January 2011 - Devel - libvirt List Archives

Re: [libvirt] virsh + ssh
by Daniel Huhardeaux 05 Jan '11

05 Jan '11

Le 04/01/2011 18:24, Eric Blake a écrit : > On 01/04/2011 10:13 AM, Daniel Huhardeaux wrote: >> My problem was that I wanted to do it with virt-manager which only >> connect using URI qemu+ssh://root@remote/system :-( > > virt-manager lets you add a new connection and specify the username. On > virt-manager-0.8.5-1.fc14 (fedora 14 box), I just tried File->add > connection; hypervisor qemu/kvm, check the box for remote host, method > ssh, and then you can specify both username and host, at which point it > displays a generated URI of qemu+ssh://user@remote/system. [...] Got it: virt-manager from Debian unstable was not uptodate due to Stable freeze. Just grab from experimental and it's like you describe it. Sorry for noise. -- Daniel

1 0

Re: [libvirt] virsh + ssh
by Daniel Huhardeaux 05 Jan '11

05 Jan '11

Eric sorry for private mail, I tought libvirt-list was the "To". Le 04/01/2011 18:13, Daniel Huhardeaux a écrit : > Hi all > > Le 03/01/2011 22:32, Eric Blake a écrit : >> On 01/01/2011 03:38 AM, Daniel Huhardeaux wrote: >>> Hi, >>> >>> connecting to a remote server using qemu+ssh is always done with root >>> user. The problem we face is that we don't allow root connection for >>> ssh. >> >> I haven't tried, but it seems to me like you should be able to do >> qemu+ssh://user@remote/system in order to connect as user instead of >> root. > > Following Justins link -which says the same that you- it works. > > My problem was that I wanted to do it with virt-manager which only > connect using URI qemu+ssh://root@remote/system :-( > > I face a strange behavior: I added my user in libvirt group and modify > libvirtd.conf to start with group libvirt. I restart libvirt-bin and: > > virsh -c qemu+ssh:///system > dh@localhost's password: > Welcome to virsh, the virtualization interactive terminal. > > Type: 'help' for help with commands > 'quit' to quit > > virsh # list > Id Name State > ---------------------------------- > 1 XPHome running > > virsh # > > Now without connecting through ssh: > > virsh > Welcome to virsh, the virtualization interactive terminal. > > Type: 'help' for help with commands > 'quit' to quit > > virsh # list > Id Name State > ---------------------------------- > > virsh # > > So using virsh I don't see VM's! If I do 'sudo virsh' it's OK. What is > wrong in my setup? > > Thanks for your help I got it: when you connect as an user with virsh it does't automatically connect to local system as it does as root user: virsh -c qemu:///system is the right command. Thanks for your time. PS: if someone knows how to connect virt-manager with ssh as a normal user, would appreciate to share :-) -- Daniel

1 0

Re: [libvirt] virsh + ssh
by Eric Blake 05 Jan '11

05 Jan '11

[forwarding back to the list, in case this helps anyone else] On 01/04/2011 10:24 AM, Eric Blake wrote: > On 01/04/2011 10:13 AM, Daniel Huhardeaux wrote: >> My problem was that I wanted to do it with virt-manager which only >> connect using URI qemu+ssh://root@remote/system :-( > > virt-manager lets you add a new connection and specify the username. On > virt-manager-0.8.5-1.fc14 (fedora 14 box), I just tried File->add > connection; hypervisor qemu/kvm, check the box for remote host, method > ssh, and then you can specify both username and host, at which point it > displays a generated URI of qemu+ssh://user@remote/system. > >> >> I face a strange behavior: I added my user in libvirt group and modify >> libvirtd.conf to start with group libvirt. I restart libvirt-bin and: >> >> virsh -c qemu+ssh:///system >> dh@localhost's password: > > so you are authenticating as yourself and not as root. > >> virsh >> Welcome to virsh, the virtualization interactive terminal. > > And here, no authentication was even attempted (which means you really > connected to qemu:///session). Try instead to use: > > virsh -c qemu:///system > > to see the difference (system and session maintain independent lists of > running domains, which is why your use of virsh without -c saw no > domains because it connected to session instead of system). > -- Eric Blake eblake(a)redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org

1 0

[libvirt] [PATCH 00/13] IPv6 support for virtual networks using bridge driver
by Laine Stump 05 Jan '11

05 Jan '11

Finally! Most of this patchset is setup for patch 09/13, which updates the network XML parser to support IPv6, and 12/13, which turns on IPv6 in the bridge driver. In order to have each patch individually pass make check and (otherwise function properly in case someone is doing a bisect), there is a bit of extra code churn (lines that are changed in one patch, only to be changed again in a later patch); I tried to minimize this as much as possible. For for IPv6 to work correctly, target *and build* systems will now need to have ip6tables and radvd available. The way I added ip6tables into autoconfigure.ac is identical to existing iptables, and the way I added radvd is identical to dnsmasq. Unfortunately this doesn't communicate the requirement downstream in a programmatic fashion, so I guess we need to make sure that this is adequately reported in the next set of release notes. There is a remaining problem, outlined in PATCH 13/13, which I would like some input on - there is a race between radvd writing its pidfile and libvirtd reading it, and a couple possible ways to fix it, neither of which is optimal. If anyone has a different idea, I'd like to hear it (I'm leaning towards option 2 right now - it creates an extra file, but behavior is 100% correct).

4 35

[libvirt] Pre-release tarball of 0.8.7 to test
by Daniel Veillard 05 Jan '11

05 Jan '11

I have uploaded a pre-release tarball as ftp://libvirt.org/libvirt/libvirt-pre0.8.7.tar.gz This seems to look okay for me with my limited current testing abilities, so I suggest people give it a try and if things looks okay we can push the real release, maybe early on Tuesday, thanks in advance for giving it a try, 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/

4 10

[libvirt] [PATCH v2] Support for qemu aio drive option
by Matthias Dahl 05 Jan '11

05 Jan '11

Revised patch against libvirt 0.7.6 to support qemu's aio option. qemu allows the user to choose what io storage api should be used, either the default (threads) or native (linux aio) which in the latter case can result in better performance. This patch exposes this functionality through libvirt. Thanks a lot to Eric Blake and Matthias Bolte for their comments. --- docs/formatdomain.html.in | 4 +++- docs/schemas/domain.rng | 13 +++++++++++-- src/conf/domain_conf.c | 23 +++++++++++++++++++++++ src/conf/domain_conf.h | 10 ++++++++++ src/qemu/qemu_conf.c | 14 ++++++++++++++ src/qemu/qemu_conf.h | 2 ++ 6 files changed, 63 insertions(+), 3 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index ce49f7d..ec85ef3 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -478,7 +478,9 @@ attribute is the primary backend driver name, while the optional <code>type</code> attribute provides the sub-type. The optional <code>cache</code> attribute controls the cache mechanism, possible values are "default", "none", - "writethrough" and "writeback". <span class="since">Since 0.1.8</span> + "writethrough" and "writeback". Specific to KVM guests is the optional <code>aio</code> + attribute which controls what storage api is used for io operations, its possible + values are "threads" and "native". <span class="since">Since 0.1.8; "aio" attribute since 0.7.7</span> </dd> <dt><code>encryption</code></dt> <dd>If present, specifies how the volume is encrypted. See diff --git a/docs/schemas/domain.rng b/docs/schemas/domain.rng index bb6d00d..5d2dafe 100644 --- a/docs/schemas/domain.rng +++ b/docs/schemas/domain.rng @@ -499,8 +499,9 @@ <ref name="driverCache"/> </group> </choice> - <empty/> - </element> + <optional> + <ref name="driverAIO"/> + </optional> </define> <define name="driverFormat"> <attribute name="name"> @@ -521,6 +522,14 @@ </choice> </attribute> </define> + <define name="driverAIO"> + <attribute name="aio"> + <choice> + <value>threads</value> + <value>native</value> + </choice> + </attribute> + </define> <define name="controller"> <element name="controller"> <optional> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 766993c..8bee7b8 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -120,6 +120,11 @@ VIR_ENUM_IMPL(virDomainDiskCache, VIR_DOMAIN_DISK_CACHE_LAST, "writethrough", "writeback") +VIR_ENUM_IMPL(virDomainDiskAIO, VIR_DOMAIN_DISK_AIO_LAST, + "default", + "native", + "threads") + VIR_ENUM_IMPL(virDomainController, VIR_DOMAIN_CONTROLLER_TYPE_LAST, "ide", "fdc", @@ -1228,6 +1233,7 @@ virDomainDiskDefParseXML(virConnectPtr conn, char *target = NULL; char *bus = NULL; char *cachetag = NULL; + char *aiotag = NULL; char *devaddr = NULL; virStorageEncryptionPtr encryption = NULL; char *serial = NULL; @@ -1293,6 +1299,7 @@ virDomainDiskDefParseXML(virConnectPtr conn, driverName = virXMLPropString(cur, "name"); driverType = virXMLPropString(cur, "type"); cachetag = virXMLPropString(cur, "cache"); + aiotag = virXMLPropString(cur, "aio"); } else if (xmlStrEqual(cur->name, BAD_CAST "readonly")) { def->readonly = 1; } else if (xmlStrEqual(cur->name, BAD_CAST "shareable")) { @@ -1409,6 +1416,13 @@ virDomainDiskDefParseXML(virConnectPtr conn, goto error; } + if (aiotag && + (def->aiomode = virDomainDiskAIOTypeFromString(aiotag)) < 0) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown disk aio mode '%s'"), aiotag); + goto error; + } + if (devaddr) { if (sscanf(devaddr, "%x:%x:%x", &def->info.addr.pci.domain, @@ -1450,6 +1464,7 @@ cleanup: VIR_FREE(driverType); VIR_FREE(driverName); VIR_FREE(cachetag); + VIR_FREE(aiotag); VIR_FREE(devaddr); VIR_FREE(serial); virStorageEncryptionFree(encryption); @@ -4565,6 +4580,7 @@ virDomainDiskDefFormat(virConnectPtr conn, const char *device = virDomainDiskDeviceTypeToString(def->device); const char *bus = virDomainDiskBusTypeToString(def->bus); const char *cachemode = virDomainDiskCacheTypeToString(def->cachemode); + const char *aiomode = virDomainDiskAIOTypeToString(def->aiomode); if (!type) { virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, @@ -4586,6 +4602,11 @@ virDomainDiskDefFormat(virConnectPtr conn, _("unexpected disk cache mode %d"), def->cachemode); return -1; } + if (!aiomode) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("unexpected disk aio mode %d"), def->aiomode); + return -1; + } virBufferVSprintf(buf, " <disk type='%s' device='%s'>\n", @@ -4597,6 +4618,8 @@ virDomainDiskDefFormat(virConnectPtr conn, virBufferVSprintf(buf, " type='%s'", def->driverType); if (def->cachemode) virBufferVSprintf(buf, " cache='%s'", cachemode); + if (def->aiomode) + virBufferVSprintf(buf, " aio='%s'", aiomode); virBufferVSprintf(buf, "/>\n"); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 0b79e88..07805a6 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -141,6 +141,14 @@ enum virDomainDiskCache { VIR_DOMAIN_DISK_CACHE_LAST }; +enum virDomainDiskAIO { + VIR_DOMAIN_DISK_AIO_DEFAULT, + VIR_DOMAIN_DISK_AIO_NATIVE, + VIR_DOMAIN_DISK_AIO_THREADS, + + VIR_DOMAIN_DISK_AIO_LAST +}; + /* Stores the virtual disk configuration */ typedef struct _virDomainDiskDef virDomainDiskDef; typedef virDomainDiskDef *virDomainDiskDefPtr; @@ -154,6 +162,7 @@ struct _virDomainDiskDef { char *driverType; char *serial; int cachemode; + int aiomode; unsigned int readonly : 1; unsigned int shared : 1; virDomainDeviceInfo info; @@ -897,6 +906,7 @@ VIR_ENUM_DECL(virDomainDisk) VIR_ENUM_DECL(virDomainDiskDevice) VIR_ENUM_DECL(virDomainDiskBus) VIR_ENUM_DECL(virDomainDiskCache) +VIR_ENUM_DECL(virDomainDiskAIO) VIR_ENUM_DECL(virDomainController) VIR_ENUM_DECL(virDomainFS) VIR_ENUM_DECL(virDomainNet) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 3d83a8f..fd3a670 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -83,6 +83,12 @@ VIR_ENUM_IMPL(qemuDiskCacheV2, VIR_DOMAIN_DISK_CACHE_LAST, "writethrough", "writeback"); +VIR_ENUM_DECL(qemuDiskAIO) +VIR_ENUM_IMPL(qemuDiskAIO, VIR_DOMAIN_DISK_AIO_LAST, + "default", + "native", + "threads"); + VIR_ENUM_DECL(qemuVideo) VIR_ENUM_IMPL(qemuVideo, VIR_DOMAIN_VIDEO_TYPE_LAST, @@ -1137,6 +1143,8 @@ static unsigned int qemudComputeCmdFlags(const char *help, flags |= QEMUD_CMD_FLAG_DRIVE_CACHE_V2; if (strstr(help, "format=")) flags |= QEMUD_CMD_FLAG_DRIVE_FORMAT; + if (strstr(help, "aio=threads|native")) + flags |= QEMUD_CMD_FLAG_DRIVE_AIO; } if (strstr(help, "-vga") && !strstr(help, "-std-vga")) flags |= QEMUD_CMD_FLAG_VGA; @@ -2340,6 +2348,12 @@ qemuBuildDriveStr(virDomainDiskDefPtr disk, virBufferAddLit(&opt, ",cache=off"); } + if (disk->aiomode && (qemuCmdFlags & QEMUD_CMD_FLAG_DRIVE_AIO)) { + const char * mode = qemuDiskAIOTypeToString(disk->aiomode); + + virBufferVSprintf(&opt, ",aio=%s", mode); + } + if (virBufferError(&opt)) { virReportOOMError(NULL); goto error; diff --git a/src/qemu/qemu_conf.h b/src/qemu/qemu_conf.h index 101f187..780ae6e 100644 --- a/src/qemu/qemu_conf.h +++ b/src/qemu/qemu_conf.h @@ -82,6 +82,8 @@ enum qemud_cmd_flags { QEMUD_CMD_FLAG_SDL = (1 << 27), /* Is the new -sdl arg available */ QEMUD_CMD_FLAG_SMP_TOPOLOGY = (1 << 28), /* Is sockets=s,cores=c,threads=t available for -smp? */ QEMUD_CMD_FLAG_NETDEV = (1 << 29), /* The -netdev flag & netdev_add/remove monitor commands */ + + QEMUD_CMD_FLAG_DRIVE_AIO = (1 << 30), /* Is -drive aio= avail */ }; /* Main driver state */ -- 1.7.0.4

4 14

[libvirt] [PATCH] Support <ipv[46]routing> network parameters
by Paweł Krześniak 05 Jan '11

05 Jan '11

Support for optional parameters <ipv[46]routing> in XML network definition. These parameters allows to set up alternative binary or script to be run during network start. Such customization allows to run additional services for started network (besides default DHCP/DNS/NDP) or modify/pass additional options to dnsmasq and/or radvd. If <ipv[46]routing> parameters point to default /usr/sbin/dnsmasq or /usr/sbin/radvd they are removed from XML network definition. --- docs/schemas/network.rng | 28 +++++++++++++++ src/conf/network_conf.c | 37 ++++++++++++++++++++ src/conf/network_conf.h | 5 +++ src/network/bridge_driver.c | 8 ++-- .../networkxml2xmlin/with-ipv4routing-network.xml | 11 ++++++ .../networkxml2xmlout/with-ipv4routing-network.xml | 11 ++++++ tests/networkxml2xmltest.c | 1 + 7 files changed, 97 insertions(+), 4 deletions(-) diff --git a/docs/schemas/network.rng b/docs/schemas/network.rng index 4252f30..71234a3 100644 --- a/docs/schemas/network.rng +++ b/docs/schemas/network.rng @@ -21,6 +21,16 @@ <element name="uuid"><text/></element> </optional> +  + <optional> + <ref name="ipv4routing"/> + </optional> + +  + <optional> + <ref name="ipv6routing"/> + </optional> +  <optional>  + <define name="ipv4routing"> + <element name="ipv4routing"> + <ref name="absFilePath"/> + </element> + </define> + + <define name="ipv6routing"> + <element name="ipv6routing"> + <ref name="absFilePath"/> + </element> + </define> + + <define name="absFilePath"> + <data type="string"> + <param name="pattern">/[a-zA-Z0-9_\.\+\-\\&"'<>/%]+</param> + </data> + </define> </grammar> diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index 420b94a..b70397f 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -115,6 +115,8 @@ void virNetworkDefFree(virNetworkDefPtr def) VIR_FREE(def->bridge); VIR_FREE(def->forwardDev); VIR_FREE(def->domain); + VIR_FREE(def->ipv4routing); + VIR_FREE(def->ipv6routing); for (ii = 0 ; ii < def->nips && def->ips ; ii++) { virNetworkIpDefClear(&def->ips[ii]); @@ -612,6 +614,36 @@ virNetworkDefParseXML(xmlXPathContextPtr ctxt) VIR_FREE(tmp); } + /* Extract path to ipv4routing binary */ + def->ipv4routing = virXPathString("string(./ipv4routing[1])", ctxt); + if (def->ipv4routing) { + if (access(def->ipv4routing, X_OK) < 0) { + virNetworkReportError(VIR_ERR_INTERNAL_ERROR, _("Cannot find IPv4Routing binary %s"), def->ipv4routing); + goto error; + } + } else { + def->ipv4routing = strdup(DNSMASQ); + if (!def->ipv4routing) { + virReportOOMError(); + goto error; + } + } + + /* Extract path to ipv6routing binary */ + def->ipv6routing = virXPathString("string(./ipv6routing[1])", ctxt); + if (def->ipv6routing) { + if (access(def->ipv6routing, X_OK) < 0) { + virNetworkReportError(VIR_ERR_INTERNAL_ERROR, _("Cannot find IPv6Routing binary %s"), def->ipv6routing); + goto error; + } + } else { + def->ipv6routing = strdup(RADVD); + if (!def->ipv6routing) { + virReportOOMError(); + goto error; + } + } + /* Parse network domain information */ def->domain = virXPathString("string(./domain[1]/@name)", ctxt); @@ -831,6 +863,11 @@ char *virNetworkDefFormat(const virNetworkDefPtr def) virUUIDFormat(uuid, uuidstr); virBufferVSprintf(&buf, " <uuid>%s</uuid>\n", uuidstr); + if (def->ipv4routing && STRNEQ(def->ipv4routing, DNSMASQ)) + virBufferEscapeString(&buf, " <ipv4routing>%s</ipv4routing>\n", def->ipv4routing); + if (def->ipv6routing && STRNEQ(def->ipv6routing, RADVD)) + virBufferEscapeString(&buf, " <ipv6routing>%s</ipv6routing>\n", def->ipv6routing); + if (def->forwardType != VIR_NETWORK_FORWARD_NONE) { const char *mode = virNetworkForwardTypeToString(def->forwardType); if (mode) { diff --git a/src/conf/network_conf.h b/src/conf/network_conf.h index fd96c36..8df600c 100644 --- a/src/conf/network_conf.h +++ b/src/conf/network_conf.h @@ -98,6 +98,11 @@ struct _virNetworkDef { size_t nips; virNetworkIpDefPtr ips; /* ptr to array of IP addresses on this network */ + + char *ipv4routing; /* path to binary for configuring IPv4 network (DNS/DHCP). + Defaults to DNSMASQ */ + char *ipv6routing; /* path to binary for configuring IPv6 network. + Defaults to RADVD */ }; typedef struct _virNetworkObj virNetworkObj; diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 7d43ef5..444cbdc 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -177,7 +177,7 @@ networkFindActiveConfigs(struct network_driver *driver) { virReportOOMError(); goto cleanup; } - if (virFileLinkPointsTo(pidpath, DNSMASQ) == 0) + if (virFileLinkPointsTo(pidpath, obj->def->ipv4routing) == 0) obj->dnsmasqPid = -1; VIR_FREE(pidpath); } @@ -196,7 +196,7 @@ networkFindActiveConfigs(struct network_driver *driver) { VIR_FREE(radvdpidbase); goto cleanup; } - if (virFileLinkPointsTo(pidpath, RADVD) == 0) + if (virFileLinkPointsTo(pidpath, obj->def->ipv6routing) == 0) obj->radvdPid = -1; VIR_FREE(pidpath); } @@ -596,7 +596,7 @@ networkStartDhcpDaemon(virNetworkObjPtr network) goto cleanup; } - cmd = virCommandNew(DNSMASQ); + cmd = virCommandNew(network->def->ipv4routing); if (networkBuildDnsmasqArgv(network, ipdef, pidfile, cmd) < 0) { goto cleanup; } @@ -727,7 +727,7 @@ networkStartRadvd(virNetworkObjPtr network) * its own pidfile, so we just let it do so, with a slightly * different name. Unused, but harmless. */ - cmd = virCommandNewArgList(RADVD, "--debug", "1", + cmd = virCommandNewArgList(network->def->ipv6routing, "--debug", "1", "--config", configfile, "--pidfile", NULL); virCommandAddArgFormat(cmd, "%s-bin", pidfile); diff --git a/tests/networkxml2xmlin/with-ipv4routing-network.xml b/tests/networkxml2xmlin/with-ipv4routing-network.xml new file mode 100644 index 0000000..1cf3aa4 --- /dev/null +++ b/tests/networkxml2xmlin/with-ipv4routing-network.xml @@ -0,0 +1,11 @@ +<network> + <name>private</name> + <uuid>81ff0d90-c91e-6742-64da-4a736edb9a9b</uuid> + <ipv4routing>/usr/local/bin/my-dnsmasq</ipv4routing> + <bridge name="virbr2" /> + <ip address="192.168.152.1" netmask="255.255.255.0"> + <dhcp> + <range start="192.168.152.2" end="192.168.152.254" /> + </dhcp> + </ip> +</network> diff --git a/tests/networkxml2xmlout/with-ipv4routing-network.xml b/tests/networkxml2xmlout/with-ipv4routing-network.xml new file mode 100644 index 0000000..d649d78 --- /dev/null +++ b/tests/networkxml2xmlout/with-ipv4routing-network.xml @@ -0,0 +1,11 @@ +<network> + <name>private</name> + <uuid>81ff0d90-c91e-6742-64da-4a736edb9a9b</uuid> + <ipv4routing>/usr/local/bin/my-dnsmasq</ipv4routing> + <bridge name='virbr2' stp='on' delay='0' /> + <ip address='192.168.152.1' netmask='255.255.255.0'> + <dhcp> + <range start='192.168.152.2' end='192.168.152.254' /> + </dhcp> + </ip> +</network> diff --git a/tests/networkxml2xmltest.c b/tests/networkxml2xmltest.c index 7805548..86e3c81 100644 --- a/tests/networkxml2xmltest.c +++ b/tests/networkxml2xmltest.c @@ -90,6 +90,7 @@ mymain(int argc, char **argv) DO_TEST("nat-network"); DO_TEST("netboot-network"); DO_TEST("netboot-proxy-network"); + DO_TEST("with-ipv4routing-network"); return (ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); }

2 1

[libvirt] [PATCH] bridge: Fix uninitialized variable
by Jiri Denemark 05 Jan '11

05 Jan '11

--- Pushed under the build-breaker rule. src/util/bridge.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/bridge.c b/src/util/bridge.c index dcd3f39..81a043c 100644 --- a/src/util/bridge.c +++ b/src/util/bridge.c @@ -676,7 +676,7 @@ brAddInetAddress(brControl *ctl ATTRIBUTE_UNUSED, virSocketAddr *addr, unsigned int prefix) { - virCommandPtr cmd; + virCommandPtr cmd = NULL; char *addrstr; int ret = -1; @@ -715,7 +715,7 @@ brDelInetAddress(brControl *ctl ATTRIBUTE_UNUSED, virSocketAddr *addr, unsigned int prefix) { - virCommandPtr cmd; + virCommandPtr cmd = NULL; char *addrstr; int ret = -1; -- 1.7.3.4

5 9

[libvirt] [PATCH] docs: Add additional indentation to level 3 menu items
by Matthias Bolte 05 Jan '11

05 Jan '11

formatnetwork.html has a menu item at level 3. libvirt.css doesn't have a explicit rule for level 3 and level 3 and level 2 items end up at the same indentation level. Add an additional 1em indentation to level 3 menu items. --- docs/libvirt.css | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/docs/libvirt.css b/docs/libvirt.css index dfc93c6..049e332 100644 --- a/docs/libvirt.css +++ b/docs/libvirt.css @@ -106,6 +106,11 @@ h2, h3, h4, h5, h6 { border-left: 8px solid #c5dbd8; } +#menu ul.l3 li .inactive, +#menu ul.l3 li .active { + padding-left: 3em; +} + #headerLogo { position: absolute; -- 1.7.0.4

3 3

[libvirt] virsh help keywords vs. command prefixes
by Paul Jenner 04 Jan '11

04 Jan '11

Hi all. >From a quick play with virsh from the 0.8.7 pre-release, as an end user I wonder if it may be more intuitive for the new help keywords to match the virsh command prefixes where they exist. For example "help nwfilter" would tell me about nwfilter- commands. Some keywords follow the prefix convention (nodedev, secret, snapshot, pool) but some do not (interface vs. iface-, filter vs nwfilter-, network vs. net-, volume vs. vol-). Minor observation and close to release but I thought I would send on feedback to see what people thought. Paul

5 4