- Devel - libvirt List Archives

[libvirt] [PATCH v3 0/5] Per-guest configurable user/group for QEMU processes
by Marcelo Cerri 16 Aug '12

16 Aug '12

This is a v3 patch series that updates the libvirt's security driver mechanism to support per-guest configurable user and group for QEMU processes running together with other security drivers, such as SELinux and AppArmor. Comments and feedbacks are welcome. Marcelo Cerri (5): Internal refactory of data structures Multiple security drivers in XML data Update security layer to handle many security labels Support for multiple default security drivers in QEMU config Update the remote API daemon/remote.c | 63 ++++ docs/formatdomain.html.in | 11 +- docs/schemas/capability.rng | 18 +- docs/schemas/domaincommon.rng | 30 ++- include/libvirt/libvirt.h.in | 2 + python/generator.py | 1 + src/conf/capabilities.c | 17 +- src/conf/capabilities.h | 6 +- src/conf/domain_audit.c | 14 +- src/conf/domain_conf.c | 342 +++++++++++++++----- src/conf/domain_conf.h | 18 +- src/driver.h | 4 + src/libvirt.c | 47 +++ src/libvirt_private.syms | 5 + src/libvirt_public.syms | 5 + src/lxc/lxc_conf.c | 8 +- src/lxc/lxc_controller.c | 8 +- src/lxc/lxc_driver.c | 11 +- src/lxc/lxc_process.c | 23 +- src/qemu/qemu_conf.c | 38 ++- src/qemu/qemu_conf.h | 2 +- src/qemu/qemu_driver.c | 218 +++++++++++--- src/qemu/qemu_process.c | 50 ++- src/remote/remote_driver.c | 46 +++ src/remote/remote_protocol.x | 17 +- src/remote_protocol-structs | 1 + src/security/security_apparmor.c | 118 +++++-- src/security/security_dac.c | 324 +++++++++++++++++-- src/security/security_manager.c | 101 +++++-- src/security/security_manager.h | 8 +- src/security/security_selinux.c | 256 ++++++++++----- src/security/security_stack.c | 237 +++++++++----- src/security/security_stack.h | 13 + src/test/test_driver.c | 11 +- .../qemuxml2argv-seclabel-dynamic-override.xml | 4 +- .../qemuxml2argv-seclabel-dynamic.xml | 2 +- 36 files changed, 1634 insertions(+), 445 deletions(-)

4 22

[libvirt] RFC: a <uuid> for every <interface> in each domain
by Laine Stump 15 Aug '12

15 Aug '12

A couple of situations have come up recently that could be solved by every interface in every domain always having a unique identifier associated with it: 1) In order to properly track information as a guest is disconnected and reconnected to an Open vSwitch bridge, the initial Open vSwitch support added an "interfaceid" parameter to virtualport. If the interfaceid isn't specified by the user when the interface is defined, one is automatically generated and placed into the persistent definition. But if the guest interface is just <interface type='network'>, it isn't known at the time of definition whether this interface will be using an Open vSwitch connection, or something else. Then, by the time we get into the network driver and decide that we're going to use Open vSwitch, it's too late to conveniently/cleanly generate an interfaceid and plug it back into the <virtualport> of the interface's persistent config. The result is that each time the guest is restarted, Open vSwitch gets a new interfaceid for it, and can't properly track things. If every interface had a uuid, the code that connects to an Open vSwitch network could just use the interface's uuid if interfaceid wasn't specified. 2) I'm about to write some patches that will allow enacting config changes to a network without needing to destroy/re-start the network. One thing that came up while I was mulling over what could and couldn't be modified without a restart is that it will be possible to allow removing physical interfaces from a network's interface pool (used for macvtap and hostdev modes), but only if that physical interface isn't currently in use. There is a use counter on each interface so we can easily tell when that situation occurs, but once we know of the failure, we have no way of pointing to which guest is causing the problem. If each interface had a uuid, we could save a list of the uuid's of all interfaces currently connected to a particular physical interface, and report that in the failure message. It would then be a fairly mechanical task to find that uuid in the guests' config. Does this sound like a reasonable idea? Any reasons *not* to do it? Problems we'll need to take care of if we add it (for example, existing guest interfaces will all need to get a uuid during the upgrade process, similar to the way we add a mac address to all existing networks that don't have one). Any other things you can think of doing with uuid if we add one?

3 5

[libvirt] [PATCHv2 0/4] support <vlan> element for interfaces and networks
by Laine Stump 15 Aug '12

15 Aug '12

danpb raised a (valid) objection about the XML format in the first version of these patches, so I've redone it to be like this: <vlan trunk='yes'> <tag id='42'/> <tag id='30'/> </vlan> The downside is that for a single vlan tag, it's more verbose: <vlan> <tag id='42'/> </vlan> On the other hand, it's more consistent. The rest of the patches in the series were unchanged (and even the data definition is unchanged). === The purpose of these patches is to define a data type that can properly describe an interface's vlan configuration (including multiple vlans for trunking), provide XML parser and formatter to translate to and from the new data type, add the new <vlan> element to both domain <interface> and to <network> in appropriate places, and to provide the new vlan data to the hypervisor driver. Once these 4 patches are applied, all that will be left to support vlan configuration for any network connection that supports it, is to request the vlan data for the interface with virDomainNetGetActualVlan(netdef). If that function returns NULL, there is no vlan config for this interface; if it's non-zero, it is a pointer to virNetDevVlan, and you can grab the data from there. Since I currently know of only two types of network connection we will be able to easily support this for (sr-iov VFs using PCI passthrough, and openvswitch), any type other than those will refuse to start if a vlan has been requested (see PATCH 4/4). This covers validation for the qemu and lxc drivers (since they use common code), but I'm not sure about xen, esx, or any other driver - they will likely require extra code to either implement transparent vlan configuration, or log an error if it's requested (my guess is that, in spite of our best efforts, *many* people will misunderstand the limitations of this functionality, and try to put all sorts of interfaces on vlans; it will be much friendlier to inform them of the problem rather than simply ignore the vlan config.

4 14

[libvirt] [PATCHv2 0/9] network: properly support openvswitch in <network>
by Laine Stump 15 Aug '12

15 Aug '12

This is mainly a refresh of the initial version of the patch to rebase, but there have also been a couple of small bugs fixed. In particular, if not <virtualport/> is specified in the domain's interface element, connection of the interface to an ovs switch won't fail because of a missing interfaceid (see http://www.redhat.com/archives/libvir-list/2012-August/msg00874.html and it's parents) Many of these patches are trivial or mechanical. ====== Although it's been possible (ever since openvswitch was added to libvirt in 0.9.11) for a libvirt network to use an openvswitch bridge (by adding <virtualport type='openvswitch'>), the virtualport in the network would always have a default random interfaceid included, which would be re-used for all interfaces using that network, which doesn't really work at all. The alternative was to not specify openvswitch in the <network> definition, but to do it in the guest's <interface> definition instead - this of course goes against the principle of not having host-specific config embedded in guest config. This patch series enhances the functionality of <virtualport> elements, to allow omitting some attributes (and even the type), and to merge the interface, network, and portgroup virtualports rather than simply picking one. This not only makes openvswitch <network>s more practical (because the network can specify type='openvswitch' without also specifying an interfaceid), but also makes <virtualport> in networks and portgroups more useful in general - for example, an interface can specify an interfaceid (used only by openvswitch) *and* an instanceid (used only by 802.1Qbh), while the network's virtualport specifies only the type, and the portgroups specify the managerid, typeid, profileid, or whatever is appropriate for the type of switch used by the network. The result is that the guest config can be completely devoid of knowledge about the type of switch being used on the hardware, but can still enjoy full configurability for whatever switch ends up being used.

6 26

[libvirt] [PATCH] build: avoid warnings on older gcc
by Eric Blake 15 Aug '12

15 Aug '12

The previous patch pulled in a newer version of stat-time.h from gnulib, which tries to suppress some stupid gcc warnings (see <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54113>) by exploiting a feature of newer gcc. But we still aim to compile on RHEL 6.3, with gcc 4.4.6 (not to mention even older devel platforms like RHEL 5), which fails to build thanks to our development -Werror. It took me a while to figure out how to make our set of warnings smaller on older gcc without losing the benefit of the warnings when using newer gcc (such as the one on Fedora 17), but this should do the trick. * m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Avoid warnings that gnulib can't silence on older gcc. --- If I don't get a review by tomorrow, I will go ahead and push this under the build-breaker rule. m4/virt-compile-warnings.m4 | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/m4/virt-compile-warnings.m4 b/m4/virt-compile-warnings.m4 index 9dee000..702506d 100644 --- a/m4/virt-compile-warnings.m4 +++ b/m4/virt-compile-warnings.m4 @@ -58,6 +58,23 @@ AC_DEFUN([LIBVIRT_COMPILE_WARNINGS],[ # Gnulib's stat-time.h violates this dontwarn="$dontwarn -Waggregate-return" + # Gnulib uses '#pragma GCC diagnostic push' to silence some + # warnings, but older gcc doesn't support this. + AC_CACHE_VAL([lv_cv_gcc_pragma_push_works], [ + save_CFLAGS=$CFLAGS + CFLAGS=-Wunknown-pragmas + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ + #pragma GCC diagnostic push + #pragma GCC diagnostic pop + ]])], + [lv_cv_gcc_pragma_push_works=yes], + [lv_cv_gcc_pragma_push_works=no]) + CFLAGS=$save_CFLAGS]) + if test $lv_cv_gcc_pragma_push_works = no; then + dontwarn="$dontwarn -Wmissing-prototypes" + dontwarn="$dontwarn -Wmissing-declarations" + fi + # We might fundamentally need some of these disabled forever, but # ideally we'd turn many of them on dontwarn="$dontwarn -Wfloat-equal" -- 1.7.11.2

4 5

[libvirt] [firewalld PATCHv3] firewalld PATCH v3
by Thomas Woerner 15 Aug '12

15 Aug '12

* configura.ac, spec file: firewalld now defaults to enabled, depends on dbus * fixed comment for with_firewalld define * bridge_driver, nwfilter_driver: new dbus filters to get FirewallD1.Reloaded signal and DBus.NameOwnerChanged on org.fedoraproject.FirewallD1 * iptables, ebtables, nwfilter_ebiptables_driver: use firewall-cmd direct passthrough interface * spec file changed as requested --- configure.ac | 17 ++++++++++++++++ libvirt.spec.in | 11 +++++++++++ src/Makefile.am | 4 ++-- src/network/bridge_driver.c | 47 +++++++++++++++++++++++++++++++++++++++++++++ src/util/ebtables.c | 35 +++++++++++++++++++++++++++++++++ src/util/iptables.c | 21 ++++++++++++++++++-- 6 files changed, 131 insertions(+), 4 deletions(-) diff --git a/configure.ac b/configure.ac index 8a04d91..c742d2f 100644 --- a/configure.ac +++ b/configure.ac @@ -1282,6 +1282,22 @@ AM_CONDITIONAL([HAVE_POLKIT1], [test "x$with_polkit1" = "xyes"]) AC_SUBST([POLKIT_CFLAGS]) AC_SUBST([POLKIT_LIBS]) +dnl firewalld +AC_ARG_WITH([firewalld], + AC_HELP_STRING([--with-firewalld], [enable firewalld support @<:@default=check@:>@]), + [], + [with_firewalld=check]) +if test "x$with_firewalld" = "xcheck" ; then + with_firewalld=$with_dbus +fi +if test "x$with_firewalld" == "xyes" ; then + if test "x$with_dbus" != "xyes" ; then + AC_MSG_ERROR([You must have dbus enabled for firewalld support]) + fi + AC_DEFINE_UNQUOTED([HAVE_FIREWALLD], [1], [whether firewalld support is enabled]) +fi +AM_CONDITIONAL([HAVE_FIREWALLD], [test "x$with_firewalld" != "xno"]) + dnl Avahi library AC_ARG_WITH([avahi], AC_HELP_STRING([--with-avahi], [use avahi to advertise remote daemon @<:@default=check@:>@]), @@ -2989,6 +3005,7 @@ AC_MSG_NOTICE([ sanlock: $SANLOCK_CFLAGS $SANLOCK_LIBS]) else AC_MSG_NOTICE([ sanlock: no]) fi +AC_MSG_NOTICE([firewalld: $with_firewalld]) if test "$with_avahi" = "yes" ; then AC_MSG_NOTICE([ avahi: $AVAHI_CFLAGS $AVAHI_LIBS]) else diff --git a/libvirt.spec.in b/libvirt.spec.in index 67b955a..ea2fd88 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -106,6 +106,7 @@ %define with_sanlock 0%{!?_without_sanlock:0} %define with_systemd 0%{!?_without_systemd:0} %define with_numad 0%{!?_without_numad:0} +%define with_firewalld 0%{!?_without_firewalld:0} # Non-server/HV driver defaults which are always enabled %define with_python 0%{!?_without_python:1} @@ -146,6 +147,11 @@ %define with_systemd 1 %endif +# Fedora 18 / RHEL-7 are first where firewalld support is enabled +%if 0%{?fedora} >= 17 || 0%{?rhel} >= 7 +%define with_firewalld 1 +%endif + # RHEL-5 has restricted QEMU to x86_64 only and is too old for LXC %if 0%{?rhel} == 5 %define with_qemu_tcg 0 @@ -1182,6 +1188,10 @@ of recent versions of Linux (and other OSes). %define _without_driver_modules --without-driver-modules %endif +%if %{with_firewalld} +%define _with_firewalld --with-firewalld +%endif + %define when %(date +"%%F-%%T") %define where %(hostname) %define who %{?packager}%{!?packager:Unknown} @@ -1240,6 +1250,7 @@ autoreconf -if %{?_without_audit} \ %{?_without_dtrace} \ %{?_without_driver_modules} \ + %{?_with_firewalld} \ %{with_packager} \ %{with_packager_version} \ --with-qemu-user=%{qemu_user} \ diff --git a/src/Makefile.am b/src/Makefile.am index 79b4e59..76570db 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -989,7 +989,7 @@ libvirt_driver_network_la_SOURCES = libvirt_driver_network_la_LIBADD = libvirt_driver_network_impl.la if WITH_DRIVER_MODULES mod_LTLIBRARIES += libvirt_driver_network.la -libvirt_driver_network_la_LIBADD += ../gnulib/lib/libgnu.la $(LIBNL_LIBS) +libvirt_driver_network_la_LIBADD += ../gnulib/lib/libgnu.la $(LIBNL_LIBS) $(DBUS_LIBS) libvirt_driver_network_la_LDFLAGS = -module -avoid-version $(AM_LDFLAGS) else noinst_LTLIBRARIES += libvirt_driver_network.la @@ -999,7 +999,7 @@ endif libvirt_driver_network_impl_la_CFLAGS = \ $(LIBNL_CFLAGS) \ - -I$(top_srcdir)/src/conf $(AM_CFLAGS) + -I$(top_srcdir)/src/conf $(AM_CFLAGS) $(DBUS_CFLAGS) libvirt_driver_network_impl_la_SOURCES = $(NETWORK_DRIVER_SOURCES) endif EXTRA_DIST += network/default.xml diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index a5046f1..39c0449 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -61,6 +61,7 @@ #include "virnetdev.h" #include "virnetdevbridge.h" #include "virnetdevtap.h" +#include "virdbus.h" #define NETWORK_PID_DIR LOCALSTATEDIR "/run/libvirt/network" #define NETWORK_STATE_DIR LOCALSTATEDIR "/lib/libvirt/network" @@ -248,6 +249,25 @@ networkAutostartConfigs(struct network_driver *driver) { } } +#if HAVE_FIREWALLD +static DBusHandlerResult +firewalld_dbus_filter_bridge(DBusConnection *connection ATTRIBUTE_UNUSED, + DBusMessage *message, void *user_data) { + struct network_driver *_driverState = user_data; + + if (dbus_message_is_signal(message, DBUS_INTERFACE_DBUS, + "NameOwnerChanged") || + dbus_message_is_signal(message, "org.fedoraproject.FirewallD1", + "Reloaded")) + { + VIR_DEBUG("Reload in bridge_driver because of firewalld."); + networkReloadIptablesRules(_driverState); + } + + return DBUS_HANDLER_RESULT_NOT_YET_HANDLED; +} +#endif + /** * networkStartup: * @@ -256,6 +276,9 @@ networkAutostartConfigs(struct network_driver *driver) { static int networkStartup(int privileged) { char *base = NULL; +#ifdef HAVE_FIREWALLD + DBusConnection *sysbus = NULL; +#endif if (VIR_ALLOC(driverState) < 0) goto error; @@ -322,6 +345,30 @@ networkStartup(int privileged) { networkDriverUnlock(driverState); +#ifdef HAVE_FIREWALLD + if (!(sysbus = virDBusGetSystemBus())) { + virErrorPtr err = virGetLastError(); + VIR_WARN("DBus not available, disabling firewalld support in bridge_driver: %s", err->message); + } else { + /* add matches for + * NameOwnerChanged on org.freedesktop.DBus for firewalld start/stop + * Reloaded on org.fedoraproject.FirewallD1 for firewalld reload + */ + dbus_bus_add_match(sysbus, + "type='signal'" + ",interface='"DBUS_INTERFACE_DBUS"'" + ",member='NameOwnerChanged'" + ",arg0='org.fedoraproject.FirewallD1'", + NULL); + dbus_bus_add_match(sysbus, + "type='signal'" + ",interface='org.fedoraproject.FirewallD1'" + ",member='Reloaded'", + NULL); + dbus_connection_add_filter(sysbus, firewalld_dbus_filter_bridge, driverState, NULL); + } +#endif + return 0; out_of_memory: diff --git a/src/util/ebtables.c b/src/util/ebtables.c index ca056b1..6f4d151 100644 --- a/src/util/ebtables.c +++ b/src/util/ebtables.c @@ -176,11 +176,34 @@ ebtablesAddRemoveRule(ebtRules *rules, int action, const char *arg, ...) const char *s; int n, command_idx; +#if HAVE_FIREWALLD + int ret; + char *firewall_cmd_path = NULL; + virCommandPtr cmd = NULL; + + firewall_cmd_path = virFindFileInPath("firewall-cmd"); + if (firewall_cmd_path) { + cmd = virCommandNew(firewall_cmd_path); + virCommandAddArgList(cmd, "--state", NULL); + ret = virCommandRun(cmd, NULL); + if (ret != 0) { + VIR_FREE(firewall_cmd_path); + firewall_cmd_path = NULL; + } + virCommandFree(cmd); + } +#endif + n = 1 + /* /sbin/ebtables */ 2 + /* --table foo */ 2 + /* --insert bar */ 1; /* arg */ +#if HAVE_FIREWALLD + if (firewall_cmd_path) + n += 3; /* --direct --passthrough eb */ +#endif + va_start(args, arg); while (va_arg(args, const char *)) n++; @@ -192,6 +215,18 @@ ebtablesAddRemoveRule(ebtRules *rules, int action, const char *arg, ...) n = 0; +#if HAVE_FIREWALLD + if (firewall_cmd_path) { + if (!(argv[n++] = strdup(firewall_cmd_path))) + goto error; + if (!(argv[n++] = strdup("--direct"))) + goto error; + if (!(argv[n++] = strdup("--passthrough"))) + goto error; + if (!(argv[n++] = strdup("eb"))) + goto error; + } else +#endif if (!(argv[n++] = strdup(EBTABLES_PATH))) goto error; diff --git a/src/util/iptables.c b/src/util/iptables.c index b23aca9..1378575 100644 --- a/src/util/iptables.c +++ b/src/util/iptables.c @@ -101,9 +101,26 @@ iptablesAddRemoveRule(iptRules *rules, int family, int action, { va_list args; int ret; - virCommandPtr cmd; + virCommandPtr cmd = NULL; const char *s; - +#if HAVE_FIREWALLD + char *firewall_cmd_path = NULL; + + firewall_cmd_path = virFindFileInPath("firewall-cmd"); + if (firewall_cmd_path) { + cmd = virCommandNew(firewall_cmd_path); + virCommandAddArgList(cmd, "--state", NULL); + ret = virCommandRun(cmd, NULL); + if (ret == 0) { + cmd = virCommandNew(firewall_cmd_path); + virCommandAddArgList(cmd, "--direct", "--passthrough", + (family == AF_INET6) ? "ipv6" : "ipv4", NULL); + } else + cmd = NULL; + VIR_FREE(firewall_cmd_path); + } + if (!cmd) +#endif cmd = virCommandNew((family == AF_INET6) ? IP6TABLES_PATH : IPTABLES_PATH); -- 1.7.11.2

4 4

[libvirt] [PATCH 1/4] Fix man page file paths to real paths
by Doug Goldstein 15 Aug '12

15 Aug '12

Currently the man page has paths that start with @sysconfdir@, @localstatedir@ and @remote_pid_file@. The sed command attempts to replace these during the build but unfortunately pod2man gets to the files first and escapes the @ character resulting in the sed not working. This removes the @ character and makes the paths correct. --- daemon/Makefile.am | 6 +++--- daemon/libvirtd.pod.in | 18 +++++++++--------- 2 files changed, 12 insertions(+), 12 deletions(-) diff --git a/daemon/Makefile.am b/daemon/Makefile.am index 928aeaf..c15c785 100644 --- a/daemon/Makefile.am +++ b/daemon/Makefile.am @@ -85,9 +85,9 @@ CLEANFILES += test_libvirtd.aug libvirtd.8: $(srcdir)/libvirtd.8.in sed \ - -e 's![@]sysconfdir[@]!$(sysconfdir)!g' \ - -e 's![@]localstatedir[@]!$(localstatedir)!g' \ - -e 's![@]remote_pid_file[@]!$(REMOTE_PID_FILE)!g' \ + -e 's!SYSCONFDIR!$(sysconfdir)!g' \ + -e 's!LOCALSTATEDIR!$(localstatedir)!g' \ + -e 's!REMOTE_PID_FILE!$(REMOTE_PID_FILE)!g' \ < $< > $@-t mv $@-t $@ diff --git a/daemon/libvirtd.pod.in b/daemon/libvirtd.pod.in index 57a3b27..7dc8d83 100644 --- a/daemon/libvirtd.pod.in +++ b/daemon/libvirtd.pod.in @@ -77,14 +77,14 @@ On receipt of B<SIGHUP> libvirtd will reload its configuration. =over -=item F<@sysconfdir@/libvirtd.conf> +=item F<SYSCONFDIR/libvirtd.conf> The default configuration file used by libvirtd, unless overridden on the command line using the B<-f>|B<--config> option. -=item F<@localstatedir@/run/libvirt/libvirt-sock> +=item F<LOCALSTATEDIR/run/libvirt/libvirt-sock> -=item F<@localstatedir@/run/libvirt/libvirt-sock-ro> +=item F<LOCALSTATEDIR/run/libvirt/libvirt-sock-ro> The sockets libvirtd will use when B<run as root>. @@ -92,19 +92,19 @@ The sockets libvirtd will use when B<run as root>. The socket libvirtd will use when run as a B<non-root> user. -=item F<@sysconfdir@/pki/CA/cacert.pem> +=item F<SYSCONFDIR/pki/CA/cacert.pem> The TLS B<Certificate Authority> certificate libvirtd will use. -=item F<@sysconfdir@/pki/libvirt/servercert.pem> +=item F<SYSCONFDIR/pki/libvirt/servercert.pem> The TLS B<Server> certificate libvirtd will use. -=item F<@sysconfdir@/pki/libvirt/private/serverkey.pem> +=item F<SYSCONFDIR/pki/libvirt/private/serverkey.pem> The TLS B<Server> private key libvirtd will use. -=item F<@remote_pid_file@> +=item F<REMOTE_PID_FILE> The PID file to use, unless overridden by the B<-p>|B<--pid-file> option. @@ -121,8 +121,8 @@ To retrieve the version of libvirtd: To start libvirtd, instructing it to daemonize and create a PID file: # libvirtd -d - # ls -la @remote_pid_file@ - -rw-r--r-- 1 root root 6 Jul 9 02:40 @remote_pid_file@ + # ls -la REMOTE_PID_FILE + -rw-r--r-- 1 root root 6 Jul 9 02:40 REMOTE_PID_FILE # =head1 BUGS -- 1.7.8.6

2 4

[libvirt] the arguments of qemudDispatchServerEvent
by mzawdx wang 15 Aug '12

15 Aug '12

Hi all: I am now reading the source codes of libvirt with version 0.9.2 , and be stuck by the function "qemudDispatchServerEvent". as you can see: static void qemudDispatchServerEvent(int watch, int fd, int events, void *opaque) { ... } My question is : except for the argument "void *opaque"， when and how the other three arguments (int watch, int fd, int events) being incoming ? For it is a static function, I look through the source file "libvirt.c" and find no answer . Could someone give me a help ? Thanks a lot. Yours. mzawdx

2 2

[libvirt] Schedule for next release: entering freeze in a week
by Daniel Veillard 15 Aug '12

15 Aug '12

I would rather have the release a bit earlier than expected, I was initially hoping to get a new set of localization for Sep 4 but it is clear at this point that we won't have them in time. Since we have been working on 0.10.0 for 6 weeks now, I think pushing it 2 weeks from now is reasonnable, so I suggest to enter the freeze next week on Wed 22 for a release on the 29 Aug (or a bit earlier if things go well). Considering the amount of changes done for 0.10.0, (we are already at 370 patches in) my guess is that an extra release at the end of September not too long after 0.10.0 but with enough testing feedack will make maintainance easier. So I'm shooting for a 0.10.1 release aproximately 4 weeks after 0.10.0 itself, If such a schedule is raising problem, please tell :-) And if you have patches you want in 0.10.0 which didn't got reviews or isn't pushed, please git rebase them and send them again so they get the attention needed and get pushed within a week ! thanks, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 1

[libvirt] stateless drivers and stored lists of virDomainObjs
by Dmitry Guryanov 15 Aug '12

15 Aug '12

Hello, vmware, openvz and parallels driver, which was written with first two ones as an example, get list of domains in virConnectOpen and then use it for lookups and even for listing sometimes. This is not correct, because you can make several connections to the same driver and changes, made in one connection will not be visible to other. And you can also create or destroy domains using native tools. So I think keeping list of domains in a stateless driver is a bad idea - you should update this list each time when user requests some info. It's better to retrieve needed data only when it requested, like in xenapi driver. I asked this because I want to continue developing parallels driver and it'= s not so hard to change it now, so it should be fixed before adding any new code there. Maybe vmware and openvz drivers should be also fixed. -- Dmitry Guryanov

2 2