- Devel - libvirt List Archives

[Libvir] [PATCH] modular compilation of xen/qemu/test
by Daniel Veillard 14 Mar '07

14 Mar '07

The following patch adds 3 configure options to turn on or off the support for Xen, QEMU/KVM and the test driver, as the output of configure --help shows: ... --with-xen add XEN support (on) --with-qemu add QEMU/KVM support (on) --with-test add test driver support (on) ... It uses compile time flags, is not really intrusive, it just pointed out a few strangeness, one of which was fixed (a Xen include) and the remaining ones are: - xml.c doesn't depend on Xen except for virDomainXMLDevID calling xenStoreDomainGetNetworkID() - libvirt.c still calls xenDaemonDomainLookupByName_ids from virDomainGetUUID() , I'm not sure if this can be avoided In the tests directory, sexpr2xmltest.c and xmconfigtest.c still depends on code in xen specific modules, trying to add them without xen being present was more complex than the value it may add (it also makes the patch smaller to cut at the module level). Plan to commit this before the release, unless this raises a problem, Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

2 1

[Libvir] Reliably detecting if Intel VT is disabled in the BIOS
by Richard W.M. Jones 14 Mar '07

14 Mar '07

VMXON is the instruction which turns on Intel VT extensions[1]. This instruction can be enabled and disabled by setting a bit in a CPU register. Moreover, this CPU register itself can be locked so that no changes can be made until the CPU is power-cycled. In detail, the register is the IA32_FEATURE_CONTROL (0x3A) MSR. The relevant bits are: bit 0 Lock bit (0 = unlocked, 1 = permanently locked) bit 1 Enable VMXON in Intel Safer Mode Extensions (SMX) bit 2 Enable VMXON in normal operation So to find out if VT is possible with the CPU, use CPUID (in practice, check if "vmx" is in /proc/cpuinfo flags). To find out if VT can be turned on in the host, check bit 2 in the above register. There is a handy tool called msr-tools[2] which you can use: # ./rdmsr 0x3a ff03 (bit 2 is clear, so VT is _not_ enabled on this host). It seems that the BIOS locks the register (by writing 1 to bit 0). To find out if the BIOS has locked the register, use rdmsr again and look at the lowest bit. In the example above you can see that the BIOS disabled VT and locked the register. Once the register is locked, the only way around it is to reboot. If the register is unlocked you can enable VT by writing a 1 to bit 2. If you don't want to use the msr-tools, then direct access to the register can be had through /dev/cpu/<id>/msr. For example this is an strace of rdmsr 0x3a: open("/dev/cpu/0/msr", O_RDONLY) = 3 pread(3, "\3\377\0\0\0\0\0\0", 8, 58) = 8 However you need to be root to open /dev/cpu/0/msr. On machines which don't support the IA32_FEATURE_CONTROL MSR you will get an EIO error: pread(3, 0x7ffff81ec810, 8, 58) = -1 EIO (Input/output error) Rich. Notes: [1] I don't think this is possible with AMD's Pacifica extensions. I'm not sure if it's possible to disable these in the BIOS & lock them. [2] http://www.kernel.org/pub/linux/utils/cpu/msr-tools/ Sources: * linux/drivers/kvm/vmx.c: function vmx_disabled_by_bios * "Intel® Trusted Execution Technology Preliminary Architecture Specification" section 2.1.2 (http://download.intel.com/technology/security/downloads/31516803.pdf) -- Emerging Technologies, Red Hat http://et.redhat.com/~rjones/ 64 Baker Street, London, W1U 7DF Mobile: +44 7866 314 421 "[Negative numbers] darken the very whole doctrines of the equations and make dark of the things which are in their nature excessively obvious and simple" (Francis Maseres FRS, mathematician, 1759)

1 0

[Libvir] [PATCH] Add virConnectGetCapabilities call to return the capabilities of the driver / hypervisor
by Richard W.M. Jones 13 Mar '07

13 Mar '07

This patch: * Adds virConnectGetCapabilities which returns an XML description of the capabilities of the driver or hypervisor under this connection (see below for more about the XML format). * Adds virsh capabilities command. * Checks the Python binding works. Caveats: * Xen implementation (in xen_internal.c) is not tested because I don't have a working Xen machine right now. QEMU, test and remote work however. Example: $ virsh -c test:///default capabilities | tidy -xml -i -q <capabilities> <domain> <type>test</type> </domain> <host_features></host_features> <guest_architectures> <guest_architecture> <model>i686</model> <bits>32</bits> <guest_features> <accelerated /> <pae /> </guest_features> </guest_architecture> </guest_architectures> </capabilities> $ virsh -c qemu:///session capabilities | tidy -xml -i -q <capabilities> <domain> <type>qemu</type> </domain> <guest_architectures> <guest_architecture> <model>i686</model> <bits>32</bits> <features> <emulated /> </features> </guest_architecture> <guest_architecture> <model>x86_64</model> <bits>64</bits> <features> <emulated /> </features> </guest_architecture> [... other archs deleted ...] </guest_architectures> </capabilities> XML format: The <capabilities> node contains information about the host and a list of supported architectures for guests. About the host we return: <domain> <type> The default type acceptable when creating a domain. <host_features> <hvm> Does the host CPU support HVM? <type> Type of HVM (eg. "vmx") <bios_setting> "enabled" or "disabled" About the guest architectures we can return: <guest_architecture> <model> The (virtualised) CPU <bits> Bits <domain> <type> If different from the default type (eg. for kqemu) <features> <hvm> Full virt guest? <accelerated> Accelerated (eg. Xen or kqemu) <emulated> Emulated (eg. qemu or bochs) (and a few other esoteric flags here) Rich. -- Emerging Technologies, Red Hat http://et.redhat.com/~rjones/ 64 Baker Street, London, W1U 7DF Mobile: +44 7866 314 421 "[Negative numbers] darken the very whole doctrines of the equations and make dark of the things which are in their nature excessively obvious and simple" (Francis Maseres FRS, mathematician, 1759)

4 13

[Libvir] virsh command is getting a connecting error
by Saori Fukuta 13 Mar '07

13 Mar '07

Hi, It was working fine on March 8th, but today it failed to do virsh command with an error message. # virsh list libvir: error : no support for hypervisor lt-virsh: error: failed to connect to the hypervisor This message is printed here because nb_network_drivers is 0. --- libvirt.c : virConnectOpen 317 if (ret->nb_drivers == 0 || ret->nb_network_drivers == 0) { 318 /* we failed to find an adequate driver */ 319 virLibConnError(NULL, VIR_ERR_NO_SUPPORT, name); 320 goto failed; 321 } I think the cause is because it failed to bind the socket at qemuOpenClientUNIX. --- qemu_internal.c : qemuOpenClientUNIX 232 /* 233 * now bind the socket to that address and listen on it 234 */ 235 if (connect(fd, (struct sockaddr *) &addr, sizeof(addr)) < 0) { They are setting members of sockaddr: {sun_family = 1, sun_path = "/usr/var/run/libvirt/qemud-sock", '\0'} and connect returned -1 with errno that was 2(ENOENT). Is there any problem? Please tell me how I can use current virsh command. Thanks, Saori Fukuta.

2 2

[Libvir] Inconsistent networking between Xen & QEMU
by Daniel P. Berrange 09 Mar '07

09 Mar '07

In Xen it is possible to configure a VIF to enslave with an arbitrary bridge device using a fragment of XML like: <interface type='bridge'> <source bridge='xenbr0'/> <mac address='00:16:3E:66:94:9C'/> </interface> In this example I've enslaved to xenbr0 which is created by the Xen networking script, but this will work with any bridge device really. In QEMU the only bridging is available via the 'virtual networks' support in libvirt, by referencing a named bridge. For example <interface type='network'> <source network='default'/> <mac address='00:16:3E:66:94:9C'/> </interface> Now 'default' is merely defined to be backed by a bridge device 'virbr0' (although it does not neccessary get backed by a bridge) so the basic impl is basically just enslaving a TAP device. Anyway, we now have two different syntaxes in the XML for defining what is more or less the same thing which is a real PITA for the tools. It is useful to have the distinct representations - one is for 'managed' bridges, and the other is for 'unmanaged' bridges - we can do some much more interesting things with managed bridges. To improve life for the tools though I'd like to do two things as a high priority - Support managed networks in Xen - we can simply lookup the bridge device associated with a network and tell Xen to use vif-bridge with this device. For the reverse Xen will tell us what bridge device a guest is using, and we can reverse lookup the corresponding network - Support non-managed networks in QEMU - we can simply enslave the QEMU tap device to an arbitrary user specific bridge device. Until one of these two options is done, its going to be a PITA to support networking across QEMU & Xen in virt-manager/virt-install. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

2 10

[Libvir] [PATCH] Add a default network
by Mark McLoughlin 09 Mar '07

09 Mar '07

Hey, So, we want to install a default network which guests can connect to. This can be seen as e.g. a replacement for xenbr0 as the default bridge for xen guests. A few things concern me about the patch, see below for my thinking. However, I'm happy to punt on all of these issues for now and go ahead with the patch. 1) Ordering - we want the default network to have a predictable bridge name, so instead of relying on it being the first network and ending up auto-allocated vnet0, we use virbr0. The alternative is to actually have an ordering scheme, e.g. #define LIBVIRT_AUTOSTART_PRIORITY_DISABLE -1 #define LIBVIRT_AUTOSTART_PRIORITY_FIRST 0 #define LIBVIRT_AUTOSTART_PRIORITY_LAST 99 #define LIBVIRT_AUTOSTART_PRIORITY_DEFAULT 50 int virNetworkSetAutostart(virDomainPtr domain, int priority); autostart/00-default.xml -> ../default.xml autostart/99-MyNetwork.xml -> ../MyNetwork.xml Maybe that's useful, but probably moreso for guests than networks. The "virbr0" solution is fine, IMHO. 2) IP address choice - I've randomly chosen 192.168.122.1/24 as the IP address for the network, and this could happen to clash with an existing network. Since we masquerade outgoing traffic, I think the only problem this would cause in practice would be where the host machine is already on a 192.168.122/24 subnet and it could find itself connecting to a guest rather than another machine on the physical subnet. It's a default configuration choice which admins can change, and I don't see it causing huge problems in practice, so I'm not overly concerned. If it did turn out to be a problem, I'd probably add something like: <ip type="auto" /> But I'm not anxious to go down that route right now. 3) UUID generation - there's no UUID specified, so the default network will have a different UUID across reboots. We could trivially save the UUID at first boot, but we'd really need to put it in /var then. IMHO, it's something we can fix later if it turns out to be important. Cheers, Mark.

3 8

[Libvir] Restarting of libvirt_qemud daemon
by Daniel P. Berrange 09 Mar '07

09 Mar '07

Thinking about later RPM upgrades I think we need to think about whether it will be possible to restart the libvirt_qemud while guests & networks are running. There's a couple of issues: - We do waitpid() to cleanup qemu & dnsmasq processes when we stop domains & networks, or to detect when they crash. For the former, we could may they daemons to avoid waitpid() cleanup, but we'd still need waitpid to be able to detect shutdowns. There is also the issue of enumerating running instances. - We always try to re-create a bridge device at startup, even if it already exists. Likewise we always try to add the IPtables rules & start dnsmasq. We can easily detect if the bridge already exists. I think we can probably double check iptables rulles too., The tricky one is figuring out whether a dnsmasq instance is still running. Dealing with theses not only helps planned restarts, but will also make it possible start up the daemon again after a crash without having to kill off all guests & networks manually. So I think it is worth investigating what we can do to enable restarts. It might be worth waiting until we sort out whether we'll merge libvirt_qemud with the generic libvirtd remote daemon though so we don't have to do the work twice over. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

4 8

[Libvir] Stable UUID needed for default network
by Daniel P. Berrange 09 Mar '07

09 Mar '07

The UUID for the default network is not included in the example default config. So every time you restart the libvirt_qemud we get a randomly generated UUID. This isn't really in keeping with the idea of having a permanent UID :-) So I think we need to do two things: - In the 'make install' routine - if-and-only-if the $prefix/etc/libvirt/networks/default.xml file was newly installed, then stuff a random UUID into it, eg UUID=`uuidgen` sed -e "s,</name>,</name>\n <uuid>$UUID</uuid>," \ $prefix/etc/libvirt/networks/default.xml If the system doesn't have uuidgen, we can just ignore this step - In the RPM %post section we need to stuff in a UUID at install time if-and-only-if this is a new install - ie leave it alone during upgrades UUID=`uuidgen` sed -e "s,</name>,</name>\n <uuid>$UUID</uuid>," \ %(sysconfdir)/libvirt/networks/default.xml Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

2 1

[Libvir] Alternate dnsmasq command line args ?
by Daniel P. Berrange 09 Mar '07

09 Mar '07

Currently we launch dnsmasq with a command line looking like: dnsmasq --keep-in-foreground --bind-interfaces --pid-file --conf-file --except-interface lo --listen-address 192.168.122.1 --dhcp-range 192.168.122.2,192.168.122.254 This causes DNSMASQ to listen on any network interface configured with an IP address of '192.168.122.1', except for loopback. I know it would be kind of unusual, but it is possible two interface might be both configured with the same address. I think it may be safer if we explicitly tell it to use virbr0 instead eg --interface virbr0 --except-interface lo Yes, the --except-interface lo is still needed due to wierd ass dnsmasq rules which make it always use loopback unless you tell it not to. I'm also not clear why we use --conf-file - we're not actually telling it about any config file so this seems redundant ? Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

2 2

[Libvir] Generate python binding for network APIs
by Daniel P. Berrange 09 Mar '07

09 Mar '07

I was about to make use of the new networking APIs in virt-manager/virt-install when I discovered we don't have any python binding for it. Attached is the quickest patch I could write to support it - I basically copied all the virDomain related bits & s/Domain/Network/ throughout. A simple demo script works: #!/usr/bin/python import libvirt con = libvirt.open(None) n = con.listNetworks() print str(n) for name in n: net = con.lookupByName(name) print net.XMLDesc(0) print net.bridgeName() I think there are a couple more methods which need manual implementations though - eg UUIDString() is generated incorrectly - in fact the UUIDString() impl for virDomains is already broken, and so is the binding for VCPU pinning. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|

3 3