- Devel - libvirt List Archives

Re: [libvirt] [CFT][PATCH 00/10] Making new mounts of proc and sysfs as safe as bind mounts (take 2)
by Richard Weinberger 16 Jun '15

16 Jun '15

[CC'ing libvirt-lxc folks] Am 28.05.2015 um 23:32 schrieb Eric W. Biederman: > Richard Weinberger <richard(a)nod.at> writes: > >> Am 28.05.2015 um 21:57 schrieb Eric W. Biederman: >>>> FWIW, it breaks also libvirt-lxc: >>>> Error: internal error: guest failed to start: Failed to re-mount /proc/sys on /proc/sys flags=1021: Operation not permitted >>> >>> Interesting. I had not anticipated a failure there? And it is failing >>> in remount? Oh that is interesting. >>> >>> That implies that there is some flag of the original mount of /proc that >>> the remount of /proc/sys is clearing, and that previously >>> >>> The flags specified are current rdonly,remount,bind so I expect there >>> are some other flags on proc that libvirt-lxc is clearing by accident >>> and we did not fail before because the kernel was not enforcing things. >> >> Please see: >> http://libvirt.org/git/?p=libvirt.git;a=blob;f=src/lxc/lxc_container.c;h=9a… >> lxcContainerMountBasicFS() >> >> and: >> http://libvirt.org/git/?p=libvirt.git;a=blob;f=src/lxc/lxc_container.c;h=9a… >> lxcBasicMounts >> >>> What are the mount flags in a working libvirt-lxc? >> >> See: >> test1:~ # cat /proc/self/mountinfo >> 149 147 0:56 / /proc rw,nosuid,nodev,noexec,relatime - proc proc rw >> 150 149 0:56 /sys /proc/sys ro,nodev,relatime - proc proc rw > >> If you need more info, please let me know. :-) > > Oh interesting I had not realized libvirt-lxc had grown an unprivileged > mode using user namespaces. > > This does appear to be a classic remount bug, where you are not > preserving the permissions. It appears the fact that the code > failed to enforce locked permissions on the fresh mount of proc > was hiding this bug until now. > > I expect what you actually want is the code below: > > diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c > index 9a9ae5c2aaf0..f008a7484bfe 100644 > --- a/src/lxc/lxc_container.c > +++ b/src/lxc/lxc_container.c > @@ -850,7 +850,7 @@ typedef struct { > > static const virLXCBasicMountInfo lxcBasicMounts[] = { > { "proc", "/proc", "proc", MS_NOSUID|MS_NOEXEC|MS_NODEV, false, false, false }, > - { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_RDONLY, false, false, false }, > + { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, false, false, false }, > { "/.oldroot/proc/sys/net/ipv4", "/proc/sys/net/ipv4", NULL, MS_BIND, false, false, true }, > { "/.oldroot/proc/sys/net/ipv6", "/proc/sys/net/ipv6", NULL, MS_BIND, false, false, true }, > { "sysfs", "/sys", "sysfs", MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, false, false, false }, > > Or possibly just: > > diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c > index 9a9ae5c2aaf0..a60ccbd12bfc 100644 > --- a/src/lxc/lxc_container.c > +++ b/src/lxc/lxc_container.c > @@ -850,7 +850,7 @@ typedef struct { > > static const virLXCBasicMountInfo lxcBasicMounts[] = { > { "proc", "/proc", "proc", MS_NOSUID|MS_NOEXEC|MS_NODEV, false, false, false }, > - { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_RDONLY, false, false, false }, > + { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_RDONLY, true, false, false }, > { "/.oldroot/proc/sys/net/ipv4", "/proc/sys/net/ipv4", NULL, MS_BIND, false, false, true }, > { "/.oldroot/proc/sys/net/ipv6", "/proc/sys/net/ipv6", NULL, MS_BIND, false, false, true }, > { "sysfs", "/sys", "sysfs", MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, false, false, false }, > > As the there is little point in making /proc/sys read-only in a > user-namespace, as the permission checks are uid based and no-one should > have the global uid 0 in your container. Making mounting /proc/sys > read-only rather pointless. Eric, using the patch below I was able to spawn a user-namespace enabled container using libvirt-lxc. :-) I had to: 1. Disable the read-only mount of /proc/sys which is anyway useless in the user-namespace case. 2. Disable the /proc/sys/net/ipv{4,6} bind mounts, this ugly hack is only needed for the non user-namespace case. 3. Remove MS_RDONLY from the sysfs mount (For the non user-namespace case we'd have to keep this, though). Daniel, I'd take this as a chance to disable all the MS_RDONLY games if user-namespace are configured. With Eric's fixes they hurt us. And as I wrote many times before if root within the user-namespace is able to do nasty things in /sys and /proc that's a plain kernel bug which needs fixing. There is no point in mounting these read-only. Except for the case then no user-namespace is used. diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 9a9ae5c..497e05f 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -850,10 +850,10 @@ typedef struct { static const virLXCBasicMountInfo lxcBasicMounts[] = { { "proc", "/proc", "proc", MS_NOSUID|MS_NOEXEC|MS_NODEV, false, false, false }, - { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_RDONLY, false, false, false }, - { "/.oldroot/proc/sys/net/ipv4", "/proc/sys/net/ipv4", NULL, MS_BIND, false, false, true }, - { "/.oldroot/proc/sys/net/ipv6", "/proc/sys/net/ipv6", NULL, MS_BIND, false, false, true }, - { "sysfs", "/sys", "sysfs", MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, false, false, false }, + { "/proc/sys", "/proc/sys", NULL, MS_BIND|MS_RDONLY, true, false, false }, + { "/.oldroot/proc/sys/net/ipv4", "/proc/sys/net/ipv4", NULL, MS_BIND, true, false, true }, + { "/.oldroot/proc/sys/net/ipv6", "/proc/sys/net/ipv6", NULL, MS_BIND, true, false, true }, + { "sysfs", "/sys", "sysfs", MS_NOSUID|MS_NOEXEC|MS_NODEV, false, false, false }, { "securityfs", "/sys/kernel/security", "securityfs", MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, true, true, false }, #if WITH_SELINUX { SELINUX_MOUNT, SELINUX_MOUNT, "selinuxfs", MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_RDONLY, true, true, false }, Thanks, //richard

3 4

[libvirt] [libvirt-python PATCH] Change livbirt version to 1.3.0 for the next release
by Martin Kletzander 16 Jun '15

16 Jun '15

Since the background for Admin API is merged upstream, we are bumping the minor release version as discussed previously Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Pushed as trivial setup.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 8d34fde3e60d..04e3de41fc71 100755 --- a/setup.py +++ b/setup.py @@ -311,7 +311,7 @@ class my_clean(clean): _c_modules, _py_modules = get_module_lists() setup(name = 'libvirt-python', - version = '1.2.17', + version = '1.3.0', url = 'http://www.libvirt.org', maintainer = 'Libvirt Maintainers', maintainer_email = 'libvir-list(a)redhat.com', -- 2.4.3

1 0

[libvirt] [PATCH] Change livbirt version to 1.3.0 for the next release
by Martin Kletzander 16 Jun '15

16 Jun '15

Since the background for Admin API is merged upstream, we are bumping the minor release version as discussed previously Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Pushed as trivial configure.ac | 2 +- docs/formatdomain.html.in | 2 +- src/parallels/parallels_driver.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/configure.ac b/configure.ac index dbaf2722f8f3..20fc564a8128 100644 --- a/configure.ac +++ b/configure.ac @@ -16,7 +16,7 @@ dnl You should have received a copy of the GNU Lesser General Public dnl License along with this library. If not, see dnl <http://www.gnu.org/licenses/>. -AC_INIT([libvirt], [1.2.17], [libvir-list(a)redhat.com] [], [http://libvirt.org] +AC_INIT([libvirt], [1.3.0], [libvir-list(a)redhat.com] [], [http://libvirt.org] AC_CONFIG_SRCDIR([src/libvirt.c]) AC_CONFIG_AUX_DIR([build-aux]) AC_CONFIG_HEADERS([config.h]) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 3ccc71c55918..4e85b51040cb 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -4275,7 +4275,7 @@ qemu-kvm -net nic,model=? /dev/null element. Each queue will potentially be handled by a different processor, resulting in much higher throughput. virtio-net since 1.0.6 (QEMU and KVM only) - vhost-user since 1.2.17 (QEMU and KVM only) + vhost-user since 1.3.0 (QEMU and KVM only) </dd> <dt><code>host</code> offloading options</dt> <dd> diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c index 706229d8486c..b0a19f32c820 100644 --- a/src/parallels/parallels_driver.c +++ b/src/parallels/parallels_driver.c @@ -1371,8 +1371,8 @@ static virHypervisorDriver vzDriver = { .domainManagedSave = parallelsDomainManagedSave, /* 1.2.14 */ .domainManagedSaveRemove = parallelsDomainManagedSaveRemove, /* 1.2.14 */ .domainGetMaxMemory = parallelsDomainGetMaxMemory, /* 1.2.15 */ - .domainBlockStats = parallelsDomainBlockStats, /* 1.2.17 */ - .domainBlockStatsFlags = parallelsDomainBlockStatsFlags, /* 1.2.17 */ + .domainBlockStats = parallelsDomainBlockStats, /* 1.3.0 */ + .domainBlockStatsFlags = parallelsDomainBlockStatsFlags, /* 1.3.0 */ }; static virConnectDriver vzConnectDriver = { -- 2.4.3

1 0

[libvirt] [PATCH 0/3] Couple of storage_backend_scsi.c fixes
by Michal Privoznik 16 Jun '15

16 Jun '15

As promised here [1] I'm proposing my own patch to fix the problem referenced here [2]. 1: https://www.redhat.com/archives/libvir-list/2015-June/msg00621.html 2: https://www.redhat.com/archives/libvir-list/2015-June/msg00567.html John Ferlan (1): scsi: Adjust return status from getBlockDevice Michal Privoznik (2): getNewStyleBlockDevice: Adjust formatting getOldStyleBlockDevice: Adjust formatting src/storage/storage_backend_scsi.c | 91 ++++++++++++++++++++++---------------- 1 file changed, 53 insertions(+), 38 deletions(-) -- 2.3.6

2 5

[libvirt] [PATCH 00/13] Admin API
by Martin Kletzander 16 Jun '15

16 Jun '15

This is a real post for an Admin API to get in. Everything discussed in previous posts should be a addressed. I'll start a discussion later on today about the ideas for the APIs to be added and the general near-future desing moves. I'm also keeping the last three example patches in so you can try that it works. Martin Kletzander (13): Move daemon-related parts of virNetServer to virNetDaemon tests: Adapt virnetservertest to daemon refactor Teach gendispatch how to handle admin dispatching files Add admin protocol Build client headers for admin protocol Add admin error domain Add libvirt-admin library Add XML files with admin API specification Add configuration options for permissions on daemon's admin socket Add support for admin API in libvirt daemon rpc: Add virNetServerGetNClients admin: Add virAdmHello function Example virt-admin .gitignore | 5 + Makefile.am | 7 +- cfg.mk | 9 +- configure.ac | 12 +- daemon/Makefile.am | 32 +- daemon/admin_server.c | 140 ++++ daemon/admin_server.h | 36 + daemon/libvirtd-config.c | 27 +- daemon/libvirtd-config.h | 13 +- daemon/libvirtd.aug | 13 + daemon/libvirtd.c | 205 ++++-- daemon/libvirtd.conf | 24 + daemon/libvirtd.h | 14 +- daemon/test_libvirtd.aug.in | 9 + docs/Makefile.am | 23 +- docs/apibuild.py | 10 +- docs/internals.html.in | 4 +- docs/internals/rpc.html.in | 7 + include/libvirt/Makefile.am | 4 +- include/libvirt/libvirt-admin.h | 63 ++ include/libvirt/virterror.h | 3 +- libvirt-admin.pc.in | 13 + libvirt.spec.in | 21 + po/POTFILES.in | 4 + src/Makefile.am | 75 ++- src/admin/admin_protocol.x | 91 +++ src/admin_protocol-structs | 15 + src/datatypes.c | 30 + src/datatypes.h | 37 + src/internal.h | 1 + src/libvirt-admin.c | 412 ++++++++++++ src/libvirt_admin.syms | 20 + src/libvirt_remote.syms | 31 +- src/locking/lock_daemon.c | 75 ++- src/locking/lock_daemon_config.c | 2 +- src/locking/lock_daemon_dispatch.c | 4 +- src/lxc/lxc_controller.c | 65 +- src/rpc/gendispatch.pl | 254 +++---- src/rpc/virnetdaemon.c | 749 +++++++++++++++++++++ src/rpc/virnetdaemon.h | 82 +++ src/rpc/virnetserver.c | 530 ++------------- src/rpc/virnetserver.h | 46 +- src/rpc/virnetserverprogram.h | 3 + src/util/virerror.c | 1 + tests/Makefile.am | 10 +- tests/confdata/libvirtd.conf | 6 + tests/confdata/libvirtd.out | 5 + .../{virnetserverdata => virnetdaemondata}/README | 0 .../virnetdaemondata/input-data-admin-nomdns.json | 126 ++++ .../input-data-anon-clients.json | 0 .../input-data-initial-nomdns.json} | 1 - .../input-data-initial.json | 0 .../virnetdaemondata/output-data-admin-nomdns.json | 128 ++++ .../virnetdaemondata/output-data-anon-clients.json | 66 ++ .../output-data-initial-nomdns.json | 66 ++ tests/virnetdaemondata/output-data-initial.json | 67 ++ tests/{virnetservertest.c => virnetdaemontest.c} | 70 +- tests/virnetserverdata/output-data-initial.json | 63 -- tools/virt-admin/Makefile.am | 69 ++ tools/virt-admin/virt-admin.c | 72 ++ tools/virt-admin/virt-admin.pod | 43 ++ 61 files changed, 3180 insertions(+), 833 deletions(-) create mode 100644 daemon/admin_server.c create mode 100644 daemon/admin_server.h create mode 100644 include/libvirt/libvirt-admin.h create mode 100644 libvirt-admin.pc.in create mode 100644 src/admin/admin_protocol.x create mode 100644 src/admin_protocol-structs create mode 100644 src/libvirt-admin.c create mode 100644 src/libvirt_admin.syms create mode 100644 src/rpc/virnetdaemon.c create mode 100644 src/rpc/virnetdaemon.h rename tests/{virnetserverdata => virnetdaemondata}/README (100%) create mode 100644 tests/virnetdaemondata/input-data-admin-nomdns.json rename tests/{virnetserverdata => virnetdaemondata}/input-data-anon-clients.json (100%) rename tests/{virnetserverdata/output-data-anon-clients.json => virnetdaemondata/input-data-initial-nomdns.json} (97%) rename tests/{virnetserverdata => virnetdaemondata}/input-data-initial.json (100%) create mode 100644 tests/virnetdaemondata/output-data-admin-nomdns.json create mode 100644 tests/virnetdaemondata/output-data-anon-clients.json create mode 100644 tests/virnetdaemondata/output-data-initial-nomdns.json create mode 100644 tests/virnetdaemondata/output-data-initial.json rename tests/{virnetservertest.c => virnetdaemontest.c} (84%) delete mode 100644 tests/virnetserverdata/output-data-initial.json create mode 100644 tools/virt-admin/Makefile.am create mode 100644 tools/virt-admin/virt-admin.c create mode 100644 tools/virt-admin/virt-admin.pod -- 2.4.3

2 19

[libvirt] [PATCH] qemu: add value range check of option granularity
by Shanzhi Yu 16 Jun '15

16 Jun '15

The default value of the granularity is the image cluster size clamped between 4096 and 65536. Libvirt should add a check for this otherwise qemu will report error like below: $ virsh blockcopy r7 vda /var/lib/libvirt/images/r7.s1 --granularity 32 error: internal error: unable to execute QEMU command 'drive-mirror': Parameter 'granularity' expects a value in range [512B, 64MB] Signed-off-by: Shanzhi Yu <shyu(a)redhat.com> --- src/qemu/qemu_driver.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index d1f195c..0eca8ed 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -16990,6 +16990,10 @@ qemuDomainBlockCopy(virDomainPtr dom, const char *disk, const char *destxml, virReportError(VIR_ERR_INVALID_ARG, "%s", _("granularity must be power of 2")); goto cleanup; + } else if (param->value.ui > 65536 || param->value.ui < 4096) { + virReportError(VIR_ERR_INVALID_ARG, "%s", + _("granularity expects value in range [4096 65536]")); + goto cleanup; } granularity = param->value.ui; } else if (STREQ(param->field, VIR_DOMAIN_BLOCK_COPY_BUF_SIZE)) { -- 2.4.3

2 1

[libvirt] [sandbox v2 0/4] Getting qemu sandboxes run as root + host-image format
by Cédric Bosdonnat 16 Jun '15

16 Jun '15

Hi all, Here are a few patches to make sandboxes run with qemu:///system connection. The last one is just a new feature to allow using somethings else than RAW images in host-image mounts. This feature will later be needed to run docker container using Eren's work. Diff to v1: * Adapted the code to set the filesystem type due to the changes in libvirt-glib patch (v2). Cédric Bosdonnat (4): Make sure the sandbox state dir and config can be accessed Write /dev/vd* instead of vd* in mounts.cfg qemu: mount all host-images as ext4 Add host-image format parameter .../libvirt-sandbox-builder-container.c | 10 +++ libvirt-sandbox/libvirt-sandbox-builder-machine.c | 13 +++- .../libvirt-sandbox-config-mount-host-image.c | 91 +++++++++++++++++++++- .../libvirt-sandbox-config-mount-host-image.h | 5 +- libvirt-sandbox/libvirt-sandbox-config.c | 70 +++++++++++++++-- .../libvirt-sandbox-context-interactive.c | 4 +- libvirt-sandbox/libvirt-sandbox.sym | 5 ++ libvirt-sandbox/tests/test-config.c | 1 + 8 files changed, 186 insertions(+), 13 deletions(-) -- 2.1.4

2 8

[libvirt] [PATCH v2 0/2] lxc: properly clean up qemu-nbd
by Cédric Bosdonnat 16 Jun '15

16 Jun '15

Hi all, Here is the very same patch, but split in two patches. Well, I also moved two comments around between v1 and v2. Cédric Bosdonnat (2): Add virProcessGetPids to get all tasks of a process lxc: properly clean up qemu-nbd src/libvirt_private.syms | 1 + src/lxc/lxc_controller.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++ src/util/virprocess.c | 47 ++++++++++++++++++++++++++++++++++++++++ src/util/virprocess.h | 2 ++ 4 files changed, 106 insertions(+) -- 2.1.4

3 3

[libvirt] Publishing a Python libvirt Howto/Reference Guide
by David Ashley 16 Jun '15

16 Jun '15

All - I represent the Fedora Docs Team. We are interested in publishing a Python libvirt HowTo/Reference Guide for users. We feel that there is a gap with using virtualization from a scripting environment. While virsh fills this gap for simple scripting purposes, a more powerful scripting language is needed for deep administrative purposes. Python is the obvious choice and the libvirt team has provided a great Python module. What is really missing are examples of using the module in both simple and deeper tasks i.e. something more than just how to start and stop a domain. We propose to create such a document. While we can use pydoc to create the reference material, we need some really good examples that can teach a user how to get real work done using the classes. We also propose to share the document with the libvirt team taking either an upstream or downstream role in maintaining the document. The document would be written in DocBook and published using the Publican system. This would allow you to create your own brand for the document (look and feel) while maintaining compatible source for both our teams. It would also allow the source to be maintained by both teams. If this sounds interesting to you we would like to begin our planning for the document in the near future with one or more representatives from the libvirt team on board. Please let either myself or Laura Novich (cc'ed on this email) know if this proposal meets with your approval. W. David Ashley w.david.ashley(a)gmail.com Fedora Docs Team

3 2

[libvirt] [PATCH v2 0/3] Adjust netfs CIFS/Samba formatting
by John Ferlan 16 Jun '15

16 Jun '15

https://bugzilla.redhat.com/show_bug.cgi?id=1186969 v1 here: http://www.redhat.com/archives/libvir-list/2015-June/msg00155.html Changes since v1: * Alter v1 to make a common RNG - even though it's ACK'd - I figured I'd still send to make sure there were no other issues... * Add 2/3 to apply comment made in former 2/2 regarding checking for pool->def->type == VIR_STORAGE_POOL_FS to the glusterfs since it has the same issue * Modify 3/3 is old 2/2 and only removed the ternary check. John Ferlan (3): storage: Fix the schema and add tests for cifs pool storage: Adjust command arglist for gluster storage: Generate correct parameters for CIFS docs/formatstorage.html.in | 7 ++++-- docs/schemas/storagepool.rng | 10 ++++---- docs/storage.html.in | 3 ++- src/storage/storage_backend_fs.c | 31 ++++++++++++++++++------- tests/storagepoolxml2xmlin/pool-netfs-cifs.xml | 12 ++++++++++ tests/storagepoolxml2xmlout/pool-netfs-cifs.xml | 15 ++++++++++++ tests/storagepoolxml2xmltest.c | 1 + 7 files changed, 64 insertions(+), 15 deletions(-) create mode 100644 tests/storagepoolxml2xmlin/pool-netfs-cifs.xml create mode 100644 tests/storagepoolxml2xmlout/pool-netfs-cifs.xml -- 2.1.0

2 6