | 1182 ++++++++++------- src/qemu/qemu_driver.c | 8 +- src/remote/remote_daemon_dispatch.c | 73 + src/remote/remote_driver.c | 69 + src/remote/remote_protocol.x | 124 +- src/remote_protocol-structs | 69 + src/rpc/gendispatch.pl | 18 +- src/util/virerror.c | 9 + src/util/virhook.c | 4 +- src/util/virhook.h | 4 +- tests/Makefile.am | 7 + .../net-virtio-network-portgroup.xml | 6 +- tests/virnetdevbandwidthtest.c | 1 + .../plug-bridge-mactbl.xml | 9 + .../virnetworkportxml2xmldata/plug-bridge.xml | 15 + .../virnetworkportxml2xmldata/plug-direct.xml | 12 + .../plug-hostdev-pci.xml | 12 + .../plug-network.xml | 16 + tests/virnetworkportxml2xmldata/plug-none.xml | 8 + tests/virnetworkportxml2xmltest.c | 104 ++ tests/virschematest.c | 1 + tools/virsh-completer.c | 50 + tools/virsh-completer.h | 4 + tools/virsh-network.c | 399 +++++- tools/virsh-network.h | 5 + 61 files changed, 4506 insertions(+), 661 deletions(-) create mode 100644 docs/formatnetworkport.html.in create mode 100644 docs/schemas/networkport.rng create mode 100644 src/conf/virnetworkportdef.c create mode 100644 src/conf/virnetworkportdef.h create mode 100644 tests/virnetworkportxml2xmldata/plug-bridge-mactbl.xml create mode 100644 tests/virnetworkportxml2xmldata/plug-bridge.xml create mode 100644 tests/virnetworkportxml2xmldata/plug-direct.xml create mode 100644 tests/virnetworkportxml2xmldata/plug-hostdev-pci.xml create mode 100644 tests/virnetworkportxml2xmldata/plug-network.xml create mode 100644 tests/virnetworkportxml2xmldata/plug-none.xml create mode 100644 tests/virnetworkportxml2xmltest.c -- 2.21.0

2 36

[libvirt] [PATCH] virDomainDefPostParse: use DOMAIN_DEVICE_ITERATE_MISSING_INFO
by Ján Tomko 23 May '19

23 May '19

Apart from virDomainDefValidate, virDomainDefPostParse is another place where operating on info-less devices makes sense. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- src/conf/domain_conf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 926c2139b7..006920e954 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -5803,7 +5803,8 @@ virDomainDefPostParse(virDomainDefPtr def, /* iterate the devices */ ret = virDomainDeviceInfoIterateInternal(def, virDomainDefPostParseDeviceIterator, - DOMAIN_DEVICE_ITERATE_ALL_CONSOLES, + DOMAIN_DEVICE_ITERATE_ALL_CONSOLES | + DOMAIN_DEVICE_ITERATE_MISSING_INFO, &data); if (virDomainDefPostParseCheckFailure(def, parseFlags, ret) < 0) -- 2.19.2

2 1

[libvirt] [PATCH 0/2] Introduce virDomainDeviceIterate
by Ján Tomko 23 May '19

23 May '19

A function for iterating over all devices [0] instead of just the ones with DeviceInfo. [0] Terms and conditions apply. Leases might be included as well. Ján Tomko (2): Introduce virDomainDeviceIterate Introduce DOMAIN_DEVICE_ITERATE_MISSING_INFO src/conf/domain_conf.c | 47 ++++++++++++++++++++++++++++++------------ src/conf/domain_conf.h | 3 +++ 2 files changed, 37 insertions(+), 13 deletions(-) -- 2.19.2

2 5