- Devel - libvirt List Archives

[libvirt] qemu-agent-command via isa-serial for freebsd
by Vasiliy Tolstov 30 Oct '15

30 Oct '15

Hi! I'm need to control freebsd guest via qemu-ga, as i see qemu-ga already supports isa-serial connection. How i need to configure domain via libvirt to able virsh qemu-agent-command to this guest vm? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

3 10

[libvirt] [PATCH] util: set error if DAD is not finished
by Luyao Huang 30 Oct '15

30 Oct '15

If DAD not finished in 5 seconds, user will get an unknown error like this: # virsh net-start ipv6 error: Failed to start network ipv6 error: An error occurred, but the cause is unknown Call virReportError to set an error. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- I found the DAD will take 7 seconds on my machine, and i cannot create a network which use ipv6 now :( . Can we offer a way allow user to change this timeout ? maybe add a configuration file option in libvirtd.conf. src/util/virnetdev.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index 9789e93..c8861e9 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -1398,7 +1398,13 @@ virNetDevWaitDadFinish(virSocketAddrPtr *addrs, size_t count) VIR_FREE(resp); } /* Check timeout. */ - ret = dad ? -1 : 0; + if (dad) { + virReportError(VIR_ERR_SYSTEM_ERROR, + _("Duplicate Address Detection " + "not finished in %d seconds"), VIR_DAD_WAIT_TIMEOUT); + } else { + ret = 0; + } cleanup: VIR_FREE(resp); -- 1.8.3.1

5 5

[libvirt] Again cgroups and isolcpus
by Henning Schild 29 Oct '15

29 Oct '15

Hi folks, i already started a discussion on the interaction of cgroups and isolcpus a while ago. But now i believe i have got a better understanding of how the two interact and i can describe problems that arise from that. The scenario: A machine that runs realtime tasks on pcpus reserved with isolcpus. It also runs VMs with the help of libvirt. It might also run realtime VMs with the help of libvirt. Moving a task into a new cgroup/cpuset and some modifications of the cpus in that set imply a setaffinity by the kernel. That affinity setting will ignore isolcpus. The result is possible "interference by" or "starvation of" these tasks. Now let me describe one scenario where that implicit setaffinity becomes a problem for our realtime system. libvirt creates a superset called the machine.slice and subsets called emulator and vpuX. By default the machine.slice inherits from the root which contains all pcpus, also the isolated ones. Now moving a task into that superset will place that task on isolcpus where it might interfere or simply starve. Turns out that a fresh qemu actually is put into that superset. That is a bug that should be fixed but let me address that one in another mail. My current point of view is that we need a strong mechanism to isolate cpus. isolcpus just is not good enough. The measure of choice probably is cpusets as well, and this time with the exclusive flag turned on. That will stop every other cpuset user from messing around with those cpus by accident. I am thinking of one or more cpusets where isolated cpus are parked and not used within this cgroup. Anyone wanting to use one of them will have to take it out there and explictily put it into their a new set. Now if libvirt makes the mistake to have tasks running in supersets these tasks will spread to the newly added rt-cpu. Or new tasks that run in supersets will end up on rt-cpus already in use. But at least we have containment in libvirt and the VMs it spawned. For alloc and free of rt-cpus i am planning to use libvirt hooks to begin with, from what i read they should enable me to do what i need. What do you guys think about the general idea to address the described problem? I will implement a prototype of the alloc-free of rt-cpus. My current hope is that libvirt hooks can be abused for that. I am thinking that at some point libvirt should be able to do that without hooks. It should get a notion of reserved ressources that are currently parked in other cgroups. My current suspicion is that the cpusets might just be the tip of the iceberg. -- for now i am running libvirt without cgroups to keep my isolcpus free cgroups/cpusets offer a switch to make a cpu exclusive to a set. That switch is great because it will act as an assert, a second line of defense. Having seen how cpusets and migration mess around with affinities i guess for realtime people have to insist on that second line of defense. Especially in times where cgroups are all over the place. In openstack one would actually say that a pcpu should be "dedicated". That will result in a vcpupin on exactly one pcpu. Unfortunately one meaning of "dedicated" gets lost in translation. It could otherwise be used by libvirtd to set cpuset.cpu_exclusive in the vcpu-cgroup. And i am bringing that up here because i do not think libvirt allows me to influence the cpu_exclusive flag for my vcpu cgroups. Henning

1 0

[libvirt] libvirt guest-agent-command error
by Vasiliy Tolstov 29 Oct '15

29 Oct '15

Hi. I'm debug some issues with guest-agent inside vm and find this bug, that affected me https://bugzilla.redhat.com/show_bug.cgi?id=1090551 Does anybody knows how to properly fix this ? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

2 1

[libvirt] [PATCH] wireshark: Install to generic plugin directory
by Michal Privoznik 29 Oct '15

29 Oct '15

There has been a report on the list [1] that we are not installing the wireshark dissector into the correct plugin directory. And in fact we are not. The problem is, the plugin directory path is constructed at compile time. However, it's dependent on the wireshark version, e.g. /usr/lib/wireshark/1.12.6/plugins/ This is rather unfortunate, because if libvirt RPMs were built with one version, but installed on a system with newer one, the plugins are not really loaded. This problem lead fedora packagers to unify plugin path to: /usr/lib/wireshark/plugins/ Cool! But this was enabled just in wireshark-1.12.6-4. Therefore, we must require at least that version. And while at it, on some distributions, the wireshark.pc file already has a variable that defines where plugin dir is. Use that if possible. 1: https://www.redhat.com/archives/libvirt-users/2015-October/msg00063.html Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- libvirt.spec.in | 6 ++++-- m4/virt-wireshark.m4 | 7 ++++++- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 9dff994..469bfca 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1170,7 +1170,7 @@ virtualization capabilities of recent versions of Linux (and other OSes). %package wireshark Summary: Wireshark dissector plugin for libvirt RPC transactions Group: Development/Libraries -Requires: wireshark +Requires: wireshark >= 1.12.6-4 Requires: %{name}-client = %{version}-%{release} %description wireshark @@ -1561,6 +1561,8 @@ rm -f $RPM_BUILD_ROOT%{_libdir}/libvirt/connection-driver/*.a %endif %if %{with_wireshark} rm -f $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/*/libvirt.la +mv $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/*/libvirt.so \ + $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/libvirt.so %endif # Temporarily get rid of not-installed libvirt-admin.so @@ -2279,7 +2281,7 @@ exit 0 %if %{with_wireshark} %files wireshark -%{_libdir}/wireshark/plugins/*/libvirt.so +%{_libdir}/wireshark/plugins/libvirt.so %endif %if %{with_lxc} diff --git a/m4/virt-wireshark.m4 b/m4/virt-wireshark.m4 index 47204ed..199317e 100644 --- a/m4/virt-wireshark.m4 +++ b/m4/virt-wireshark.m4 @@ -28,7 +28,12 @@ AC_DEFUN([LIBVIRT_CHECK_WIRESHARK],[ dnl Check for system location of wireshark plugins if test "x$with_wireshark_dissector" != "xno" ; then if test "x$with_ws_plugindir" = "xcheck" ; then - ws_plugindir="$libdir/wireshark/plugins/$($PKG_CONFIG --modversion wireshark)" + ws_plugindir="$($PKG_CONFIG --variable plugindir wireshark)" + if test "x$ws_plugindir" = "x" ; then + dnl On some systems the plugindir variable may not be stored within pkg config. + dnl Fall back to older style of constructing the plugin dir path. + ws_plugindir="$libdir/wireshark/plugins/$($PKG_CONFIG --modversion wireshark)" + fi elif test "x$with_ws_plugindir" = "xno" || test "x$with_ws_plugindir" = "xyes"; then AC_MSG_ERROR([ws-plugindir must be used only with valid path]) else -- 2.4.10

2 1

[libvirt] virsh can't support VM offline blockcommit
by justlibvirt＠gmail.com 29 Oct '15

29 Oct '15

Hi everyone! I use the libvirt(version: 1.2.2) and QEMU(version: 2.2.1) to test qemu snapshot features: I tried virsh blockcommit when VM offline, the virsh blockcommit failed: the error messase as below: error: Requested operation is not valid: domain is not running when I start the VM, the virsh blockcommit work fine! my question is：we need virsh support blockcommit when VM offline。how do？ we test qemu snapshot as follows： we need create snapshot of the cycle，and by virsh blockcommit merge snapshot chain，but when VM offline， the virsh blockcommit will fail， someone suggest me：when VM offline，may use qemu-img commit to submit snapshot turn，but qemu-img can't modify snapshot chain backing relation，it need we use qemu-img rebase amend snapshot‘s backingfile， the main question is when we use qemu-img rebase modify snapshot's backingfile, if someone start the VM, it will may lead to VM filesystem readonly。 we should how to solve the question? thanks! justlibvirt(a)gmail.com

2 3

[libvirt] [PATCH] Remove new lines from debug messages
by Jiri Denemark 29 Oct '15

29 Oct '15

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/nwfilter/nwfilter_gentech_driver.c | 2 +- src/nwfilter/nwfilter_learnipaddr.c | 10 +++++----- src/rpc/virnetsocket.c | 2 +- src/util/virfile.c | 2 +- src/util/virhash.c | 2 +- src/util/virnetdevmacvlan.c | 28 +++++++++++++------------- src/util/virprocess.c | 2 +- src/xen/xend_internal.c | 2 +- tests/virhostdevtest.c | 36 +++++++++++++++++----------------- tests/virnetsockettest.c | 2 +- tests/virtimetest.c | 2 +- 11 files changed, 45 insertions(+), 45 deletions(-) diff --git a/src/nwfilter/nwfilter_gentech_driver.c b/src/nwfilter/nwfilter_gentech_driver.c index 701f8d8..5a4cff1 100644 --- a/src/nwfilter/nwfilter_gentech_driver.c +++ b/src/nwfilter/nwfilter_gentech_driver.c @@ -540,7 +540,7 @@ virNWFilterDetermineMissingVarsRec(virNWFilterDefPtr filter, if (rc) break; } else if (inc) { - VIR_DEBUG("Following filter %s\n", inc->filterref); + VIR_DEBUG("Following filter %s", inc->filterref); obj = virNWFilterObjFindByName(&driver->nwfilters, inc->filterref); if (obj) { diff --git a/src/nwfilter/nwfilter_learnipaddr.c b/src/nwfilter/nwfilter_learnipaddr.c index 5b55055..1adbadb 100644 --- a/src/nwfilter/nwfilter_learnipaddr.c +++ b/src/nwfilter/nwfilter_learnipaddr.c @@ -413,7 +413,7 @@ learnIPAddressThread(void *arg) handle = pcap_open_live(listen_if, BUFSIZ, 0, PKT_TIMEOUT_MS, errbuf); if (handle == NULL) { - VIR_DEBUG("Couldn't open device %s: %s\n", listen_if, errbuf); + VIR_DEBUG("Couldn't open device %s: %s", listen_if, errbuf); req->status = ENODEV; goto done; } @@ -448,13 +448,13 @@ learnIPAddressThread(void *arg) filter = virBufferContentAndReset(&buf); if (pcap_compile(handle, &fp, filter, 1, 0) != 0) { - VIR_DEBUG("Couldn't compile filter '%s'.\n", filter); + VIR_DEBUG("Couldn't compile filter '%s'", filter); req->status = EINVAL; goto done; } if (pcap_setfilter(handle, &fp) != 0) { - VIR_DEBUG("Couldn't set filter '%s'.\n", filter); + VIR_DEBUG("Couldn't set filter '%s'", filter); req->status = EINVAL; pcap_freecode(&fp); goto done; @@ -626,7 +626,7 @@ learnIPAddressThread(void *arg) req->filtername, req->filterparams); VIR_DEBUG("Result from applying firewall rules on " - "%s with IP addr %s : %d\n", req->ifname, inetaddr, ret); + "%s with IP addr %s : %d", req->ifname, inetaddr, ret); } } else { if (showError) @@ -638,7 +638,7 @@ learnIPAddressThread(void *arg) techdriver->applyDropAllRules(req->ifname); } - VIR_DEBUG("pcap thread terminating for interface %s\n", req->ifname); + VIR_DEBUG("pcap thread terminating for interface %s", req->ifname); virNWFilterUnlockIface(req->ifname); diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c index 5e5f1ab..526d291 100644 --- a/src/rpc/virnetsocket.c +++ b/src/rpc/virnetsocket.c @@ -196,7 +196,7 @@ int virNetSocketCheckProtocols(bool *hasIPv4, freeaddrinfo(ai); - VIR_DEBUG("Protocols: v4 %d v6 %d\n", *hasIPv4, *hasIPv6); + VIR_DEBUG("Protocols: v4 %d v6 %d", *hasIPv4, *hasIPv6); ret = 0; cleanup: diff --git a/src/util/virfile.c b/src/util/virfile.c index e5cf2c5..f45e18f 100644 --- a/src/util/virfile.c +++ b/src/util/virfile.c @@ -681,7 +681,7 @@ static int virFileLoopDeviceOpen(char **dev_name) if (virFileLoopDeviceOpenLoopCtl(dev_name, &loop_fd) < 0) return -1; - VIR_DEBUG("Return from loop-control got fd %d\n", loop_fd); + VIR_DEBUG("Return from loop-control got fd %d", loop_fd); if (loop_fd >= 0) return loop_fd; diff --git a/src/util/virhash.c b/src/util/virhash.c index bc90c44..fab621b 100644 --- a/src/util/virhash.c +++ b/src/util/virhash.c @@ -287,7 +287,7 @@ virHashGrow(virHashTablePtr table, size_t size) VIR_FREE(oldtable); #ifdef DEBUG_GROW - VIR_DEBUG("virHashGrow : from %d to %d, %ld elems\n", oldsize, + VIR_DEBUG("virHashGrow : from %d to %d, %ld elems", oldsize, size, nbElem); #endif diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index 89985b8..f85bd3e 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -473,16 +473,16 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, case RTM_DELLINK: case RTM_SETLINK: case RTM_GETLINK: - VIR_DEBUG(" IFINFOMSG\n"); - VIR_DEBUG(" ifi_family = 0x%02x\n", + VIR_DEBUG(" IFINFOMSG"); + VIR_DEBUG(" ifi_family = 0x%02x", ((struct ifinfomsg *)data)->ifi_family); - VIR_DEBUG(" ifi_type = 0x%x\n", + VIR_DEBUG(" ifi_type = 0x%x", ((struct ifinfomsg *)data)->ifi_type); - VIR_DEBUG(" ifi_index = %i\n", + VIR_DEBUG(" ifi_index = %i", ((struct ifinfomsg *)data)->ifi_index); - VIR_DEBUG(" ifi_flags = 0x%04x\n", + VIR_DEBUG(" ifi_flags = 0x%04x", ((struct ifinfomsg *)data)->ifi_flags); - VIR_DEBUG(" ifi_change = 0x%04x\n", + VIR_DEBUG(" ifi_change = 0x%04x", ((struct ifinfomsg *)data)->ifi_change); } /* DEBUG end */ @@ -544,29 +544,29 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, if (tb[IFLA_IFNAME]) { ifname = (char *)RTA_DATA(tb[IFLA_IFNAME]); - VIR_DEBUG("IFLA_IFNAME = %s\n", ifname); + VIR_DEBUG("IFLA_IFNAME = %s", ifname); } if (tb[IFLA_OPERSTATE]) { rem = *(unsigned short *)RTA_DATA(tb[IFLA_OPERSTATE]); - VIR_DEBUG("IFLA_OPERSTATE = %d\n", rem); + VIR_DEBUG("IFLA_OPERSTATE = %d", rem); } if (tb[IFLA_VF_PORTS]) { struct nlattr *tb_vf_ports; - VIR_DEBUG("found IFLA_VF_PORTS\n"); + VIR_DEBUG("found IFLA_VF_PORTS"); nla_for_each_nested(tb_vf_ports, tb[IFLA_VF_PORTS], rem) { - VIR_DEBUG("iterating\n"); + VIR_DEBUG("iterating"); if (nla_type(tb_vf_ports) != IFLA_VF_PORT) { - VIR_DEBUG("not a IFLA_VF_PORT. skipping\n"); + VIR_DEBUG("not a IFLA_VF_PORT. skipping"); continue; } if (nla_parse_nested(tb3, IFLA_PORT_MAX, tb_vf_ports, ifla_port_policy)) { VIR_DEBUG("nested parsing on level 2" - " failed."); + " failed"); } if (tb3[IFLA_PORT_VF]) { VIR_DEBUG("IFLA_PORT_VF = %d", @@ -600,7 +600,7 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, uuid = (unsigned char *) RTA_DATA(tb3[IFLA_PORT_INSTANCE_UUID]); instance2str(uuid, instance, sizeof(instance)); - VIR_DEBUG("IFLA_PORT_INSTANCE_UUID = %s\n", + VIR_DEBUG("IFLA_PORT_INSTANCE_UUID = %s", instance); } @@ -615,7 +615,7 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, } if (tb3[IFLA_PORT_RESPONSE]) { - VIR_DEBUG("IFLA_PORT_RESPONSE = %d\n", *(uint16_t *) + VIR_DEBUG("IFLA_PORT_RESPONSE = %d", *(uint16_t *) RTA_DATA(tb3[IFLA_PORT_RESPONSE])); } } diff --git a/src/util/virprocess.c b/src/util/virprocess.c index e6b78ef..d2bf360 100644 --- a/src/util/virprocess.c +++ b/src/util/virprocess.c @@ -407,7 +407,7 @@ virProcessKillPainfully(pid_t pid, bool force) int virProcessSetAffinity(pid_t pid, virBitmapPtr map) { size_t i; - VIR_DEBUG("Set process affinity on %lld\n", (long long)pid); + VIR_DEBUG("Set process affinity on %lld", (long long)pid); # ifdef CPU_ALLOC /* New method dynamically allocates cpu mask, allowing unlimted cpus */ int numcpus = 1024; diff --git a/src/xen/xend_internal.c b/src/xen/xend_internal.c index 62ce930..21d99e3 100644 --- a/src/xen/xend_internal.c +++ b/src/xen/xend_internal.c @@ -507,7 +507,7 @@ xend_op_ext(virConnectPtr xend, const char *path, const char *key, va_list ap) return -1; content = virBufferContentAndReset(&buf); - VIR_DEBUG("xend op: %s\n", content); + VIR_DEBUG("xend op: %s", content); ret = http2unix(xend_post(xend, path, content)); VIR_FREE(content); diff --git a/tests/virhostdevtest.c b/tests/virhostdevtest.c index 065b825..faebdd4 100644 --- a/tests/virhostdevtest.c +++ b/tests/virhostdevtest.c @@ -169,14 +169,14 @@ testVirHostdevPreparePCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); /* Test normal functionality */ - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, NULL, 0, 0) < 0) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); /* Test unmanaged hostdevs */ - VIR_DEBUG("Test >=1 unmanaged hostdevs\n"); + VIR_DEBUG("Test >=1 unmanaged hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, hostdevs, nhostdevs, 0) < 0) goto cleanup; @@ -186,21 +186,21 @@ testVirHostdevPreparePCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) /* Test conflict */ count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); - VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, &hostdevs[0], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); CHECK_LIST_COUNT(mgr->inactivePCIHostdevs, count2); - VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, "test_domain1", uuid, &hostdevs[1], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); CHECK_LIST_COUNT(mgr->inactivePCIHostdevs, count2); - VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, "test_driver1", dom_name, uuid, &hostdevs[2], 1, 0)) goto cleanup; @@ -223,7 +223,7 @@ testVirHostdevReAttachPCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) for (i = 0; i < nhostdevs; i++) { if (hostdevs[i]->managed != false) { - VIR_DEBUG("invalid test\n"); + VIR_DEBUG("invalid test"); return -1; } } @@ -231,11 +231,11 @@ testVirHostdevReAttachPCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, NULL, 0, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 unmanaged hostdevs\n"); + VIR_DEBUG("Test >=1 unmanaged hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, hostdevs, nhostdevs, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1 - 3); @@ -261,7 +261,7 @@ testVirHostdevPreparePCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); /* Test normal functionality */ - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, hostdevs, nhostdevs, 0) < 0) goto cleanup; @@ -269,19 +269,19 @@ testVirHostdevPreparePCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) /* Test conflict */ count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, &hostdevs[0], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, "test_domain1", uuid, &hostdevs[1], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, "test_driver1", dom_name, uuid, &hostdevs[2], 1, 0)) goto cleanup; @@ -303,18 +303,18 @@ testVirHostdevReAttachPCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) for (i = 0; i < nhostdevs; i++) { if (hostdevs[i]->managed != true) { - VIR_DEBUG("invalid test\n"); + VIR_DEBUG("invalid test"); return -1; } } count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, NULL, 0, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, hostdevs, nhostdevs, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1 - 3); @@ -392,13 +392,13 @@ testVirHostdevUpdateActivePCIHostdevs(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); if (virHostdevUpdateActivePCIDevices(mgr, NULL, 0, drv_name, dom_name) < 0) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); if (virHostdevUpdateActivePCIDevices(mgr, hostdevs, nhostdevs, drv_name, dom_name) < 0) goto cleanup; @@ -432,7 +432,7 @@ mymain(void) # define DO_TEST(fnc) \ do { \ - VIR_DEBUG("\nTesting: %s", #fnc); \ + VIR_DEBUG("Testing: %s", #fnc); \ if (virtTestRun(#fnc, fnc, NULL) < 0) \ ret = -1; \ } while (0) diff --git a/tests/virnetsockettest.c b/tests/virnetsockettest.c index ce9eeab..5786870 100644 --- a/tests/virnetsockettest.c +++ b/tests/virnetsockettest.c @@ -105,7 +105,7 @@ checkProtocols(bool *hasIPv4, bool *hasIPv6, break; } - VIR_DEBUG("Choose port %d\n", *freePort); + VIR_DEBUG("Choose port %d", *freePort); ret = 0; diff --git a/tests/virtimetest.c b/tests/virtimetest.c index 9ab38ae..49040d3 100644 --- a/tests/virtimetest.c +++ b/tests/virtimetest.c @@ -92,7 +92,7 @@ testTimeLocalOffset(const void *args) return -1; if (data->offset != actual) { - VIR_DEBUG("Expect Offset %ld got %ld\n", + VIR_DEBUG("Expect Offset %ld got %ld", data->offset, actual); return -1; } -- 2.6.2

2 3

[libvirt] [libvirt-test-api][PATCH] Add config flag when calling setVcpusFlags with maximum flag
by Hongming Zhang 29 Oct '15

29 Oct '15

Flag 'VIR_DOMAIN_AFFECT_CONFIG' is required by flag 'VIR_DOMAIN_VCPU_MAXIMUM' modified: set_vcpus_config.py --- repos/setVcpus/set_vcpus_config.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/repos/setVcpus/set_vcpus_config.py b/repos/setVcpus/set_vcpus_config.py index 3bb3984..3ef0992 100644 --- a/repos/setVcpus/set_vcpus_config.py +++ b/repos/setVcpus/set_vcpus_config.py @@ -80,7 +80,7 @@ def set_vcpus_config(params): return 1 if maxvcpu: - flags = libvirt.VIR_DOMAIN_VCPU_MAXIMUM + flags = libvirt.VIR_DOMAIN_VCPU_MAXIMUM|libvirt.VIR_DOMAIN_AFFECT_CONFIG logger.info("the given max vcpu number is %s" % maxvcpu) logger.info("set domain maximum vcpu as %s with flag: %s" % (maxvcpu, flags)) -- 1.9.3

2 1

[libvirt] [libvirt-test-api][PATCH 0/2]Fix issues in the method of getting vcpu thread id of vcpupin_live and cpu_affinity
by Hongming Zhang 29 Oct '15

29 Oct '15

Hongming Zhang (2): Fix the issues in checking method of vcpupin_live Modify the checking method via passing the vcpu thread id repos/domain/cpu_affinity.py | 16 +++++++++------- repos/setVcpus/vcpupin_live.py | 36 ++++++++++++++++++++---------------- 2 files changed, 29 insertions(+), 23 deletions(-) -- 1.9.3

2 3

[libvirt] [PATCH] Fix virNetDevWaitDadFinish stub
by Roman Bogorodskiy 29 Oct '15

29 Oct '15

Build on non-Linux fails because the virNetDevWaitDadFinish() stub has unused parameters. Fix by adding appropriate ATTRIBUTE_UNUSED for these parameters. Pushing under build-breaker rule. --- src/util/virnetdev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index b84437e..9789e93 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -1527,7 +1527,8 @@ int virNetDevClearIPAddress(const char *ifname, /* return after DAD finishes for all known IPv6 addresses or an error */ int -virNetDevWaitDadFinish(virSocketAddrPtr *addrs, size_t count) +virNetDevWaitDadFinish(virSocketAddrPtr *addrs ATTRIBUTE_UNUSED, + size_t count ATTRIBUTE_UNUSED) { virReportSystemError(ENOSYS, "%s", _("Unable to wait for IPv6 DAD on this platform")); -- 2.4.6

1 0

[libvirt] [PATCH] util: remove unnecessary code to avoid error overwrite
by Luyao Huang 29 Oct '15

29 Oct '15

virCgroupRemove use VIR_ERROR to log the error and won't change the virLastErr in the thread, so no need do this. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/util/vircgroup.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index 0379c2e..a120a15 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -1670,13 +1670,8 @@ virCgroupNewMachineSystemd(const char *name, } if (virCgroupAddTask(*group, pidleader) < 0) { - virErrorPtr saved = virSaveLastError(); virCgroupRemove(*group); virCgroupFree(group); - if (saved) { - virSetError(saved); - virFreeError(saved); - } } ret = 0; @@ -1728,13 +1723,8 @@ virCgroupNewMachineManual(const char *name, goto cleanup; if (virCgroupAddTask(*group, pidleader) < 0) { - virErrorPtr saved = virSaveLastError(); virCgroupRemove(*group); virCgroupFree(group); - if (saved) { - virSetError(saved); - virFreeError(saved); - } } done: -- 1.8.3.1

3 2

[libvirt] [PATCH] qemu: fix wrong nodeset info when numatune placement is auto
by Luyao Huang 29 Oct '15

29 Oct '15

https://bugzilla.redhat.com/show_bug.cgi?id=1270715 Since commit 9deb96f remove check the nodeset from cgroup for a running vm, the numatune nodeset reporting for those guest numatune placement is auto is not right. Pass the autonodeset to virDomainNumatuneFormatNodeset() to fix this. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_driver.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 8cd5ee3..8ca55dd 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -10177,6 +10177,7 @@ qemuDomainGetNumaParameters(virDomainPtr dom, size_t i; virDomainObjPtr vm = NULL; virDomainNumatuneMemMode tmpmode = VIR_DOMAIN_NUMATUNE_MEM_STRICT; + qemuDomainObjPrivatePtr priv; char *nodeset = NULL; int ret = -1; virDomainDefPtr def = NULL; @@ -10187,6 +10188,7 @@ qemuDomainGetNumaParameters(virDomainPtr dom, if (!(vm = qemuDomObjFromDomain(dom))) return -1; + priv = vm->privateData; if (virDomainGetNumaParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; @@ -10214,7 +10216,8 @@ qemuDomainGetNumaParameters(virDomainPtr dom, break; case 1: /* fill numa nodeset here */ - nodeset = virDomainNumatuneFormatNodeset(def->numa, NULL, -1); + nodeset = virDomainNumatuneFormatNodeset(def->numa, + priv->autoNodeset, -1); if (!nodeset || virTypedParameterAssign(param, VIR_DOMAIN_NUMA_NODESET, VIR_TYPED_PARAM_STRING, nodeset) < 0) -- 1.8.3.1

3 2

[libvirt] [PATCHv4 0/2] Added waiting for DAD to finish for bridge address.
by Maxim Perevedentsev 29 Oct '15

29 Oct '15

This is a fix for commit db488c79173b240459c7754f38c3c6af9b432970 dnsmasq main process which is relied on when waiting for DAD to complete exits without actually waiting for DAD. This is dnsmasq daemon's task. It seems to be a race that DAD finished before dnsmasq main process exited. The above commit needs the execution to block until DAD finishes for bridge IPv6 address because then it closes dummy tap device. Thus we need to ensure this ourselves. So we periodically poll the kernel using netlink and check whether there are any IPv6 addresses assigned to bridge which have 'tentative' state. After DAD is finished, execution continues. I guess that is what dnsmasq was assumed to do. We use netlink to dump information about existing IPv6 addresses. Netlink's response is a multi-part message. Unfortunately, the current implementation of virNetlink treats such messages as faulty and throws an error. So the patch 2/2 adds multi-part nelink response support. Update v2: fixed syntax. Update v3: moved to virnetdev. Update v4: added DAD timeout, minor fixes. Maxim Perevedentsev (2): network: added waiting for DAD to finish for bridge address. netlink: add support for multi-part netlink messages. src/libvirt_private.syms | 1 + src/network/bridge_driver.c | 34 +++++++++++++- src/util/virnetdev.c | 110 ++++++++++++++++++++++++++++++++++++++++++++ src/util/virnetdev.h | 2 + src/util/virnetlink.c | 4 +- 5 files changed, 149 insertions(+), 2 deletions(-) -- 1.8.3.1

2 4

[libvirt] [libvirt-glib] gobject: Add wrapper virDomainSetTime()
by Zeeshan Ali (Khattak) 28 Oct '15

28 Oct '15

--- libvirt-gobject/libvirt-gobject-domain.c | 126 +++++++++++++++++++++++++++++++ libvirt-gobject/libvirt-gobject-domain.h | 25 ++++++ libvirt-gobject/libvirt-gobject.sym | 9 +++ 3 files changed, 160 insertions(+) diff --git a/libvirt-gobject/libvirt-gobject-domain.c b/libvirt-gobject/libvirt-gobject-domain.c index 34eb7ca..26d0cba 100644 --- a/libvirt-gobject/libvirt-gobject-domain.c +++ b/libvirt-gobject/libvirt-gobject-domain.c @@ -1886,3 +1886,129 @@ gboolean gvir_domain_get_has_current_snapshot(GVirDomain *dom, return TRUE; } + +/** + * gvir_domain_set_time: + * @dom: the domain + * @seconds: The seconds since Epoch in UTC. + * @nseconds: The microseconds. + * @flags: bitwise-OR of #GVirDomainSetTimeFlags. + * + * This function tries to set guest time to the given value. The time to set + * (@seconds and @nseconds) should be in seconds relative to the Epoch of + * 1970-01-01 00:00:00 in UTC. + * + * If you pass #GVIR_DOMAIN_TIME_SYNC as @flags, the time is reset using + * domain's RTC and @seconds and @nseconds arguments are ignored. + * + * Please note that some hypervisors may require guest agent to be configured + * and running in order for this function to work. + */ +gboolean gvir_domain_set_time(GVirDomain *dom, + gint64 seconds, + guint nseconds, + guint flags, + GError **err) +{ + GVirDomainPrivate *priv; + int ret; + + g_return_val_if_fail(GVIR_IS_DOMAIN(dom), FALSE); + g_return_val_if_fail(err == NULL || *err == NULL, FALSE); + g_return_val_if_fail(flags == 0 || flags & GVIR_DOMAIN_TIME_SYNC, FALSE); + + priv = dom->priv; + ret = virDomainSetTime(priv->handle, seconds, nseconds, flags); + if (ret < 0) { + gvir_set_error_literal(err, GVIR_DOMAIN_ERROR, + 0, + "Unable to set domain time"); + return FALSE; + } + + return TRUE; +} + +typedef struct { + gint64 seconds; + guint nseconds; + guint flags; +} DomainSetTimeData; + +static void domain_set_time_data_free(DomainSetTimeData *data) +{ + g_slice_free(DomainSetTimeData, data); +} + +static void +gvir_domain_set_time_helper(GTask *task, + gpointer object, + gpointer task_data, + GCancellable *cancellable G_GNUC_UNUSED) +{ + GVirDomain *dom = GVIR_DOMAIN(object); + DomainSetTimeData *data = (DomainSetTimeData *) task_data; + GError *err = NULL; + + if (!gvir_domain_set_time(dom, + data->seconds, + data->nseconds, + data->flags, + &err)) + g_task_return_error(task, err); + else + g_task_return_boolean(task, TRUE); +} + +/** + * gvir_domain_set_time_async: + * @dom: the domain + * @dom: the domain + * @seconds: The seconds since Epoch in UTC. + * @nseconds: The microseconds. + * @flags: bitwise-OR of #GVirDomainSetTimeFlags. + * @cancellable: (allow-none)(transfer none): cancellation object + * @callback: (scope async): completion callback + * @user_data: (closure): opaque data for callback + * + * Asynchronous variant of #gvir_domain_set_time. + */ +void gvir_domain_set_time_async(GVirDomain *dom, + gint64 seconds, + guint nseconds, + guint flags, + GCancellable *cancellable, + GAsyncReadyCallback callback, + gpointer user_data) +{ + GTask *task; + DomainSetTimeData *data; + + g_return_if_fail(GVIR_IS_DOMAIN(dom)); + g_return_if_fail((cancellable == NULL) || G_IS_CANCELLABLE(cancellable)); + + data = g_slice_new0(DomainSetTimeData); + data->seconds = seconds; + data->nseconds = nseconds; + data->flags = flags; + + task = g_task_new (G_OBJECT(dom), + cancellable, + callback, + user_data); + g_task_set_task_data (task, data, (GDestroyNotify)domain_set_time_data_free); + g_task_run_in_thread(task, gvir_domain_set_time_helper); + + g_object_unref(task); +} + +gboolean gvir_domain_set_time_finish(GVirDomain *dom, + GAsyncResult *result, + GError **err) +{ + g_return_val_if_fail(GVIR_IS_DOMAIN(dom), FALSE); + g_return_val_if_fail(g_task_is_valid(result, G_OBJECT(dom)), FALSE); + g_return_val_if_fail(err == NULL || *err == NULL, FALSE); + + return g_task_propagate_boolean(G_TASK(result), err); +} diff --git a/libvirt-gobject/libvirt-gobject-domain.h b/libvirt-gobject/libvirt-gobject-domain.h index 4fe381e..fbf7173 100644 --- a/libvirt-gobject/libvirt-gobject-domain.h +++ b/libvirt-gobject/libvirt-gobject-domain.h @@ -215,6 +215,15 @@ typedef enum { GVIR_DOMAIN_SNAPSHOT_LIST_EXTERNAL = VIR_DOMAIN_SNAPSHOT_LIST_EXTERNAL } GVirDomainSnapshotListFlags; +/** + * GVirDomainSetTimeFlags: + * @GVIR_DOMAIN_TIME_NONE: Re-sync domain time to given values. + * @GVIR_DOMAIN_TIME_SYNC: Re-sync domain time from domain's RTC. + */ +typedef enum { + GVIR_DOMAIN_TIME_NONE = 0, + GVIR_DOMAIN_TIME_SYNC = VIR_DOMAIN_TIME_SYNC, +} GVirDomainSetTimeFlags; typedef struct _GVirDomainInfo GVirDomainInfo; struct _GVirDomainInfo @@ -401,6 +410,22 @@ gboolean gvir_domain_get_has_current_snapshot(GVirDomain *dom, gboolean *has_current_snapshot, GError **error); +gboolean gvir_domain_set_time(GVirDomain *dom, + gint64 seconds, + guint nseconds, + guint flags, + GError **err); +void gvir_domain_set_time_async(GVirDomain *dom, + gint64 seconds, + guint nseconds, + guint flags, + GCancellable *cancellable, + GAsyncReadyCallback callback, + gpointer user_data); +gboolean gvir_domain_set_time_finish(GVirDomain *dom, + GAsyncResult *result, + GError **err); + G_END_DECLS #endif /* __LIBVIRT_GOBJECT_DOMAIN_H__ */ diff --git a/libvirt-gobject/libvirt-gobject.sym b/libvirt-gobject/libvirt-gobject.sym index ca89a45..cbfaa71 100644 --- a/libvirt-gobject/libvirt-gobject.sym +++ b/libvirt-gobject/libvirt-gobject.sym @@ -304,4 +304,13 @@ LIBVIRT_GOBJECT_0.2.2 { gvir_network_get_dhcp_leases; } LIBVIRT_GOBJECT_0.2.1; +LIBVIRT_GOBJECT_0.2.3 { + global: + gvir_domain_set_time_flags_get_type; + + gvir_domain_set_time; + gvir_domain_set_time_async; + gvir_domain_set_time_finish; +} LIBVIRT_GOBJECT_0.2.2; + # .... define new API here using predicted next version number .... -- 2.5.0

2 3

[libvirt] [PATCH] Use correct pci addresses during device-detach
by Nitesh_Konkar 28 Oct '15

28 Oct '15

From: Nitesh_Konkar <niteshkonkar(a)in.ibm.com> The attach-device on live and persistent copies can be done independently. Thus devices can end up having different pci addresses in live and persistent copies. The detach device should try to detach the device from their respective addresses instead of using the same from live/persistent. Signed-off-by:nitkon12@linux.vnet.ibm.com --- src/driver-nodedev.h | 1 + src/qemu/qemu_driver.c | 25 ++++++++++--------------- 2 files changed, 11 insertions(+), 15 deletions(-) diff --git a/src/driver-nodedev.h b/src/driver-nodedev.h index e846612..dea79bd 100644 --- a/src/driver-nodedev.h +++ b/src/driver-nodedev.h @@ -59,6 +59,7 @@ typedef char * typedef char * (*virDrvNodeDeviceGetParent)(virNodeDevicePtr dev); + typedef int (*virDrvNodeDeviceNumOfCaps)(virNodeDevicePtr dev); diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f133b45..6fd58c2 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8708,23 +8708,12 @@ static int qemuDomainDetachDeviceFlags(virDomainPtr dom, const char *xml, !(flags & VIR_DOMAIN_AFFECT_LIVE)) parse_flags |= VIR_DOMAIN_DEF_PARSE_INACTIVE; - dev = dev_copy = virDomainDeviceDefParse(xml, vm->def, + dev_copy = virDomainDeviceDefParse(xml, vm->def, caps, driver->xmlopt, parse_flags); - if (dev == NULL) + if (dev_copy == NULL) goto endjob; - if (flags & VIR_DOMAIN_AFFECT_CONFIG && - flags & VIR_DOMAIN_AFFECT_LIVE) { - /* If we are affecting both CONFIG and LIVE - * create a deep copy of device as adding - * to CONFIG takes one instance. - */ - dev_copy = virDomainDeviceDefCopy(dev, vm->def, caps, driver->xmlopt); - if (!dev_copy) - goto endjob; - } - if (priv->qemuCaps) qemuCaps = virObjectRef(priv->qemuCaps); else if (!(qemuCaps = virQEMUCapsCacheLookup(driver->qemuCapsCache, vm->def->emulator))) @@ -8736,6 +8725,13 @@ static int qemuDomainDetachDeviceFlags(virDomainPtr dom, const char *xml, if (!vmdef) goto endjob; + dev = virDomainDeviceDefParse(xml, vmdef, + caps, driver->xmlopt, + parse_flags); + if (!dev) + goto endjob; + + if (virDomainDefCompatibleDevice(vmdef, dev, VIR_DOMAIN_DEVICE_ACTION_DETACH) < 0) goto endjob; @@ -8777,8 +8773,7 @@ static int qemuDomainDetachDeviceFlags(virDomainPtr dom, const char *xml, cleanup: virObjectUnref(qemuCaps); virDomainDefFree(vmdef); - if (dev != dev_copy) - virDomainDeviceDefFree(dev_copy); + virDomainDeviceDefFree(dev_copy); virDomainDeviceDefFree(dev); virDomainObjEndAPI(&vm); virObjectUnref(caps); -- 2.4.0

2 1

[libvirt] [PATCH] qemu: fix migration flags undefinesource cannot work
by Luyao Huang 28 Oct '15

28 Oct '15

In commit f41be296, we moved vm->persistent check into qemuDomainRemoveInactive, but we didn't change the vm->persistent before call qemuDomainRemoveInactive in some place before and just call it to remove the inactive vm. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_migration.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index b53491a..2abf2ee 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -3912,8 +3912,10 @@ qemuMigrationConfirm(virConnectPtr conn, qemuMigrationJobFinish(driver, vm); if (!virDomainObjIsActive(vm)) { - if (flags & VIR_MIGRATE_UNDEFINE_SOURCE) + if (flags & VIR_MIGRATE_UNDEFINE_SOURCE) { virDomainDeleteConfig(cfg->configDir, cfg->autostartDir, vm); + vm->persistent = 0; + } qemuDomainRemoveInactive(driver, vm); } @@ -5405,8 +5407,10 @@ qemuMigrationPerformJob(virQEMUDriverPtr driver, qemuMigrationJobFinish(driver, vm); if (!virDomainObjIsActive(vm) && ret == 0) { - if (flags & VIR_MIGRATE_UNDEFINE_SOURCE) + if (flags & VIR_MIGRATE_UNDEFINE_SOURCE) { virDomainDeleteConfig(cfg->configDir, cfg->autostartDir, vm); + vm->persistent = 0; + } qemuDomainRemoveInactive(driver, vm); } -- 1.8.3.1

3 4

[libvirt] [PATCH] conf: Add serial target type to ABI stability check
by Luyao Huang 28 Oct '15

28 Oct '15

https://bugzilla.redhat.com/show_bug.cgi?id=1273686 There is no ABI check for serial target type attribute, just add it. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/conf/domain_conf.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 02cc8ad..a31bc05 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -17225,6 +17225,14 @@ static bool virDomainSerialDefCheckABIStability(virDomainChrDefPtr src, virDomainChrDefPtr dst) { + if (src->targetType != dst->targetType) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("Target serial type %s does not match source %s"), + virDomainChrSerialTargetTypeToString(dst->targetType), + virDomainChrSerialTargetTypeToString(src->targetType)); + return false; + } + if (src->target.port != dst->target.port) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Target serial port %d does not match source %d"), -- 1.8.3.1

3 2

[libvirt] [PATCH] rbd: Remove snapshots if the FORCED flag has been provided
by Wido den Hollander 27 Oct '15

27 Oct '15

When a RBD volume has snapshots it can not be removed. This patch introduces a new flag to force volume removal, VIR_STORAGE_VOL_DELETE_FORCED. With this flag any existing snapshots will be removed prior to removing the volume. No existing mechanism in libvirt allowed us to pass such information, so that's why a new flag was introduced. Signed-off-by: Wido den Hollander <wido(a)widodh.nl> --- include/libvirt/libvirt-storage.h | 1 + src/storage/storage_backend_rbd.c | 58 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+) diff --git a/include/libvirt/libvirt-storage.h b/include/libvirt/libvirt-storage.h index 453089e..36ff979 100644 --- a/include/libvirt/libvirt-storage.h +++ b/include/libvirt/libvirt-storage.h @@ -115,6 +115,7 @@ typedef enum { typedef enum { VIR_STORAGE_VOL_DELETE_NORMAL = 0, /* Delete metadata only (fast) */ VIR_STORAGE_VOL_DELETE_ZEROED = 1 << 0, /* Clear all data to zeros (slow) */ + VIR_STORAGE_VOL_DELETE_FORCED = 2, /* Force removal of volume, even if in use */ } virStorageVolDeleteFlags; typedef enum { diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c index ac5085a..ca5e802 100644 --- a/src/storage/storage_backend_rbd.c +++ b/src/storage/storage_backend_rbd.c @@ -428,9 +428,13 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, { int ret = -1; int r = 0; + int max_snaps = 128; + int i, snap_count, protected; virStorageBackendRBDState ptr; ptr.cluster = NULL; ptr.ioctx = NULL; + rbd_snap_info_t *snaps; + rbd_image_t image = NULL; VIR_DEBUG("Removing RBD image %s/%s", pool->def->source.name, vol->name); @@ -443,6 +447,59 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, if (virStorageBackendRBDOpenIoCTX(&ptr, pool) < 0) goto cleanup; + r = rbd_open(ptr.ioctx, vol->name, &image, NULL); + if (r < 0) { + virReportSystemError(-r, _("failed to open the RBD image '%s'"), + vol->name); + goto cleanup; + } + + do { + if (VIR_ALLOC_N(snaps, max_snaps)) + goto cleanup; + + snap_count = rbd_snap_list(image, snaps, &max_snaps); + if (snap_count <= 0) { + VIR_FREE(snaps); + } + } while (snap_count == -ERANGE); + + VIR_DEBUG("Found %d snapshots for volume %s/%s", snap_count, + pool->def->source.name, vol->name); + + if (snap_count > 0 && (flags & VIR_STORAGE_VOL_DELETE_FORCED)) { + for (i = 0; i < snap_count; i++) { + if (rbd_snap_is_protected(image, snaps[i].name, &protected)) + goto cleanup; + + if (protected == 1) { + VIR_DEBUG("Snapshot %s/%s@%s is protected needs to be " + "unprotected", pool->def->source.name, vol->name, + snaps[i].name); + + if (rbd_snap_unprotect(image, snaps[i].name) < 0) + goto cleanup; + } + + VIR_DEBUG("Removing snapshot %s/%s@%s", pool->def->source.name, + vol->name, snaps[i].name); + + if (rbd_snap_remove(image, snaps[i].name) < 0) { + virReportSystemError(-r, _("failed to remove snapshot '%s/%s@%s'"), + pool->def->source.name, vol->name, + snaps[i].name); + goto cleanup; + } + } + + rbd_snap_list_end(snaps); + } + + if (rbd_close(image) < 0) + goto cleanup; + + VIR_DEBUG("Removing volume %s/%s", pool->def->source.name, vol->name); + r = rbd_remove(ptr.ioctx, vol->name); if (r < 0 && (-r) != ENOENT) { virReportSystemError(-r, _("failed to remove volume '%s/%s'"), @@ -453,6 +510,7 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, ret = 0; cleanup: + VIR_FREE(snaps); virStorageBackendRBDCloseRADOSConn(&ptr); return ret; } -- 1.9.1

3 9

[libvirt] Plan for next release
by Daniel Veillard 27 Oct '15

27 Oct '15

I'm a bit late, sorry, but I think we should freeze on Thursday for the next release, and then shoot for middle of next week for the actual release, I hope this last moment warning isn't a problem, thanks, Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] util: implement virProcessGetStartTime on GNU/kFreeBSD
by Pino Toscano 27 Oct '15

27 Oct '15

Use the virProcessGetStartTime implementation also when only the kernel is FreeBSD, such as on GNU/kFreeBSD. --- src/util/virprocess.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/util/virprocess.c b/src/util/virprocess.c index e6b78ef..43118f8 100644 --- a/src/util/virprocess.c +++ b/src/util/virprocess.c @@ -36,11 +36,11 @@ # include <sched.h> #endif -#if defined(__FreeBSD__) || HAVE_BSD_CPU_AFFINITY +#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || HAVE_BSD_CPU_AFFINITY # include <sys/param.h> #endif -#ifdef __FreeBSD__ +#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__) # include <sys/sysctl.h> # include <sys/user.h> #endif @@ -937,7 +937,7 @@ int virProcessGetStartTime(pid_t pid, VIR_FREE(buf); return ret; } -#elif defined(__FreeBSD__) +#elif defined(__FreeBSD__) || defined(__FreeBSD_kernel__) int virProcessGetStartTime(pid_t pid, unsigned long long *timestamp) { -- 2.1.0

4 5

[libvirt] [PATCH] bhyve: implement domainGetOSType
by Roman Bogorodskiy 27 Oct '15

27 Oct '15

--- src/bhyve/bhyve_driver.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/src/bhyve/bhyve_driver.c b/src/bhyve/bhyve_driver.c index d44cf2c..efba0ae 100644 --- a/src/bhyve/bhyve_driver.c +++ b/src/bhyve/bhyve_driver.c @@ -465,6 +465,27 @@ bhyveDomainIsPersistent(virDomainPtr domain) } static char * +bhyveDomainGetOSType(virDomainPtr dom) +{ + virDomainObjPtr vm; + char *ret = NULL; + + if (!(vm = bhyveDomObjFromDomain(dom))) + goto cleanup; + + if (virDomainGetOSTypeEnsureACL(dom->conn, vm->def) < 0) + goto cleanup; + + if (VIR_STRDUP(ret, virDomainOSTypeToString(vm->def->os.type)) < 0) + goto cleanup; + + cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static char * bhyveDomainGetXMLDesc(virDomainPtr domain, unsigned int flags) { virDomainObjPtr vm; @@ -1487,6 +1508,7 @@ static virHypervisorDriver bhyveHypervisorDriver = { .domainDefineXML = bhyveDomainDefineXML, /* 1.2.2 */ .domainDefineXMLFlags = bhyveDomainDefineXMLFlags, /* 1.2.12 */ .domainUndefine = bhyveDomainUndefine, /* 1.2.2 */ + .domainGetOSType = bhyveDomainGetOSType, /* 1.2.21 */ .domainGetXMLDesc = bhyveDomainGetXMLDesc, /* 1.2.2 */ .domainIsActive = bhyveDomainIsActive, /* 1.2.2 */ .domainIsPersistent = bhyveDomainIsPersistent, /* 1.2.2 */ -- 2.6.1

2 2

[libvirt] [PATCH] qemu: Resolve agent deadlock
by John Ferlan 27 Oct '15

27 Oct '15

If we have a shutdown of a VM by a qemu agent while an agent EOF occurs at relatively the same time, it's possible that a deadlock occurs depending on what happens first. Assume qemuProcessHandleAgentEOF runs first, with the vm->lock it clears priv->agent, then clears the vm->lock. After some agent call such as qemuDomainShutdownFlags will call qemuDomainObjExitAgent in a worker thread with the agent lock. It will then reference the priv->agent field: hasRefs = virObjectUnref(priv->agent); in order to determine whether there are references. However, since priv->agent is NULL, this returns false, so the subsquent: if (hasRefs) virObjectUnlock(priv->agent); will not unlock the agent. Eventually the EOF thread will attempt to call qemuAgentClose which will try to lock the agent, but cannot since it's still locked. This patch resolves this issue by keeping a copy of the priv->agent address prior to releasing the vm->lock in qemuDomainObjEnterAgent. Then rather than assuming priv->agent is valid, use the copy of the agent pointer for the subsequent agent call and qemuDomainObjExitAgent. Within qemuDomainObjExitAgent, use the agent copy in order to release the agent lock, then get the vm->lock again and decide whether to clear the priv->agent based on whether any references still exist. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Recent upstream patch: http://www.redhat.com/archives/libvir-list/2015-September/msg00971.html attempted to undo commit id '1020a504' which appears to be the wrong approach since it would still have the vm->lock during the qemuAgentClose. src/qemu/qemu_domain.c | 11 +++--- src/qemu/qemu_domain.h | 4 +-- src/qemu/qemu_driver.c | 90 +++++++++++++++++++++++++++++++++----------------- 3 files changed, 67 insertions(+), 38 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 890d8ed..a908df8 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1868,19 +1868,20 @@ qemuDomainObjEnterAgent(virDomainObjPtr obj) * Should be paired with an earlier qemuDomainObjEnterAgent() call */ void -qemuDomainObjExitAgent(virDomainObjPtr obj) +qemuDomainObjExitAgent(virDomainObjPtr obj, + qemuAgentPtr agent) { qemuDomainObjPrivatePtr priv = obj->privateData; bool hasRefs; - hasRefs = virObjectUnref(priv->agent); + hasRefs = virObjectUnref(agent); if (hasRefs) - virObjectUnlock(priv->agent); + virObjectUnlock(agent); virObjectLock(obj); - VIR_DEBUG("Exited agent (agent=%p vm=%p name=%s)", - priv->agent, obj, obj->def->name); + VIR_DEBUG("Exited agent (agent=%p vm=%p name=%s hasRefs=%d)", + agent, obj, obj->def->name, hasRefs); priv->agentStart = 0; if (!hasRefs) diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 64cd7e1..def31a9 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -296,8 +296,8 @@ int qemuDomainObjEnterMonitorAsync(virQEMUDriverPtr driver, void qemuDomainObjEnterAgent(virDomainObjPtr obj) ATTRIBUTE_NONNULL(1); -void qemuDomainObjExitAgent(virDomainObjPtr obj) - ATTRIBUTE_NONNULL(1); +void qemuDomainObjExitAgent(virDomainObjPtr obj, qemuAgentPtr agent) + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2); void qemuDomainObjEnterRemote(virDomainObjPtr obj) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index a2cc002..b8c9ff7 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1994,9 +1994,10 @@ static int qemuDomainShutdownFlags(virDomainPtr dom, unsigned int flags) qemuDomainSetFakeReboot(driver, vm, isReboot); if (useAgent) { + qemuAgentPtr agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentShutdown(priv->agent, agentFlag); - qemuDomainObjExitAgent(vm); + ret = qemuAgentShutdown(agent, agentFlag); + qemuDomainObjExitAgent(vm, agent); } /* If we are not enforced to use just an agent, try ACPI @@ -2087,9 +2088,10 @@ qemuDomainReboot(virDomainPtr dom, unsigned int flags) qemuDomainSetFakeReboot(driver, vm, isReboot); if (useAgent) { + qemuAgentPtr agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentShutdown(priv->agent, agentFlag); - qemuDomainObjExitAgent(vm); + ret = qemuAgentShutdown(agent, agentFlag); + qemuDomainObjExitAgent(vm, agent); } /* If we are not enforced to use just an agent, try ACPI @@ -4864,6 +4866,7 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, qemuAgentCPUInfoPtr cpuinfo = NULL; int ncpuinfo; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; size_t i; virCgroupPtr cgroup_temp = NULL; char *mem_mask = NULL; @@ -4924,6 +4927,7 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, } if (flags & VIR_DOMAIN_VCPU_GUEST) { + if (!qemuDomainAgentAvailable(vm, true)) goto endjob; @@ -4935,19 +4939,20 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, goto endjob; } + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ncpuinfo = qemuAgentGetVCPUs(priv->agent, &cpuinfo); - qemuDomainObjExitAgent(vm); + ncpuinfo = qemuAgentGetVCPUs(agent, &cpuinfo); + qemuDomainObjExitAgent(vm, agent); - if (ncpuinfo < 0) + if (ncpuinfo < 0 || agent != priv->agent) goto endjob; if (qemuAgentUpdateCPUInfo(nvcpus, cpuinfo, ncpuinfo) < 0) goto endjob; qemuDomainObjEnterAgent(vm); - ret = qemuAgentSetVCPUs(priv->agent, cpuinfo, ncpuinfo); - qemuDomainObjExitAgent(vm); + ret = qemuAgentSetVCPUs(agent, cpuinfo, ncpuinfo); + qemuDomainObjExitAgent(vm, agent); if (ret < 0) goto endjob; @@ -5515,6 +5520,8 @@ qemuDomainGetVcpusFlags(virDomainPtr dom, unsigned int flags) goto cleanup; if (flags & VIR_DOMAIN_VCPU_GUEST) { + qemuAgentPtr agent; + if (!virDomainObjIsActive(vm)) { virReportError(VIR_ERR_INVALID_ARG, "%s", _("vCPU count provided by the guest agent can only be " @@ -5534,9 +5541,10 @@ qemuDomainGetVcpusFlags(virDomainPtr dom, unsigned int flags) goto endjob; } + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ncpuinfo = qemuAgentGetVCPUs(priv->agent, &cpuinfo); - qemuDomainObjExitAgent(vm); + ncpuinfo = qemuAgentGetVCPUs(agent, &cpuinfo); + qemuDomainObjExitAgent(vm, agent); endjob: qemuDomainObjEndJob(driver, vm); @@ -13513,14 +13521,16 @@ qemuDomainSnapshotFSFreeze(virQEMUDriverPtr driver ATTRIBUTE_UNUSED, unsigned int nmountpoints) { qemuDomainObjPrivatePtr priv = vm->privateData; + qemuAgentPtr agent; int frozen; if (!qemuDomainAgentAvailable(vm, true)) return -1; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - frozen = qemuAgentFSFreeze(priv->agent, mountpoints, nmountpoints); - qemuDomainObjExitAgent(vm); + frozen = qemuAgentFSFreeze(agent, mountpoints, nmountpoints); + qemuDomainObjExitAgent(vm, agent); return frozen < 0 ? -2 : frozen; } @@ -13532,19 +13542,21 @@ qemuDomainSnapshotFSThaw(virQEMUDriverPtr driver ATTRIBUTE_UNUSED, bool report) { qemuDomainObjPrivatePtr priv = vm->privateData; + qemuAgentPtr agent; int thawed; virErrorPtr err = NULL; if (!qemuDomainAgentAvailable(vm, report)) return -1; + agent = priv->agent; qemuDomainObjEnterAgent(vm); if (!report) err = virSaveLastError(); - thawed = qemuAgentFSThaw(priv->agent); + thawed = qemuAgentFSThaw(agent); if (!report) virSetError(err); - qemuDomainObjExitAgent(vm); + qemuDomainObjExitAgent(vm, agent); virFreeError(err); @@ -18146,6 +18158,7 @@ qemuDomainPMSuspendForDuration(virDomainPtr dom, { virQEMUDriverPtr driver = dom->conn->privateData; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virDomainObjPtr vm; int ret = -1; @@ -18218,9 +18231,10 @@ qemuDomainPMSuspendForDuration(virDomainPtr dom, goto endjob; } + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentSuspend(priv->agent, target); - qemuDomainObjExitAgent(vm); + ret = qemuAgentSuspend(agent, target); + qemuDomainObjExitAgent(vm, agent); endjob: qemuDomainObjEndJob(driver, vm); @@ -18309,6 +18323,7 @@ qemuDomainQemuAgentCommand(virDomainPtr domain, int ret = -1; char *result = NULL; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virCheckFlags(0, NULL); @@ -18338,9 +18353,10 @@ qemuDomainQemuAgentCommand(virDomainPtr domain, goto endjob; } + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentArbitraryCommand(priv->agent, cmd, &result, timeout); - qemuDomainObjExitAgent(vm); + ret = qemuAgentArbitraryCommand(agent, cmd, &result, timeout); + qemuDomainObjExitAgent(vm, agent); if (ret < 0) VIR_FREE(result); @@ -18411,6 +18427,7 @@ qemuDomainFSTrim(virDomainPtr dom, virDomainObjPtr vm; int ret = -1; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virCheckFlags(0, -1); @@ -18447,9 +18464,10 @@ qemuDomainFSTrim(virDomainPtr dom, goto endjob; } + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentFSTrim(priv->agent, minimum); - qemuDomainObjExitAgent(vm); + ret = qemuAgentFSTrim(agent, minimum); + qemuDomainObjExitAgent(vm, agent); endjob: qemuDomainObjEndJob(driver, vm); @@ -18600,6 +18618,7 @@ qemuDomainGetTime(virDomainPtr dom, virQEMUDriverPtr driver = dom->conn->privateData; virDomainObjPtr vm = NULL; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; int ret = -1; int rv; @@ -18625,9 +18644,10 @@ qemuDomainGetTime(virDomainPtr dom, if (!qemuDomainAgentAvailable(vm, true)) goto endjob; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - rv = qemuAgentGetTime(priv->agent, seconds, nseconds); - qemuDomainObjExitAgent(vm); + rv = qemuAgentGetTime(agent, seconds, nseconds); + qemuDomainObjExitAgent(vm, agent); if (rv < 0) goto endjob; @@ -18650,6 +18670,7 @@ qemuDomainSetTime(virDomainPtr dom, { virQEMUDriverPtr driver = dom->conn->privateData; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virDomainObjPtr vm; bool rtcSync = flags & VIR_DOMAIN_TIME_SYNC; int ret = -1; @@ -18689,9 +18710,10 @@ qemuDomainSetTime(virDomainPtr dom, if (!qemuDomainAgentAvailable(vm, true)) goto endjob; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - rv = qemuAgentSetTime(priv->agent, seconds, nseconds, rtcSync); - qemuDomainObjExitAgent(vm); + rv = qemuAgentSetTime(agent, seconds, nseconds, rtcSync); + qemuDomainObjExitAgent(vm, agent); if (rv < 0) goto endjob; @@ -19647,6 +19669,7 @@ qemuDomainGetFSInfo(virDomainPtr dom, { virQEMUDriverPtr driver = dom->conn->privateData; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virDomainObjPtr vm; int ret = -1; @@ -19672,9 +19695,10 @@ qemuDomainGetFSInfo(virDomainPtr dom, if (!qemuDomainAgentAvailable(vm, true)) goto endjob; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentGetFSInfo(priv->agent, info, vm->def); - qemuDomainObjExitAgent(vm); + ret = qemuAgentGetFSInfo(agent, info, vm->def); + qemuDomainObjExitAgent(vm, agent); endjob: qemuDomainObjEndJob(driver, vm); @@ -19692,6 +19716,7 @@ qemuDomainInterfaceAddresses(virDomainPtr dom, { virQEMUDriverPtr driver = dom->conn->privateData; qemuDomainObjPrivatePtr priv = NULL; + qemuAgentPtr agent; virDomainObjPtr vm = NULL; int ret = -1; @@ -19723,9 +19748,10 @@ qemuDomainInterfaceAddresses(virDomainPtr dom, if (!qemuDomainAgentAvailable(vm, true)) goto endjob; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - ret = qemuAgentGetInterfaces(priv->agent, ifaces); - qemuDomainObjExitAgent(vm); + ret = qemuAgentGetInterfaces(agent, ifaces); + qemuDomainObjExitAgent(vm, agent); endjob: qemuDomainObjEndJob(driver, vm); @@ -19850,6 +19876,7 @@ qemuDomainSetUserPassword(virDomainPtr dom, { virQEMUDriverPtr driver = dom->conn->privateData; qemuDomainObjPrivatePtr priv; + qemuAgentPtr agent; virDomainObjPtr vm; int ret = -1; int rv; @@ -19876,10 +19903,11 @@ qemuDomainSetUserPassword(virDomainPtr dom, if (!qemuDomainAgentAvailable(vm, true)) goto endjob; + agent = priv->agent; qemuDomainObjEnterAgent(vm); - rv = qemuAgentSetUserPassword(priv->agent, user, password, + rv = qemuAgentSetUserPassword(agent, user, password, flags & VIR_DOMAIN_PASSWORD_ENCRYPTED); - qemuDomainObjExitAgent(vm); + qemuDomainObjExitAgent(vm, agent); if (rv < 0) goto endjob; -- 2.1.0

2 4

[libvirt] [PATCH 0/4] improve virsh attache-interface
by Pavel Hrdina 27 Oct '15

27 Oct '15

Pavel Hrdina (4): virsh-domain: use correct base for virStrToLong_ui virsh-nodedev: makes struct and functions for NodeDevice list available virsh-domain: update attach-interface to support type=hostdev virsh.pod: update and improve a attach-interface section tools/virsh-domain.c | 120 ++++++++++++++++++++++++++++++++++++++++++-------- tools/virsh-nodedev.c | 16 +++---- tools/virsh-nodedev.h | 11 +++++ tools/virsh.pod | 85 +++++++++++++++++++++-------------- 4 files changed, 170 insertions(+), 62 deletions(-) -- 2.6.2

3 13

[libvirt] [PATCH] locking: Add io_timeout to sanlock
by Michal Privoznik 27 Oct '15

27 Oct '15

https://bugzilla.redhat.com/show_bug.cgi?id=1251190 So, if domain loses access to storage, sanlock tries to kill it after some timeout. So far, the default is 80 seconds. But for some scenarios this might not be enough. We should allow users to adjust the timeout according to their needs. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/locking/libvirt_sanlock.aug | 1 + src/locking/lock_driver_sanlock.c | 8 +++++++- src/locking/sanlock.conf | 7 +++++++ src/locking/test_libvirt_sanlock.aug.in | 1 + 4 files changed, 16 insertions(+), 1 deletion(-) diff --git a/src/locking/libvirt_sanlock.aug b/src/locking/libvirt_sanlock.aug index a78a444..8843590 100644 --- a/src/locking/libvirt_sanlock.aug +++ b/src/locking/libvirt_sanlock.aug @@ -22,6 +22,7 @@ module Libvirt_sanlock = | int_entry "host_id" | bool_entry "require_lease_for_disks" | bool_entry "ignore_readonly_and_shared_disks" + | int_entry "io_timeout" | str_entry "user" | str_entry "group" let comment = [ label "#comment" . del /#[ \t]*/ "# " . store /([^ \t\n][^\n]*)?/ . del /\n/ "\n" ] diff --git a/src/locking/lock_driver_sanlock.c b/src/locking/lock_driver_sanlock.c index e052875..f1dbbaf 100644 --- a/src/locking/lock_driver_sanlock.c +++ b/src/locking/lock_driver_sanlock.c @@ -73,6 +73,7 @@ struct _virLockManagerSanlockDriver { int hostID; bool autoDiskLease; char *autoDiskLeasePath; + unsigned int io_timeout; /* under which permissions does sanlock run */ uid_t user; @@ -151,6 +152,10 @@ static int virLockManagerSanlockLoadConfig(const char *configFile) else driver->requireLeaseForDisks = !driver->autoDiskLease; + p = virConfGetValue(conf, "io_timeout"); + CHECK_TYPE("io_timeout", VIR_CONF_ULONG); + if (p) driver->io_timeout = p->l; + p = virConfGetValue(conf, "user"); CHECK_TYPE("user", VIR_CONF_STRING); if (p) { @@ -338,7 +343,7 @@ static int virLockManagerSanlockSetupLockspace(void) * or we can fallback to polling. */ retry: - if ((rv = sanlock_add_lockspace(&ls, 0)) < 0) { + if ((rv = sanlock_add_lockspace_timeout(&ls, 0, driver->io_timeout)) < 0) { if (-rv == EINPROGRESS && --retries) { #ifdef HAVE_SANLOCK_INQ_LOCKSPACE /* we have this function which blocks until lockspace change the @@ -404,6 +409,7 @@ static int virLockManagerSanlockInit(unsigned int version, driver->requireLeaseForDisks = true; driver->hostID = 0; driver->autoDiskLease = false; + driver->io_timeout = 0; driver->user = (uid_t) -1; driver->group = (gid_t) -1; if (VIR_STRDUP(driver->autoDiskLeasePath, LOCALSTATEDIR "/lib/libvirt/sanlock") < 0) { diff --git a/src/locking/sanlock.conf b/src/locking/sanlock.conf index e5566ef..3a1a51c 100644 --- a/src/locking/sanlock.conf +++ b/src/locking/sanlock.conf @@ -54,6 +54,13 @@ #require_lease_for_disks = 1 # +# Sanlock is able to kill qemu processes on IO timeout. By its internal +# implementation, the current default is 80 seconds. If you need to adjust +# the value change the following variable. Value of zero means use the +# default sanlock timeout. +#io_timeout = 0 + +# # The combination of user and group under which the sanlock # daemon runs. Libvirt will chown created files (like # content of disk_lease_dir) to make sure sanlock daemon can diff --git a/src/locking/test_libvirt_sanlock.aug.in b/src/locking/test_libvirt_sanlock.aug.in index ef98ea6..7f66f81 100644 --- a/src/locking/test_libvirt_sanlock.aug.in +++ b/src/locking/test_libvirt_sanlock.aug.in @@ -6,5 +6,6 @@ module Test_libvirt_sanlock = { "disk_lease_dir" = "/var/lib/libvirt/sanlock" } { "host_id" = "1" } { "require_lease_for_disks" = "1" } +{ "io_timeout" = "0" } { "user" = "root" } { "group" = "root" } -- 2.4.10

2 2

[libvirt] [PATCH 0/4] HACKING: Improve handling
by Andrea Bolognani 27 Oct '15

27 Oct '15

Generate the HACKING file at dist time like we're already doing for NEWS, AUTHORS and ChangeLog. The file shouldn't be tracked by git as it's not a source file but rather a generated one; because of this, it should also be generated in $(builddir) as opposed to $(top_srcdir). Cheers. Andrea Bolognani (4): HACKING: Generate inside the build directory HACKING: Remove generated copy HACKING: Add to ignore file HACKING: Include in EXTRA_DIST .gitignore | 1 + HACKING | 1008 ----------------------------------------------------------- Makefile.am | 3 +- cfg.mk | 3 +- 4 files changed, 4 insertions(+), 1011 deletions(-) delete mode 100644 HACKING -- 2.4.3

3 7