- Devel - libvirt List Archives

[libvirt] [PATCHv2 0/5] manage the shmem device source
by Luyao Huang 27 Aug '15

27 Aug '15

v1 link: https://www.redhat.com/archives/libvir-list/2015-July/msg00926.html v2 different: -Introduce new xml element shareable to indicate whether the device is intended to be shared across multiple guests. -Make virSecuritySELinuxRestoreSecurityAllLabel and virSecuritySELinuxSetSecurityAllLabel call set/restore label -Move struct _virShmObject and _virShmObjectList in virshm.c and introduce some functions to get the parameter. -Change the return value to -1 when the function are not supported on some platform. -add the code in src/security/security_nop.c -some small pieces fix Since there is a shmobj leak when let qemu create shmobj by themselves, also the label of shmobj/shmem-server socket is not right. Guest cannot direct use the shmem-server if users enabled selinux. So it will be better to manage it in libvirt. The way i chosed is region the shmem deivce in a list, and save it status to a local file to avoid losing it after restart libvirtd, and count the guest which use it, and let the callers know if there is no guest is using it (then we can relabel/cleanup some resource). About shmem-server we decided introduce some new selinux label to fix that, however ivshmem server still not finished right now (i mean patches were still being reviewed in qemu-devel), we could implement the set up ivshmem-server part later. BTW, during some research i noticed the vhost-user network should have the same problem like the ivshmem-server, since it use the same way to connect to guest (a host app connect to guest via socket), and after some test i found the guest cannot connect to the sockect which created by app if i enable the Selinux. So i think maybe we could fix these two issue together, if there is already have a way to make vhost-usr works well with SElinux, i think we could use that way in this place. Luyao Huang (5): conf: introduce seclabels in shmem device element conf: introduce shareable in shmem device element util: introduce new helpers to manage shmem device security: add security part for shmem device qemu: call the helpers in virshm.c to manage shmem device configure.ac | 10 + docs/formatdomain.html.in | 12 + docs/schemas/domaincommon.rng | 8 + po/POTFILES.in | 3 +- src/Makefile.am | 7 +- src/conf/domain_conf.c | 103 ++- src/conf/domain_conf.h | 6 + src/libvirt_private.syms | 26 + src/qemu/qemu_conf.h | 3 + src/qemu/qemu_driver.c | 4 + src/qemu/qemu_process.c | 157 +++++ src/security/security_dac.c | 100 +++ src/security/security_driver.h | 9 + src/security/security_manager.c | 35 + src/security/security_manager.h | 6 + src/security/security_nop.c | 19 + src/security/security_selinux.c | 97 +++ src/security/security_stack.c | 39 ++ src/util/virshm.c | 739 +++++++++++++++++++++ src/util/virshm.h | 92 +++ .../qemuxml2argv-shmem-seclabel.xml | 55 ++ .../qemuxml2argv-shmem-shareable.xml | 43 ++ tests/qemuxml2xmltest.c | 5 + 23 files changed, 1545 insertions(+), 33 deletions(-) create mode 100644 src/util/virshm.c create mode 100644 src/util/virshm.h create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-shmem-seclabel.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-shmem-shareable.xml -- 1.8.3.1

1 5

[libvirt] [python PATCH] examples: small fix for nodestats.py example
by Luyao Huang 27 Aug '15

27 Aug '15

Add nodestats.py in MANIFEST.in and add a small description for nodestats.py in README Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- MANIFEST.in | 1 + examples/README | 3 +++ 2 files changed, 4 insertions(+) diff --git a/MANIFEST.in b/MANIFEST.in index b5ba783..0d66f9c 100644 --- a/MANIFEST.in +++ b/MANIFEST.in @@ -13,6 +13,7 @@ include examples/esxlist.py include examples/event-test.py include examples/guest-vcpus/guest-vcpu-daemon.py include examples/guest-vcpus/guest-vcpu.py +include examples/nodestats.py include examples/topology.py include generator.py include libvirt-lxc-override-api.xml diff --git a/examples/README b/examples/README index 0cb4513..13211b6 100644 --- a/examples/README +++ b/examples/README @@ -14,6 +14,9 @@ dhcpleases.py - list dhcp leases for a given virtual network domipaddrs.py - list IP addresses for guest domains guest-vcpus - two helpers to make the guest agent event useful with agent based vCPU state modification +nodestats.py - print total memory and free memory for each host NUMA node and + the memory strictly bound to certain host nodes for each running + domain. The XML files in this directory are examples of the XML format that libvirt expects, and will have to be adapted for your setup. They are only needed -- 1.8.3.1

3 2

[libvirt] [PATCHv2 0/2] qemu: fix the audit log is not correct for memory device
by Luyao Huang 27 Aug '15

27 Aug '15

First review: http://www.redhat.com/archives/libvir-list/2015-July/msg00982.html Change in v2: - split it to two patches - fix some small mistake in hot-unplug part I change the code in qemuDomainAttachMemory like what we do in other attach deivce functions. And I have removed the jump in qemuDomainRemoveMemoryDevice, since i just see only one place need jump to audit, so i refactor the code (thanks John's advise). Luyao Huang (2): qemu: fix the audit log is not correct after hot-plug memory success qemu: fix the audit log is not correct after hot-unplug memory device src/qemu/qemu_hotplug.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-) -- 1.8.3.1

3 4

[libvirt] [PATCH] virsh: Remove <backingStore> when changing cdrom media source
by Peter Krempa 27 Aug '15

27 Aug '15

Since the code is changing the source image path by modifying the existing XML snippet the <backingStore> element does not get dropped. The element is ignored though but it might start being used in the future. Before this patch, you'd get: $ virsh change-media --eject --print-xml 10 hdc <disk type="file" device="cdrom"> <driver name="qemu" type="qcow2"/> <backingStore type="file" index="1"> <format type="qcow2"/> <source file="/var/lib/libvirt/images/vm.1436949097"/> <backingStore/> </backingStore> <target dev="hdc" bus="ide"/> ... </disk> After: $ virsh change-media --eject --print-xml 10 hdc <disk type="file" device="cdrom"> <driver name="qemu" type="qcow2"/> <target dev="hdc" bus="ide"/> ... </disk> --- tools/virsh-domain.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index a61656f..d4d6987 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -11415,6 +11415,7 @@ vshUpdateDiskXML(xmlNodePtr disk_node, vshUpdateDiskXMLType type) { xmlNodePtr source = NULL; + xmlNodePtr child; char *device_type = NULL; char *ret = NULL; @@ -11430,10 +11431,16 @@ vshUpdateDiskXML(xmlNodePtr disk_node, } /* find the current source subelement */ - for (source = disk_node->children; source; source = source->next) { - if (source->type == XML_ELEMENT_NODE && - xmlStrEqual(source->name, BAD_CAST "source")) - break; + for (child = disk_node->children; child; child = child->next) { + if (child->type == XML_ELEMENT_NODE && + xmlStrEqual(child->name, BAD_CAST "source")) + source = child; + + if (child->type == XML_ELEMENT_NODE && + xmlStrEqual(child->name, BAD_CAST "backingStore")) { + xmlUnlinkNode(child); + xmlFreeNode(child); + } } if (type == VSH_UPDATE_DISK_XML_EJECT) { -- 2.4.5

2 1

[libvirt] [PATCH] virt-aa-helper: add NVRAM store file for read/write
by Peter Kieser 26 Aug '15

26 Aug '15

Some UEFI firmwares may want to use a non-volatile memory to store some variables. If AppArmor is enabled, and NVRAM store file is set currently virt-aa-helper does not add the NVRAM store file to the template. Add this file for read/write when this functionality is defined in domain XML. Signed-off-by: Peter Kieser <peter(a)kieser.ca> --- src/security/virt-aa-helper.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 4ce1e7a..2f93172 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -1047,6 +1047,10 @@ get_files(vahControl * ctl) if (vah_add_file(&buf, ctl->def->os.loader->path, "r") != 0) goto cleanup; + if (ctl->def->os.loader && ctl->def->os.loader->nvram) + if (vah_add_file(&buf, ctl->def->os.loader->nvram, "rw") != 0) + goto cleanup; + for (i = 0; i < ctl->def->ngraphics; i++) { if (ctl->def->graphics[i]->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC && ctl->def->graphics[i]->data.vnc.socket &&

3 2

[libvirt] [PATCH] Fix link to page for Virtuozzo driver
by Sergey Bronnikov 26 Aug '15

26 Aug '15

--- docs/index.html.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/index.html.in b/docs/index.html.in index 7b0c149..c4025de 100644 --- a/docs/index.html.in +++ b/docs/index.html.in @@ -72,7 +72,7 @@ The <a href="http://libvirt.org/drvphyp.html">IBM PowerVM</a> hypervisor </li> <li> - The <a href="http://libvirt.org/drvirtuozzo.html">Virtuozzo</a> hypervisor + The <a href="http://libvirt.org/drvvirtuozzo.html">Virtuozzo</a> hypervisor </li> <li> The <a href="http://libvirt.org/drvbhyve.html">Bhyve</a> hypervisor

2 1

[libvirt] [PATCH] hostdev: skip ACS check when using VFIO for device assignment
by Laine Stump 26 Aug '15

26 Aug '15

The ACS checks are meaningless when using the more modern VFIO driver for device assignment since VFIO has its own more complete and exact checks, but I didn't realize that when I added support for VFIO. This patch eliminates the ACS check when preparing PCI devices for assignment if VFIO is being used. This resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1256486 --- src/util/virhostdev.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/src/util/virhostdev.c b/src/util/virhostdev.c index 529753c..0bc5120 100644 --- a/src/util/virhostdev.c +++ b/src/util/virhostdev.c @@ -1,6 +1,6 @@ /* virhostdev.c: hostdev management * - * Copyright (C) 2006-2007, 2009-2013 Red Hat, Inc. + * Copyright (C) 2006-2007, 2009-2015 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * Copyright (C) 2014 SUSE LINUX Products GmbH, Nuernberg, Germany. * @@ -230,7 +230,6 @@ virHostdevGetPCIHostDeviceList(virDomainHostdevDefPtr *hostdevs, int nhostdevs) virObjectUnref(list); return NULL; } - if (virPCIDeviceListAdd(list, dev) < 0) { virPCIDeviceFree(dev); virObjectUnref(list); @@ -579,7 +578,7 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, struct virHostdevIsPCINodeDeviceUsedData data = {hostdev_mgr, dom_name, usesVfio}; - if (!virPCIDeviceIsAssignable(dev, strict_acs_check)) { + if (!usesVfio && !virPCIDeviceIsAssignable(dev, strict_acs_check)) { virReportError(VIR_ERR_OPERATION_INVALID, _("PCI device %s is not assignable"), virPCIDeviceGetName(dev)); -- 2.1.0

2 2

[libvirt] [PATCH] Start daemon only after filesystems are mounted
by Martin Kletzander 26 Aug '15

26 Aug '15

When images are on mounted filesystems, there is no guarantee libvirtd will start before they are mounted. In systemd world, this is done with 'After=systemd.mount' in the service file. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- daemon/libvirtd.service.in | 1 + 1 file changed, 1 insertion(+) diff --git a/daemon/libvirtd.service.in b/daemon/libvirtd.service.in index 1759ac8a0946..c461f9462ae5 100644 --- a/daemon/libvirtd.service.in +++ b/daemon/libvirtd.service.in @@ -5,6 +5,7 @@ After=network.target After=dbus.service After=iscsid.service After=apparmor.service +After=systemd.mount Documentation=man:libvirtd(8) Documentation=http://libvirt.org -- 2.5.0

2 3

[libvirt] [PATCH v4] lxc: Inherit namespace feature
by ik.nitk 26 Aug '15

26 Aug '15

This patch adds feature for lxc containers to inherit namespaces. This is very similar to what lxc-tools or docker provides. Look for "man lxc-start" and you will find that you can pass command args as [ --share-[net|ipc|uts] name|pid ]. Or check out docker networking option in which you can give --net=container:NAME_or_ID as an option for sharing +namespace. >From this patch you can add extra libvirt option to share namespace in following way. <lxc:namespace> <lxc:sharenet type='netns' value='red'/> <lxc:shareipc type='pid' value='12345'/> <lxc:shareuts type='name' value='container1'/> </lxc:namespace> The netns option is specific to sharenet. It can be used to inherit from existing network namespace. --- docs/drvlxc.html.in | 21 +++++ docs/schemas/domaincommon.rng | 42 +++++++++ po/POTFILES.in | 1 + src/Makefile.am | 7 +- src/lxc/lxc_conf.c | 2 +- src/lxc/lxc_container.c | 71 +++++++++++++-- src/lxc/lxc_container.h | 2 + src/lxc/lxc_controller.c | 57 +++++++++++- src/lxc/lxc_domain.c | 149 ++++++++++++++++++++++++++++++++ src/lxc/lxc_domain.h | 26 ++++++ src/lxc/lxc_process.c | 157 ++++++++++++++++++++++++++++++++++ tests/lxcxml2xmldata/lxc-sharenet.xml | 33 +++++++ tests/lxcxml2xmltest.c | 1 + 13 files changed, 560 insertions(+), 9 deletions(-) create mode 100644 tests/lxcxml2xmldata/lxc-sharenet.xml diff --git a/docs/drvlxc.html.in b/docs/drvlxc.html.in index a094bd9..d6c57c4 100644 --- a/docs/drvlxc.html.in +++ b/docs/drvlxc.html.in @@ -590,6 +590,27 @@ Note that allowing capabilities that are normally dropped by default can serious affect the security of the container and the host. </p> +<h2><a name="share">Inherit namespaces</a></h2> + +<p> +Libvirt allows you to inherit the namespace from container/process just like lxc tools +or docker provides to share the network namespace. The following can be used to share +required namespaces. If we want to share only one then the other namespaces can be ignored. +The netns option is specific to sharenet. It can be used in cases we want to use existing network namespace +rather than creating new network namespace for the container. In this case privnet option will be +ignored. +</p> +<pre> +<domain type='lxc' xmlns:lxc='http://libvirt.org/schemas/domain/lxc/1.0'> +... +<lxc:namespace> + <lxc:sharenet type='netns' value='red'/> + <lxc:shareuts type='name' value='container1'/> + <lxc:shareipc type='pid' value='12345'/> +</lxc:namespace> +</domain> +</pre> + <h2><a name="usage">Container usage / management</a></h2> <p> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 043c975..fa026cd 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -68,6 +68,9 @@ <ref name='qemucmdline'/> </optional> <optional> + <ref name='lxcsharens'/> + </optional> + <optional> <ref name='keywrap'/> </optional> </interleave> @@ -5057,6 +5060,45 @@ </element> </define> +  + <define name="lxcsharens"> + <element name="namespace" ns="http://libvirt.org/schemas/domain/lxc/1.0"> + <zeroOrMore> + <element name="sharenet"> + <attribute name="type"> + <choice> + <value>netns</value> + <value>name</value> + <value>pid</value> + </choice> + </attribute> + <attribute name='value'/> + </element> + <element name="shareipc"> + <attribute name="type"> + <choice> + <value>name</value> + <value>pid</value> + </choice> + </attribute> + <attribute name='value'/> + </element> + <element name="shareuts"> + <attribute name="type"> + <choice> + <value>name</value> + <value>pid</value> + </choice> + </attribute> + <attribute name='value'/> + </element> + </zeroOrMore> + </element> + </define> + <define name="metadata"> <element name="metadata"> <zeroOrMore> diff --git a/po/POTFILES.in b/po/POTFILES.in index 1e52e6a..46220f7 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -85,6 +85,7 @@ src/lxc/lxc_native.c src/lxc/lxc_container.c src/lxc/lxc_conf.c src/lxc/lxc_controller.c +src/lxc/lxc_domain.c src/lxc/lxc_driver.c src/lxc/lxc_process.c src/libxl/libxl_domain.c diff --git a/src/Makefile.am b/src/Makefile.am index c4d49a5..24d31e1 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -1320,7 +1320,12 @@ libvirt_driver_lxc_impl_la_CFLAGS = \ -I$(srcdir)/access \ -I$(srcdir)/conf \ $(AM_CFLAGS) -libvirt_driver_lxc_impl_la_LIBADD = $(CAPNG_LIBS) $(LIBNL_LIBS) $(FUSE_LIBS) +libvirt_driver_lxc_impl_la_LIBADD = \ + $(CAPNG_LIBS) \ + $(LIBNL_LIBS) \ + $(LIBXML_LIBS) \ + $(FUSE_LIBS) + if WITH_BLKID libvirt_driver_lxc_impl_la_CFLAGS += $(BLKID_CFLAGS) libvirt_driver_lxc_impl_la_LIBADD += $(BLKID_LIBS) diff --git a/src/lxc/lxc_conf.c b/src/lxc/lxc_conf.c index b689b92..8ada531 100644 --- a/src/lxc/lxc_conf.c +++ b/src/lxc/lxc_conf.c @@ -213,7 +213,7 @@ lxcDomainXMLConfInit(void) { return virDomainXMLOptionNew(&virLXCDriverDomainDefParserConfig, &virLXCDriverPrivateDataCallbacks, - NULL); + &virLXCDriverDomainXMLNamespace); } diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 11e9514..8011ed0 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -27,6 +27,7 @@ #include <config.h> #include <fcntl.h> +#include <sched.h> #include <limits.h> #include <stdlib.h> #include <stdio.h> @@ -38,7 +39,6 @@ #include <mntent.h> #include <sys/reboot.h> #include <linux/reboot.h> - /* Yes, we want linux private one, for _syscall2() macro */ #include <linux/unistd.h> @@ -111,6 +111,7 @@ struct __lxc_child_argv { size_t nttyPaths; char **ttyPaths; int handshakefd; + int *nsInheritFDs; }; static int lxcContainerMountFSBlock(virDomainFSDefPtr fs, @@ -2144,6 +2145,35 @@ static int lxcContainerDropCapabilities(virDomainDefPtr def ATTRIBUTE_UNUSED, /** + * lxcAttach_ns: + * @ns_fd: array of namespaces to attach + */ +static int lxcAttachNS(int *ns_fd) +{ + size_t i; + if (ns_fd) + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) { + if (ns_fd[i] < 0) + continue; + VIR_DEBUG("Setting into namespace\n"); + /* We get EINVAL if new NS is same as the current + * NS, or if the fd namespace doesn't match the + * type passed to setns()'s second param. Since we + * pass 0, we know the EINVAL is harmless + */ + if (setns(ns_fd[i], 0) < 0 && + errno != EINVAL) { + virReportSystemError(errno, _("failed to set namespace '%s'"), + virLXCDomainNamespaceTypeToString(i)); + return -1; + } + VIR_FORCE_CLOSE(ns_fd[i]); + } + return 0; +} + + +/** * lxcContainerChild: * @data: pointer to container arguments * @@ -2172,6 +2202,12 @@ static int lxcContainerChild(void *data) goto cleanup; } + if (lxcAttachNS(argv->nsInheritFDs) < 0) { + virReportError(VIR_ERR_SYSTEM_ERROR, "%s", + _("failed to attach the namespace")); + return -1; + } + /* Wait for controller to finish setup tasks, including * things like move of network interfaces, uid/gid mapping */ @@ -2342,6 +2378,7 @@ int lxcContainerStart(virDomainDefPtr def, int *passFDs, int control, int handshakefd, + int *nsInheritFDs, size_t nttyPaths, char **ttyPaths) { @@ -2359,7 +2396,8 @@ int lxcContainerStart(virDomainDefPtr def, .monitor = control, .nttyPaths = nttyPaths, .ttyPaths = ttyPaths, - .handshakefd = handshakefd + .handshakefd = handshakefd, + .nsInheritFDs = nsInheritFDs, }; /* allocate a stack for the container */ @@ -2368,7 +2406,7 @@ int lxcContainerStart(virDomainDefPtr def, stacktop = stack + stacksize; - cflags = CLONE_NEWPID|CLONE_NEWNS|CLONE_NEWUTS|CLONE_NEWIPC|SIGCHLD; + cflags = CLONE_NEWPID|CLONE_NEWNS|SIGCHLD; if (userns_required(def)) { if (userns_supported()) { @@ -2381,10 +2419,31 @@ int lxcContainerStart(virDomainDefPtr def, return -1; } } + if (!nsInheritFDs || nsInheritFDs[VIR_LXC_DOMAIN_NAMESPACE_SHARENET] == -1) { + if (lxcNeedNetworkNamespace(def)) { + VIR_DEBUG("Enable network namespaces"); + cflags |= CLONE_NEWNET; + } + } else { + if (lxcNeedNetworkNamespace(def)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Config askes for inherit net namespace " + "as well as private network interfaces")); + return -1; + } + VIR_DEBUG("Inheriting a net namespace"); + } + + if (!nsInheritFDs || nsInheritFDs[VIR_LXC_DOMAIN_NAMESPACE_SHAREIPC] == -1) { + cflags |= CLONE_NEWIPC; + } else { + VIR_DEBUG("Inheriting an IPC namespace"); + } - if (lxcNeedNetworkNamespace(def)) { - VIR_DEBUG("Enable network namespaces"); - cflags |= CLONE_NEWNET; + if (!nsInheritFDs || nsInheritFDs[VIR_LXC_DOMAIN_NAMESPACE_SHAREUTS] == -1) { + cflags |= CLONE_NEWUTS; + } else { + VIR_DEBUG("Inheriting a UTS namespace"); } VIR_DEBUG("Cloning container init process"); diff --git a/src/lxc/lxc_container.h b/src/lxc/lxc_container.h index 67292ab..33eaab4 100644 --- a/src/lxc/lxc_container.h +++ b/src/lxc/lxc_container.h @@ -25,6 +25,7 @@ # define LXC_CONTAINER_H # include "lxc_conf.h" +# include "lxc_domain.h" # include "security/security_manager.h" enum { @@ -60,6 +61,7 @@ int lxcContainerStart(virDomainDefPtr def, int *passFDs, int control, int handshakefd, + int *nsInheritFDs, size_t nttyPaths, char **ttyPaths); diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index 48a3597..a94e819 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -119,6 +119,8 @@ struct _virLXCController { size_t npassFDs; int *passFDs; + int *nsFDs; + size_t nconsoles; virLXCControllerConsolePtr consoles; char *devptmx; @@ -287,6 +289,7 @@ static void virLXCControllerFree(virLXCControllerPtr ctrl) VIR_FREE(ctrl->nbdpids); + VIR_FREE(ctrl->nsFDs); virCgroupFree(&ctrl->cgroup); /* This must always be the last thing to be closed */ @@ -2391,6 +2394,7 @@ virLXCControllerRun(virLXCControllerPtr ctrl) ctrl->passFDs, control[1], containerhandshake[1], + ctrl->nsFDs, ctrl->nconsoles, containerTTYPaths)) < 0) goto cleanup; @@ -2400,6 +2404,10 @@ virLXCControllerRun(virLXCControllerPtr ctrl) for (i = 0; i < ctrl->npassFDs; i++) VIR_FORCE_CLOSE(ctrl->passFDs[i]); + if (ctrl->nsFDs) + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) + VIR_FORCE_CLOSE(ctrl->nsFDs[i]); + if (virLXCControllerSetupCgroupLimits(ctrl) < 0) goto cleanup; @@ -2468,6 +2476,7 @@ int main(int argc, char *argv[]) const char *name = NULL; size_t nveths = 0; char **veths = NULL; + int ns_fd[VIR_LXC_DOMAIN_NAMESPACE_LAST]; int handshakeFd = -1; bool bg = false; const struct option options[] = { @@ -2478,6 +2487,9 @@ int main(int argc, char *argv[]) { "passfd", 1, NULL, 'p' }, { "handshakefd", 1, NULL, 's' }, { "security", 1, NULL, 'S' }, + { "share-net", 1, NULL, 'N' }, + { "share-ipc", 1, NULL, 'I' }, + { "share-uts", 1, NULL, 'U' }, { "help", 0, NULL, 'h' }, { 0, 0, 0, 0 }, }; @@ -2489,6 +2501,9 @@ int main(int argc, char *argv[]) size_t i; const char *securityDriver = "none"; + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) + ns_fd[i] = -1; + if (setlocale(LC_ALL, "") == NULL || bindtextdomain(PACKAGE, LOCALEDIR) == NULL || textdomain(PACKAGE) == NULL || @@ -2504,7 +2519,7 @@ int main(int argc, char *argv[]) while (1) { int c; - c = getopt_long(argc, argv, "dn:v:p:m:c:s:h:S:", + c = getopt_long(argc, argv, "dn:v:p:m:c:s:h:S:N:I:U:", options, NULL); if (c == -1) @@ -2552,6 +2567,30 @@ int main(int argc, char *argv[]) } break; + case 'N': + if (virStrToLong_i(optarg, NULL, 10, &ns_fd[VIR_LXC_DOMAIN_NAMESPACE_SHARENET]) < 0) { + fprintf(stderr, "malformed --share-net argument '%s'", + optarg); + goto cleanup; + } + break; + + case 'I': + if (virStrToLong_i(optarg, NULL, 10, &ns_fd[VIR_LXC_DOMAIN_NAMESPACE_SHAREIPC]) < 0) { + fprintf(stderr, "malformed --share-ipc argument '%s'", + optarg); + goto cleanup; + } + break; + + case 'U': + if (virStrToLong_i(optarg, NULL, 10, &ns_fd[VIR_LXC_DOMAIN_NAMESPACE_SHAREUTS]) < 0) { + fprintf(stderr, "malformed --share-uts argument '%s'", + optarg); + goto cleanup; + } + break; + case 'S': securityDriver = optarg; break; @@ -2569,6 +2608,9 @@ int main(int argc, char *argv[]) fprintf(stderr, " -v VETH, --veth VETH\n"); fprintf(stderr, " -s FD, --handshakefd FD\n"); fprintf(stderr, " -S NAME, --security NAME\n"); + fprintf(stderr, " -N FD, --share-net FD\n"); + fprintf(stderr, " -I FD, --share-ipc FD\n"); + fprintf(stderr, " -U FD, --share-uts FD\n"); fprintf(stderr, " -h, --help\n"); fprintf(stderr, "\n"); goto cleanup; @@ -2621,6 +2663,19 @@ int main(int argc, char *argv[]) ctrl->passFDs = passFDs; ctrl->npassFDs = npassFDs; + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) { + if (ns_fd[i] != -1) { + if (!ctrl->nsFDs) {/*allocate only once */ + size_t j = 0; + if (VIR_ALLOC_N(ctrl->nsFDs, VIR_LXC_DOMAIN_NAMESPACE_LAST) < 0) + goto cleanup; + for (j = 0; j < VIR_LXC_DOMAIN_NAMESPACE_LAST; j++) + ctrl->nsFDs[j] = -1; + } + ctrl->nsFDs[i] = ns_fd[i]; + } + } + for (i = 0; i < nttyFDs; i++) { if (virLXCControllerAddConsole(ctrl, ttyFDs[i]) < 0) goto cleanup; diff --git a/src/lxc/lxc_domain.c b/src/lxc/lxc_domain.c index 2f377d8..e3da9f0 100644 --- a/src/lxc/lxc_domain.c +++ b/src/lxc/lxc_domain.c @@ -26,8 +26,13 @@ #include "viralloc.h" #include "virlog.h" #include "virerror.h" +#include <libxml/xpathInternals.h> +#include "virstring.h" +#include "virutil.h" +#include "virfile.h" #define VIR_FROM_THIS VIR_FROM_LXC +#define LXC_NAMESPACE_HREF "http://libvirt.org/schemas/domain/lxc/1.0" VIR_LOG_INIT("lxc.lxc_domain"); @@ -41,6 +46,150 @@ static void *virLXCDomainObjPrivateAlloc(void) return priv; } +VIR_ENUM_IMPL(virLXCDomainNamespace, + VIR_LXC_DOMAIN_NAMESPACE_LAST, + "sharenet", + "shareipc", + "shareuts") + +VIR_ENUM_IMPL(virLXCDomainNamespaceSource, + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_LAST, + "none", + "name", + "pid", + "netns") + +static void +lxcDomainDefNamespaceFree(void *nsdata) +{ + size_t i; + lxcDomainDefPtr lxcDef = nsdata; + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) + VIR_FREE(lxcDef->ns_val[i]); + VIR_FREE(nsdata); +} + +static int +lxcDomainDefNamespaceParse(xmlDocPtr xml ATTRIBUTE_UNUSED, + xmlNodePtr root ATTRIBUTE_UNUSED, + xmlXPathContextPtr ctxt, + void **data) +{ + lxcDomainDefPtr lxcDef = NULL; + xmlNodePtr *nodes = NULL; + bool uses_lxc_ns = false; + xmlNodePtr node; + int feature; + int n; + char *tmp = NULL; + size_t i; + + if (xmlXPathRegisterNs(ctxt, BAD_CAST "lxc", BAD_CAST LXC_NAMESPACE_HREF) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to register xml namespace '%s'"), + LXC_NAMESPACE_HREF); + return -1; + } + + if (VIR_ALLOC(lxcDef) < 0) + return -1; + + node = ctxt->node; + if ((n = virXPathNodeSet("./lxc:namespace/*", ctxt, &nodes)) < 0) + goto error; + uses_lxc_ns |= n > 0; + + for (i = 0; i < n; i++) { + if ((feature = virLXCDomainNamespaceTypeFromString( + (const char *) nodes[i]->name)) < 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unsupported Namespace feature: %s"), + nodes[i]->name); + goto error; + } + + ctxt->node = nodes[i]; + + if (!(tmp = virXMLPropString(nodes[i], "type"))) { + virReportError(VIR_ERR_INTERNAL_ERROR, + "%s", _("No lxc environment type specified")); + goto error; + } + if ((lxcDef->ns_source[feature] = + virLXCDomainNamespaceSourceTypeFromString(tmp)) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unknown LXC namespace source '%s'"), + tmp); + VIR_FREE(tmp); + goto error; + } + VIR_FREE(tmp); + + if (!(lxcDef->ns_val[feature] = + virXMLPropString(nodes[i], "value"))) { + virReportError(VIR_ERR_INTERNAL_ERROR, + "%s", _("No lxc environment type specified")); + goto error; + } + } + VIR_FREE(nodes); + ctxt->node = node; + if (uses_lxc_ns) + *data = lxcDef; + else + VIR_FREE(lxcDef); + return 0; + error: + VIR_FREE(nodes); + lxcDomainDefNamespaceFree(lxcDef); + return -1; +} + + +static int +lxcDomainDefNamespaceFormatXML(virBufferPtr buf, + void *nsdata) +{ + lxcDomainDefPtr lxcDef = nsdata; + size_t i; + + if (!lxcDef) + return 0; + + virBufferAddLit(buf, "<lxc:namespace>\n"); + virBufferAdjustIndent(buf, 2); + + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) { + if (lxcDef->ns_source[i] == VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NONE) + continue; + + virBufferAsprintf(buf, "<lxc:%s type='%s' value='%s'/>\n", + virLXCDomainNamespaceTypeToString(i), + virLXCDomainNamespaceSourceTypeToString( + lxcDef->ns_source[i]), + lxcDef->ns_val[i]); + } + + virBufferAdjustIndent(buf, -2); + virBufferAddLit(buf, "</lxc:namespace>\n"); + return 0; +} + +static const char * +lxcDomainDefNamespaceHref(void) +{ + return "xmlns:lxc='" LXC_NAMESPACE_HREF "'"; +} + + +virDomainXMLNamespace virLXCDriverDomainXMLNamespace = { + .parse = lxcDomainDefNamespaceParse, + .free = lxcDomainDefNamespaceFree, + .format = lxcDomainDefNamespaceFormatXML, + .href = lxcDomainDefNamespaceHref, +}; + + static void virLXCDomainObjPrivateFree(void *data) { virLXCDomainObjPrivatePtr priv = data; diff --git a/src/lxc/lxc_domain.h b/src/lxc/lxc_domain.h index 751aece..2119c78 100644 --- a/src/lxc/lxc_domain.h +++ b/src/lxc/lxc_domain.h @@ -27,6 +27,31 @@ # include "lxc_conf.h" # include "lxc_monitor.h" +typedef enum { + VIR_LXC_DOMAIN_NAMESPACE_SHARENET = 0, + VIR_LXC_DOMAIN_NAMESPACE_SHAREIPC, + VIR_LXC_DOMAIN_NAMESPACE_SHAREUTS, + VIR_LXC_DOMAIN_NAMESPACE_LAST, +} virLXCDomainNamespace; + +typedef enum { + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NONE, + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NAME, + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_PID, + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NETNS, + + VIR_LXC_DOMAIN_NAMESPACE_SOURCE_LAST, +} virLXCDomainNamespaceSource; + +VIR_ENUM_DECL(virLXCDomainNamespace) +VIR_ENUM_DECL(virLXCDomainNamespaceSource) + +typedef struct _lxcDomainDef lxcDomainDef; +typedef lxcDomainDef *lxcDomainDefPtr; +struct _lxcDomainDef { + int ns_source[VIR_LXC_DOMAIN_NAMESPACE_LAST]; /* virLXCDomainNamespaceSource */ + char *ns_val[VIR_LXC_DOMAIN_NAMESPACE_LAST]; +}; typedef struct _virLXCDomainObjPrivate virLXCDomainObjPrivate; typedef virLXCDomainObjPrivate *virLXCDomainObjPrivatePtr; @@ -41,6 +66,7 @@ struct _virLXCDomainObjPrivate { virCgroupPtr cgroup; }; +extern virDomainXMLNamespace virLXCDriverDomainXMLNamespace; extern virDomainXMLPrivateDataCallbacks virLXCDriverPrivateDataCallbacks; extern virDomainDefParserConfig virLXCDriverDomainDefParserConfig; diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index e99b039..ade0ed7 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -359,6 +359,143 @@ char *virLXCProcessSetupInterfaceDirect(virConnectPtr conn, return ret; } +static const char *nsInfoLocal[VIR_LXC_DOMAIN_NAMESPACE_LAST] = { + [VIR_LXC_DOMAIN_NAMESPACE_SHARENET] = "net", + [VIR_LXC_DOMAIN_NAMESPACE_SHAREIPC] = "ipc", + [VIR_LXC_DOMAIN_NAMESPACE_SHAREUTS] = "uts", +}; + +static int virLXCProcessSetupNamespaceName(virConnectPtr conn, int ns_type, const char *name) +{ + virLXCDriverPtr driver = conn->privateData; + int fd = -1; + virDomainObjPtr vm; + virLXCDomainObjPrivatePtr priv; + char *path; + + vm = virDomainObjListFindByName(driver->domains, name); + if (!vm) { + virReportError(VIR_ERR_NO_DOMAIN, + _("No domain with matching name '%s'"), name); + return -1; + } + + priv = vm->privateData; + if (!priv->initpid) { + virReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("Init pid is not yet available")); + goto cleanup; + } + + if (virAsprintf(&path, "/proc/%lld/ns/%s", + (long long int)priv->initpid, + nsInfoLocal[ns_type]) < 0) + goto cleanup; + + if ((fd = open(path, O_RDONLY)) < 0) { + virReportSystemError(errno, + _("failed to open ns %s"), + virLXCDomainNamespaceTypeToString(ns_type)); + goto cleanup; + } + + cleanup: + VIR_FREE(path); + virObjectUnlock(vm); + virObjectUnref(vm); + return fd; +} + + +static int virLXCProcessSetupNamespacePID(int ns_type, const char *name) +{ + int fd; + char *path; + + if (virAsprintf(&path, "/proc/%s/ns/%s", + name, + nsInfoLocal[ns_type]) < 0) + return -1; + fd = open(path, O_RDONLY); + VIR_FREE(path); + if (fd < 0) { + virReportSystemError(errno, + _("failed to open ns %s"), + virLXCDomainNamespaceTypeToString(ns_type)); + return -1; + } + return fd; +} + + +static int virLXCProcessSetupNamespaceNet(int ns_type, const char *name) +{ + char *path; + int fd; + if (ns_type != VIR_LXC_DOMAIN_NAMESPACE_SHARENET) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("'netns' namespace source can only be " + "used with sharenet")); + return -1; + } + + if (virAsprintf(&path, "/var/run/netns/%s", name) < 0) + return -1; + fd = open(path, O_RDONLY); + VIR_FREE(path); + if (fd < 0) { + virReportSystemError(errno, + _("failed to open netns %s"), name); + return -1; + } + return fd; +} + + +/** + * virLXCProcessSetupNamespaces: + * @conn: pointer to connection + * @def: pointer to virtual machines namespaceData + * @nsFDs: out parameter to store the namespace FD + * + * Opens the specified namespace that needs to be shared and + * will moved into the container namespace later after clone has been called. + * + * Returns 0 on success or -1 in case of error + */ +static int virLXCProcessSetupNamespaces(virConnectPtr conn, + lxcDomainDefPtr lxcDef, + int *nsFDs) +{ + size_t i; + + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) + nsFDs[i] = -1; + /*If there are no namespace to be opened just return success*/ + if (lxcDef == NULL) + return 0; + + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) { + switch (lxcDef->ns_source[i]) { + case VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NONE: + continue; + case VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NAME: + if ((nsFDs[i] = virLXCProcessSetupNamespaceName(conn, i, lxcDef->ns_val[i])) < 0) + return -1; + break; + case VIR_LXC_DOMAIN_NAMESPACE_SOURCE_PID: + if ((nsFDs[i] = virLXCProcessSetupNamespacePID(i, lxcDef->ns_val[i])) < 0) + return -1; + break; + case VIR_LXC_DOMAIN_NAMESPACE_SOURCE_NETNS: + if ((nsFDs[i] = virLXCProcessSetupNamespaceNet(i, lxcDef->ns_val[i])) < 0) + return -1; + break; + } + } + + return 0; +} /** * virLXCProcessSetupInterfaces: @@ -764,6 +901,7 @@ virLXCProcessBuildControllerCmd(virLXCDriverPtr driver, char **veths, int *ttyFDs, size_t nttyFDs, + int *nsInheritFDs, int *files, size_t nfiles, int handshakefd, @@ -825,6 +963,19 @@ virLXCProcessBuildControllerCmd(virLXCDriverPtr driver, virCommandPassFD(cmd, files[i], 0); } + for (i = 0; i < VIR_LXC_DOMAIN_NAMESPACE_LAST; i++) { + if (nsInheritFDs[i] > 0) { + char *tmp = NULL; + if (virAsprintf(&tmp, "--share-%s", + nsInfoLocal[i]) < 0) + goto cleanup; + virCommandAddArg(cmd, tmp); + virCommandAddArgFormat(cmd, "%d", nsInheritFDs[i]); + virCommandPassFD(cmd, nsInheritFDs[i], 0); + VIR_FREE(tmp); + } + } + virCommandAddArgPair(cmd, "--security", virSecurityManagerGetModel(driver->securityManager)); @@ -1032,6 +1183,7 @@ int virLXCProcessStart(virConnectPtr conn, off_t pos = -1; char ebuf[1024]; char *timestamp; + int nsInheritFDs[VIR_LXC_DOMAIN_NAMESPACE_LAST]; virCommandPtr cmd = NULL; virLXCDomainObjPrivatePtr priv = vm->privateData; virCapsPtr caps = NULL; @@ -1204,6 +1356,10 @@ int virLXCProcessStart(virConnectPtr conn, if (virLXCProcessSetupInterfaces(conn, vm->def, &nveths, &veths) < 0) goto cleanup; + VIR_DEBUG("Setting up namespaces if any"); + if (virLXCProcessSetupNamespaces(conn, vm->def->namespaceData, nsInheritFDs) < 0) + goto cleanup; + VIR_DEBUG("Preparing to launch"); if ((logfd = open(logfile, O_WRONLY | O_APPEND | O_CREAT, S_IRUSR|S_IWUSR)) < 0) { @@ -1223,6 +1379,7 @@ int virLXCProcessStart(virConnectPtr conn, vm, nveths, veths, ttyFDs, nttyFDs, + nsInheritFDs, files, nfiles, handshakefds[1], &logfd, diff --git a/tests/lxcxml2xmldata/lxc-sharenet.xml b/tests/lxcxml2xmldata/lxc-sharenet.xml new file mode 100644 index 0000000..a2b8d1b --- /dev/null +++ b/tests/lxcxml2xmldata/lxc-sharenet.xml @@ -0,0 +1,33 @@ +<domain type='lxc' xmlns:lxc='http://libvirt.org/schemas/domain/lxc/1.0'> + <name>jessie</name> + <uuid>e21987a5-e98e-9c99-0e35-803e4d9ad1fe</uuid> + <memory unit='KiB'>1048576</memory> + <currentMemory unit='KiB'>1048576</currentMemory> + <vcpu placement='static'>1</vcpu> + <resource> + <partition>/machine</partition> + </resource> + <os> + <type arch='x86_64'>exe</type> + <init>/sbin/init</init> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>restart</on_crash> + <devices> + <emulator>/usr/libexec/libvirt_lxc</emulator> + <filesystem type='mount' accessmode='passthrough'> + <source dir='/mach/jessie'/> + <target dir='/'/> + </filesystem> + <console type='pty'> + <target type='lxc' port='0'/> + </console> + </devices> + <lxc:namespace> + <lxc:sharenet type='netns' value='red'/> + <lxc:shareipc type='pid' value='12345'/> + <lxc:shareuts type='name' value='container1'/> + </lxc:namespace> +</domain> diff --git a/tests/lxcxml2xmltest.c b/tests/lxcxml2xmltest.c index 3e00347..8d824b9 100644 --- a/tests/lxcxml2xmltest.c +++ b/tests/lxcxml2xmltest.c @@ -133,6 +133,7 @@ mymain(void) DO_TEST("filesystem-root"); DO_TEST("idmap"); DO_TEST("capabilities"); + DO_TEST("sharenet"); virObjectUnref(caps); virObjectUnref(xmlopt); -- 1.9.1

2 1

[libvirt] Libvirt Application Development Guide Using Python is now complete
by David Ashley 26 Aug '15

26 Aug '15

All - I have finished work on the "Libvirt Application Development Guide Using Python" book for the Fedora Docs Team. It is currently under review before being published. I have confirmed that I can offer the guide back the the libvirt team for their own use. The guide currently resides in a git repository and can be obtained with git clone git://git.fedorahosted.org/git/docs/libvirt_application_development_guide_u… Please let me know if you publish the guide as I would like to keep track of all the publishers so I can notify them about future updates. Thanks, W. David Ashley Fedora Docs Team

2 1