- Devel - libvirt List Archives

[libvirt] [PATCH] Move QEMU capabilities initialization later in QEMU startup
by Daniel P. Berrange 25 Jan '13

25 Jan '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Currently QEMU capabilities are initialized before the QEMU driver sets ownership on its various directories. The upshot is that if you change the user/group in the qemu.conf file, libvirtd will fail to probe QEMU the first time it is run after the config change. Moving QEMU capabilities initialization to after the chown() calls fixes this Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/qemu/qemu_driver.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 1f43d54..9ff920e 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -825,13 +825,6 @@ qemuStartup(bool privileged, if (qemuSecurityInit(qemu_driver) < 0) goto error; - qemu_driver->capsCache = qemuCapsCacheNew(qemu_driver->libDir, - qemu_driver->stateDir, - qemu_driver->user, - qemu_driver->group); - if (!qemu_driver->capsCache) - goto error; - if ((qemu_driver->activePciHostdevs = pciDeviceListNew()) == NULL) goto error; @@ -871,6 +864,12 @@ qemuStartup(bool privileged, } } + qemu_driver->capsCache = qemuCapsCacheNew(qemu_driver->libDir, + qemu_driver->user, + qemu_driver->group); + if (!qemu_driver->capsCache) + goto error; + if ((qemu_driver->caps = qemuCreateCapabilities(qemu_driver)) == NULL) goto error; -- 1.8.0.2

2 1

[libvirt] [PATCH v3 0/2] selinux: Only create the selabel_handle once.
by Richard W.M. Jones 24 Jan '13

24 Jan '13

Version 3 of this patch. Because I now have to pass the 'mgr' pointer around, the patch is considerably more complicated than before. Patch 1/2 is required so that I can use virReportSystemError when I don't need to have any optional arguments, ie. the equivalent of: printf ("foo\n"); I have tested these two patches against the libguestfs test suite, and it works for me and removes the memory leak. Rich.

3 6

[libvirt] not able to stop vmware player node through libvirt API
by varun bhatnagar 24 Jan '13

24 Jan '13

Hi, I have started one node through libvirt API using its start method. I am trying to stop that node using the *destroy *method provided by the API but I am not able to do so. Instead, I am getting one message saying: *libvir: error: internal error Child process (vmrun -T player stop /root/test_folder/myImage.vmx soft) unexpected exit status 255* Can anybody tell me why I am not able to stop the node. Thanks in advance. :)

1 0

[libvirt] [libvirt-java] Add various block, snapshot and migrate methods
by Wido den Hollander 24 Jan '13

24 Jan '13

Hi, I've sent a series of patches like these about two weeks ago and got some great feedback from Claudio on those! The feedback from Claudio has been used for writing this series of patches. Backwards compatibility has been preserved by still using virDomainMigrate and virtDomainMigrateToUri for the exisiting methods, just to be sure. Thank you, Wido

2 17

[libvirt] libvirt -- add virtio-scsi disk with scsi-generic
by 韩诚 24 Jan '13

24 Jan '13

Hi, Paolo, All, I read "virtio-scsi support proposal v2"(http://permalink.gmane.org/gmane.comp.emulators.libvirt/50428) Try to add a virtio-scsi with scsi-generic like this: <hostdev type='scsi'> <source> <adapter name='scsi_host0'/> <address type='scsi' bus='0' target='0' unit='0'/> </source> <target> <address type='scsi' controller='0' bus='0' target='0' unit='2'/> </target> </hostdev> But It turn out to be wrong, showing: error: Failed to define domain from rhel63ga error: XML error: unknown host device source address type 'scsi_host' I'd like to know how shall to add a virtio-scsi disk with scsi-generic option. Thanks~ -- Best regards, Cheng

3 4

[libvirt] This patch adds the label to lxc-enter-namespace
by Daniel J Walsh 24 Jan '13

24 Jan '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 (Resend to the correct list) lxc-enter-namespace allows a process from outside a container to start a process inside a container. One problem with the current code is the process running within the container would run with the label of the process that created it. For example if the admin process is running as unconfined_t and executes the following command # virsh -c lxc:/// lxc-enter-namespace --nolabel 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 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, 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 I also add a --nolabel command to virsh, which can go back to the original behaviour. virsh -c lxc:/// lxc-enter-namespace --nolabel 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 staff_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 37 ? 00:00:00 ps One problem I had when I originally did the patch is lxcDomainGetSecurityLabel was returning the incorrect label, I needed the label of the initpid within the container not its parent process, so I changed this function to match OpenNamespaces function. One last strangeness, about half the time I run this, virsh hangs and never returns. Seems like if (conn->driver->domainGetSecurityLabel(domain, seclabel) == 0) { Gets hung up. I have attached the strace in out1.gz -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.13 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlEBO2EACgkQrlYvE4MpobMS5ACg3Ih4Iu0lD9BofF4iP0QXarAL jpQAoLyWWNhnnFw2TRDJsXqvrTTVujyZ =hUZ/ -----END PGP SIGNATURE-----

2 1

[libvirt] [PATCH 0/2] Two additional build fixes
by Jiri Denemark 24 Jan '13

24 Jan '13

Both pushed as trivial and build-breakers. Jiri Denemark (2): apparmor: Avoid freeing uninitialized pointer selinux: Properly indent preprocessor directives src/security/security_apparmor.c | 2 +- src/security/security_selinux.c | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) -- 1.8.1.1

1 2

[libvirt] [PATCH 0/2] Fix build after commit 87b4c10c6cf02251dd8c29b5b895bebc6ec297f9
by Peter Krempa 24 Jan '13

24 Jan '13

My gcc didn't whine about the uninitialized variable. This series tries to fix and then actually fixes the bug. Sorry for the inconvenience. I'm putting on my brown bag and upgrading my GCC. Series pushed as a build-breaker. Peter Krempa (2): xen: Initialize variable before using xen: Actually fix the uninitialized variable src/xen/xend_internal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) -- 1.8.1.1

1 2

[libvirt] [libvirt-glib 1/2] Use g_strlcpy instead of strncpy
by Christophe Fergeau 24 Jan '13

24 Jan '13

This guarantees that the string will be nul-terminated. Coverity warned about this issue. --- libvirt-gobject/libvirt-gobject-connection.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt-gobject/libvirt-gobject-connection.c b/libvirt-gobject/libvirt-gobject-connection.c index 91cc535..0525323 100644 --- a/libvirt-gobject/libvirt-gobject-connection.c +++ b/libvirt-gobject/libvirt-gobject-connection.c @@ -1583,7 +1583,7 @@ GVirNodeInfo *gvir_connection_get_node_info(GVirConnection *conn, } ret = g_slice_new(GVirNodeInfo); - strncpy (ret->model, info.model, sizeof (ret->model)); + g_strlcpy (ret->model, info.model, sizeof (ret->model)); ret->memory = info.memory; ret->cpus = info.cpus; ret->mhz = info.mhz; -- 1.8.0.2

2 6

[libvirt] [PATCH 0/5] Resolve parallels and virCommandPtr resource leaks
by John Ferlan 24 Jan '13

24 Jan '13

This is v3 of the parallels_driver.c changes. The most recent review pointed out that virCommandPtr's need to be free'd. I found a few other instances where they weren't free'd. I also found a few instances where the status check from virCommandRun was not < 0 and cleaned them to be more consistent. John Ferlan (5): parallels: Resolve some resource leaks security: Need to add virCommandFree() storage: Need to add virCommandFree() util: Need to add virCommandFree() parallels_utils: Check return status properly from virCommandRun() src/parallels/parallels_driver.c | 47 ++++++++++++++++++++++++---------------- src/parallels/parallels_utils.c | 2 +- src/security/security_apparmor.c | 1 + src/storage/storage_backend_fs.c | 2 ++ src/util/virnetdevopenvswitch.c | 2 ++ 5 files changed, 34 insertions(+), 20 deletions(-) -- 1.7.11.7

2 8