- Devel - libvirt List Archives

[libvirt] [libvirt-glib 1/2] spec: Get min libvirt version from configure.ac
by Christophe Fergeau 03 Oct '12

03 Oct '12

The minimum libvirt version we need is duplicated in configure.ac and libvirt.spec.in. Since libvirt.spec.in already gets some variables substituted, we can also substitute LIBVIRT_REQUIRED to avoid the duplication. --- configure.ac | 1 + libvirt-glib.spec.in | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index 7262322..369b368 100644 --- a/configure.ac +++ b/configure.ac @@ -10,6 +10,7 @@ AC_CANONICAL_HOST AM_SILENT_RULES([yes]) LIBVIRT_REQUIRED=0.9.10 +AC_SUBST([LIBVIRT_REQUIRED]) dnl used in the .spec file GLIB2_REQUIRED=2.22.0 GOBJECT2_REQUIRED=2.10.0 GIO_REQUIRED=2.10.0 diff --git a/libvirt-glib.spec.in b/libvirt-glib.spec.in index e312926..f10c38a 100644 --- a/libvirt-glib.spec.in +++ b/libvirt-glib.spec.in @@ -18,7 +18,7 @@ %endif %define with_vala %{with_introspection} -%define libvirt_version 0.9.10 +%define libvirt_version @LIBVIRT_REQUIRED@ Name: @PACKAGE@ Version: @VERSION@ -- 1.7.12.1

2 2

[libvirt] [libvirt-glib 1/2] gobject: Handle PMSUSPENDED state from libvirt
by Christophe Fergeau 03 Oct '12

03 Oct '12

Emit a signal when switching to the PMSUSPENDED state, and add an enum entry to describe this state. This avoids runtime warnings with newer libvirt. --- libvirt-gobject/libvirt-gobject-connection.c | 7 +++++++ libvirt-gobject/libvirt-gobject-domain.c | 11 +++++++++++ libvirt-gobject/libvirt-gobject-domain.h | 19 +++++++++++-------- 3 files changed, 29 insertions(+), 8 deletions(-) diff --git a/libvirt-gobject/libvirt-gobject-connection.c b/libvirt-gobject/libvirt-gobject-connection.c index 428ae38..ad7aa07 100644 --- a/libvirt-gobject/libvirt-gobject-connection.c +++ b/libvirt-gobject/libvirt-gobject-connection.c @@ -389,6 +389,13 @@ static int domain_event_cb(virConnectPtr conn G_GNUC_UNUSED, case VIR_DOMAIN_EVENT_SHUTDOWN: break; + case VIR_DOMAIN_EVENT_PMSUSPENDED: + if (detail == VIR_DOMAIN_EVENT_PMSUSPENDED_MEMORY) + g_signal_emit_by_name(gdom, "pmsuspended::memory"); + else + g_warn_if_reached(); + break; + default: g_warn_if_reached(); } diff --git a/libvirt-gobject/libvirt-gobject-domain.c b/libvirt-gobject/libvirt-gobject-domain.c index bcfad2a..f8ad493 100644 --- a/libvirt-gobject/libvirt-gobject-domain.c +++ b/libvirt-gobject/libvirt-gobject-domain.c @@ -55,6 +55,7 @@ enum { VIR_RESUMED, VIR_STOPPED, VIR_UPDATED, + VIR_PMSUSPENDED, LAST_SIGNAL }; @@ -225,6 +226,16 @@ static void gvir_domain_class_init(GVirDomainClass *klass) G_TYPE_NONE, 0); + signals[VIR_PMSUSPENDED] = g_signal_new("pmsuspended", + G_OBJECT_CLASS_TYPE(object_class), + G_SIGNAL_RUN_LAST | G_SIGNAL_NO_RECURSE | + G_SIGNAL_NO_HOOKS | G_SIGNAL_DETAILED, + G_STRUCT_OFFSET(GVirDomainClass, pmsuspended), + NULL, NULL, + g_cclosure_marshal_VOID__VOID, + G_TYPE_NONE, + 0); + g_type_class_add_private(klass, sizeof(GVirDomainPrivate)); } diff --git a/libvirt-gobject/libvirt-gobject-domain.h b/libvirt-gobject/libvirt-gobject-domain.h index 7810d1b..fc2db7c 100644 --- a/libvirt-gobject/libvirt-gobject-domain.h +++ b/libvirt-gobject/libvirt-gobject-domain.h @@ -66,19 +66,22 @@ struct _GVirDomainClass void (*resumed)(GVirDomain *dom); void (*updated)(GVirDomain *dom); void (*suspended)(GVirDomain *dom); + void (*pmsuspended)(GVirDomain *dom); - gpointer padding[20]; + gpointer padding[19]; }; typedef enum { - GVIR_DOMAIN_STATE_NONE = 0, /* no state */ - GVIR_DOMAIN_STATE_RUNNING = 1, /* the domain is running */ - GVIR_DOMAIN_STATE_BLOCKED = 2, /* the domain is blocked on resource */ - GVIR_DOMAIN_STATE_PAUSED = 3, /* the domain is paused by user */ - GVIR_DOMAIN_STATE_SHUTDOWN= 4, /* the domain is being shut down */ - GVIR_DOMAIN_STATE_SHUTOFF = 5, /* the domain is shut off */ - GVIR_DOMAIN_STATE_CRASHED = 6 /* the domain is crashed */ + GVIR_DOMAIN_STATE_NONE = 0, /* no state */ + GVIR_DOMAIN_STATE_RUNNING = 1, /* the domain is running */ + GVIR_DOMAIN_STATE_BLOCKED = 2, /* the domain is blocked on resource */ + GVIR_DOMAIN_STATE_PAUSED = 3, /* the domain is paused by user */ + GVIR_DOMAIN_STATE_SHUTDOWN= 4, /* the domain is being shut down */ + GVIR_DOMAIN_STATE_SHUTOFF = 5, /* the domain is shut off */ + GVIR_DOMAIN_STATE_CRASHED = 6, /* the domain is crashed */ + GVIR_DOMAIN_STATE_PMSUSPENDED = 7 /* the domain is suspended by guest + power management */ } GVirDomainState; -- 1.7.11.4

2 4

[libvirt] [PATCH] build: update gnulib for FreeBSD build
by Eric Blake 03 Oct '12

03 Oct '12

* .gnulib: Update to latest, for gcc 4.2 support and fixed FreeBSD tests. --- Pushing under the build-breaker rule; this fixes the problems I found on FreeBSD 8.2 during 'make check' with CFLAGS=-g. * .gnulib b493832...bc8b4e9 (6): > pselect: reject invalid file descriptors > select: reject invalid file descriptors > ptsname: reject invalid file descriptors > hash-pjw-bare: new module > manywarnings: cater to more gcc infelicities > select, poll tests: Make setsockopt invocation effective. .gnulib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index b493832..bc8b4e9 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit b4938324b7113c9d73f8390d21f3ecbd842e74b9 +Subproject commit bc8b4e9f9d325e01d83e3ddff360a8b7d1f36cb3 -- 1.7.11.4

1 0

[libvirt] ipv6 related filtering through libvirt interfaces
by R P Herrold 03 Oct '12

03 Oct '12

Looking mid page at: http://libvirt.org/formatnwfilter.html#nwfelemsReservedVars I see that there are some un-implemented filtering variables for IPv6. Conspicuously absent is a variable set for the content handed out by radvd. The changes back at 0.8.0 [Apr 12 2010] seem to be the last time this area was touched [ The interaction of the dhcpd, a dhcp6d, and the radvd as to DNS server assignment is also on the horizon with the roll in of the systemd ] I used the search tool to look for a current 'ROADMAP' without getting meaningful results; The 'TODO' page is empty [1]; the patch queue page seems unmaintained [2]. I cannot find a 'roadmap' posting in my archive this year ;( With the exhaustion of 'easy' ipv4 space allocations, we are turning to deploying ipv6 only VM's, but finding that we are manually maintaining filtering rule sets, rather than handing them out in the XML configs Is there movement in this space 'behind the scenes'? Thank you -- Russ herrold [1] http://libvirt.org/todo.html [2] http://libvirt.org/pending.html

1 0

[libvirt] [PATCH] build: avoid -Wno-format on new-enough gcc
by Eric Blake 02 Oct '12

02 Oct '12

Commit c579d6b added a sledgehammer to silence spurious warnings from gcc 4.2, but in the process, it also silenced useful warnings from newer gcc (such as 4.4). As a result, a bug slipped in to commit 0caccb58. Tested with FreeBSD (gcc 4.2.1), RHEL 6.3 (gcc 4.4), and F17 (gcc 4.7.2). * m4/virt-compile-warnings.m4 (-Wno-format): Probe for the actual spurious message, to once again allow gcc 4.4 to use -Wformat. --- m4/virt-compile-warnings.m4 | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/m4/virt-compile-warnings.m4 b/m4/virt-compile-warnings.m4 index d1173eb..f1b8f39 100644 --- a/m4/virt-compile-warnings.m4 +++ b/m4/virt-compile-warnings.m4 @@ -60,6 +60,23 @@ AC_DEFUN([LIBVIRT_COMPILE_WARNINGS],[ # gcc 4.4.6 complains this is C++ only; gcc 4.7.0 implies this from -Wall dontwarn="$dontwarn -Wenum-compare" + # gcc 4.2 treats attribute(format) as an implicit attribute(nonnull), + # which triggers spurious warnings for our usage + AC_CACHE_CHECK([whether gcc -Wformat allows NULL strings], + [lv_cv_gcc_wformat_null_works], [ + save_CFLAGS=$CFLAGS + CFLAGS='-Wunknown-pragmas -Werror -Wformat' + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ + #include <stddef.h> + static __attribute__ ((__format__ (__printf__, 1, 2))) int + foo (const char *fmt, ...) { return !fmt; } + ]], [[ + return foo(NULL); + ]])], + [lv_cv_gcc_wformat_null_works=yes], + [lv_cv_gcc_wformat_null_works=no]) + CFLAGS=$save_CFLAGS]) + # Gnulib uses '#pragma GCC diagnostic push' to silence some # warnings, but older gcc doesn't support this. AC_CACHE_CHECK([whether pragma GCC diagnostic push works], @@ -117,7 +134,7 @@ AC_DEFUN([LIBVIRT_COMPILE_WARNINGS],[ # ATTRIBUTE_FMT_PRINT, which causes -Wformat failure on our # intentional use of virReportError(code, NULL). gl_WARN_ADD([-Wno-format-nonliteral]) - if test $lv_cv_gcc_pragma_push_works = no; then + if test $lv_cv_gcc_wformat_null_works = no; then gl_WARN_ADD([-Wno-format]) fi -- 1.7.11.4

2 2

[libvirt] [PATCH] build: fix typo in debug message
by Eric Blake 02 Oct '12

02 Oct '12

Introduced in commit 0caccb58. CC libvirt_driver_qemu_impl_la-qemu_capabilities.lo ../../src/qemu/qemu_capabilities.c: In function 'qemuCapsInitQMP': ../../src/qemu/qemu_capabilities.c:2327:13: error: format '%d' expects argument of type 'int', but argument 8 has type 'const char *' [-Werror=format] * src/qemu/qemu_capabilities.c (qemuCapsInitQMP): Use correct format. --- Pushing under the build-breaker rule. src/qemu/qemu_capabilities.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 759f5bb..a5eb995 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -2324,7 +2324,7 @@ cleanup: int rc; if ((rc = virPidFileReadPath(pidfile, &pid)) < 0) { - VIR_DEBUG("Failed to read pidfile %s: %d", + VIR_DEBUG("Failed to read pidfile %s: %s", pidfile, virStrerror(-rc, ebuf, sizeof(ebuf))); } else { VIR_DEBUG("Killing QMP caps process %lld", (long long) pid); -- 1.7.11.4

1 1

[libvirt] [PATCH] qemu: Kill processes used for QMP caps probing
by Jiri Denemark 02 Oct '12

02 Oct '12

Since libvirt switched to QMP capabilities probing recently, it starts QEMU process used for this probing with -daemonize, which means virCommandAbort can no longer reach these processes. As a result of that, restarting libvirtd will leave several new QEMU processes behind. Let's use QEMU's -pidfile and use it to kill the process when QMP caps probing is done. --- src/qemu/qemu_capabilities.c | 51 ++++++++++++++++++++++++++++++++++++++------ src/qemu/qemu_capabilities.h | 5 +++-- src/qemu/qemu_driver.c | 4 +++- 3 files changed, 51 insertions(+), 9 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 20b350a..bfefa92 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -29,6 +29,8 @@ #include "virterror_internal.h" #include "util.h" #include "virfile.h" +#include "virpidfile.h" +#include "virprocess.h" #include "nodeinfo.h" #include "cpu/cpu.h" #include "domain_conf.h" @@ -214,6 +216,7 @@ struct _qemuCapsCache { virMutex lock; virHashTablePtr binaries; char *libDir; + char *runDir; }; @@ -2180,7 +2183,8 @@ qemuCapsInitQMPBasic(qemuCapsPtr caps) static int qemuCapsInitQMP(qemuCapsPtr caps, - const char *libDir) + const char *libDir, + const char *runDir) { int ret = -1; virCommandPtr cmd = NULL; @@ -2191,7 +2195,11 @@ qemuCapsInitQMP(qemuCapsPtr caps, virDomainChrSourceDef config; char *monarg = NULL; char *monpath = NULL; + char *pidfile = NULL; + /* the ".sock" sufix is important to avoid a possible clash with a qemu + * domain called "capabilities" + */ if (virAsprintf(&monpath, "%s/%s", libDir, "capabilities.monitor.sock") < 0) { virReportOOMError(); goto cleanup; @@ -2201,6 +2209,14 @@ qemuCapsInitQMP(qemuCapsPtr caps, goto cleanup; } + /* ".pidfile" suffix is used rather than ".pid" to avoid a possible clash + * with a qemu domain called "capabilities" + */ + if (virAsprintf(&pidfile, "%s/%s", runDir, "capabilities.pidfile") < 0) { + virReportOOMError(); + goto cleanup; + } + memset(&config, 0, sizeof(config)); config.type = VIR_DOMAIN_CHR_TYPE_UNIX; config.data.nix.path = monpath; @@ -2215,6 +2231,7 @@ qemuCapsInitQMP(qemuCapsPtr caps, "-nographic", "-M", "none", "-qmp", monarg, + "-pidfile", pidfile, "-daemonize", NULL); virCommandAddEnvPassCommon(cmd); @@ -2300,12 +2317,32 @@ cleanup: virCommandFree(cmd); VIR_FREE(monarg); VIR_FREE(monpath); + + if (pidfile) { + char ebuf[1024]; + pid_t pid; + int rc; + + if ((rc = virPidFileReadPath(pidfile, &pid)) < 0) { + VIR_DEBUG("Failed to read pidfile %s: %d", + pidfile, virStrerror(-rc, ebuf, sizeof(ebuf))); + } else { + VIR_DEBUG("Killing QMP caps process %lld", (long long) pid); + if (virProcessKill(pid, SIGKILL) < 0) + VIR_DEBUG("Failed to kill process %lld: %s", + (long long) pid, + virStrerror(errno, ebuf, sizeof(ebuf))); + } + unlink(pidfile); + VIR_FREE(pidfile); + } return ret; } qemuCapsPtr qemuCapsNewForBinary(const char *binary, - const char *libDir) + const char *libDir, + const char *runDir) { qemuCapsPtr caps = qemuCapsNew(); struct stat sb; @@ -2333,7 +2370,7 @@ qemuCapsPtr qemuCapsNewForBinary(const char *binary, goto error; } - if ((rv = qemuCapsInitQMP(caps, libDir)) < 0) + if ((rv = qemuCapsInitQMP(caps, libDir, runDir)) < 0) goto error; if (!caps->usedQMP && @@ -2373,7 +2410,7 @@ qemuCapsHashDataFree(void *payload, const void *key ATTRIBUTE_UNUSED) qemuCapsCachePtr -qemuCapsCacheNew(const char *libDir) +qemuCapsCacheNew(const char *libDir, const char *runDir) { qemuCapsCachePtr cache; @@ -2391,7 +2428,8 @@ qemuCapsCacheNew(const char *libDir) if (!(cache->binaries = virHashCreate(10, qemuCapsHashDataFree))) goto error; - if (!(cache->libDir = strdup(libDir))) { + if (!(cache->libDir = strdup(libDir)) || + !(cache->runDir = strdup(runDir))) { virReportOOMError(); goto error; } @@ -2420,7 +2458,7 @@ qemuCapsCacheLookup(qemuCapsCachePtr cache, const char *binary) if (!ret) { VIR_DEBUG("Creating capabilities for %s", binary); - ret = qemuCapsNewForBinary(binary, cache->libDir); + ret = qemuCapsNewForBinary(binary, cache->libDir, cache->runDir); if (ret) { VIR_DEBUG("Caching capabilities %p for %s", ret, binary); @@ -2459,6 +2497,7 @@ qemuCapsCacheFree(qemuCapsCachePtr cache) return; VIR_FREE(cache->libDir); + VIR_FREE(cache->runDir); virHashFree(cache->binaries); virMutexDestroy(&cache->lock); VIR_FREE(cache); diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 89f351c..5d343c1 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -163,7 +163,8 @@ typedef qemuCapsCache *qemuCapsCachePtr; qemuCapsPtr qemuCapsNew(void); qemuCapsPtr qemuCapsNewCopy(qemuCapsPtr caps); qemuCapsPtr qemuCapsNewForBinary(const char *binary, - const char *libDir); + const char *libDir, + const char *runDir); int qemuCapsProbeQMP(qemuCapsPtr caps, qemuMonitorPtr mon); @@ -201,7 +202,7 @@ int qemuCapsGetMachineTypesCaps(qemuCapsPtr caps, bool qemuCapsIsValid(qemuCapsPtr caps); -qemuCapsCachePtr qemuCapsCacheNew(const char *libDir); +qemuCapsCachePtr qemuCapsCacheNew(const char *libDir, const char *runDir); qemuCapsPtr qemuCapsCacheLookup(qemuCapsCachePtr cache, const char *binary); qemuCapsPtr qemuCapsCacheLookupCopy(qemuCapsCachePtr cache, const char *binary); void qemuCapsCacheFree(qemuCapsCachePtr cache); diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index b0a0bb5..55d8027 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -767,7 +767,9 @@ qemudStartup(int privileged) { if (qemuSecurityInit(qemu_driver) < 0) goto error; - if ((qemu_driver->capsCache = qemuCapsCacheNew(qemu_driver->libDir)) == NULL) + qemu_driver->capsCache = qemuCapsCacheNew(qemu_driver->libDir, + qemu_driver->stateDir); + if (!qemu_driver->capsCache) goto error; if ((qemu_driver->caps = qemuCreateCapabilities(qemu_driver)) == NULL) -- 1.7.12

2 2

[libvirt] installing libvirt
by Felipe Oliveira Gutierrez 02 Oct '12

02 Oct '12

Hi everyone, I am new using libvirt, Do I have to install libvirt-0.10.2.tar.gz alone, or I need to install libvirt-java-0.4.9.tar.gz after to developing codes using java? I runned "# ant build" from libvirt-java-0.4.9, and got this: Buildfile: /libvirt-java-0.4.9/build.xml init: [copy] Copying 1 file to /libvirt-java-0.4.9 build: [javac] /libvirt-java-0.4.9/build.xml:42: warning: 'includeantruntime' was not set, defaulting to build.sysclasspath=last; set to false for repeatable builds [javac] /libvirt-java-0.4.9/build.xml:43: warning: 'includeantruntime' was not set, defaulting to build.sysclasspath=last; set to false for repeatable builds BUILD SUCCESSFUL On the README file he asks: 2. You must have the libvirt.jar file in your classpath. By default the installs it to /usr/share/java/0.4.9.jar But I don't find the libvirt.jar file to copy. Does anyone can help me please? Bets Regards, Felipe -- *-- Felipe Oliveira Gutierrez* *-- **lipe.82(a)gmail.com* <lipe.82(a)gmail.com> *-- https://sites.google.com/site/lipe82/Home/diaadia*

2 2

[libvirt] [PATCH] qemu: Use proper agent entering function when freezing filesystems
by Peter Krempa 02 Oct '12

02 Oct '12

When doing snapshots, the filesystem freeze function used the agent entering function that expects the qemud_driver unlocked. This might cause a deadlock of the qemu driver if the agent does not respond. The only call path of this function has the qemud_driver locked, so this patch changes the entering functions to those expecting the driver locked. --- src/qemu/qemu_driver.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index b0a0bb5..50f44ce 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -10375,6 +10375,7 @@ qemuDomainSnapshotIsAllowed(virDomainObjPtr vm) return true; } +/* this function expects the driver lock to be held by the caller */ static int qemuDomainSnapshotFSFreeze(struct qemud_driver *driver, virDomainObjPtr vm) { @@ -10393,9 +10394,9 @@ qemuDomainSnapshotFSFreeze(struct qemud_driver *driver, return -1; } - qemuDomainObjEnterAgent(driver, vm); + qemuDomainObjEnterAgentWithDriver(driver, vm); freezed = qemuAgentFSFreeze(priv->agent); - qemuDomainObjExitAgent(driver, vm); + qemuDomainObjExitAgentWithDriver(driver, vm); return freezed; } -- 1.7.12

2 2

[libvirt] only 256 active networks
by Dan Kenigsberg 02 Oct '12

02 Oct '12

Hi List, Laine, Maybe you can help here. We define 256 bridged networks such as <network> <name>vdsm-test244</name> <uuid>c3acd4c3-37be-3f9d-ee6b-317513015f9e</uuid> <forward mode='bridge'/> <bridge name='test244' /> </network> and end up killing `virsh net-list` with error: Failed to list active networks error: too many remote undefineds: 257 > 256 error: Reconnected to the hypervisor What is this limitation? Why is it there? How can this be extended/avoided? Dan.

2 2