May 2009 - Devel - libvirt List Archives

[libvirt] suspend/hibernate guests on host shutdown
by Gerry Reno 26 May '09

26 May '09

There are times when you just have to reboot the host machine to clear a problem or after some maintenance. The problem right now is that this forces all the guests to go through an entire reboot as well. What would be good is if the host shutdown would invoke a suspend/hibernate of the guests and then restore the guests to their same condition on the reboot. The guests should notice nothing other that a small loss of wallclock time. Regards, Gerry

1 0

[libvirt] HELP: after host upgrade to F11, guest runs extremely slow
by Gerry Reno 26 May '09

26 May '09

I upgraded the host from F10 to F11 (x86_64) with no issues. Now when I start a F10 (i386) guest it runs very very slow. I also see messages on the guest boot console about "clocksource tsc unstable" and some kernel oops. Once it got far enough to start network I logged in and checked the clocksource and it currently is 'acpi_pm' even though the kernel line says clocksource=pit. The available clocksources are acpi_pm, jiffies, and tsc. I do not see 'pit' in the list. How do I fix this issue? Regards, Gerry

4 7

[libvirt] VMware ESX driver announcement
by Matthias Bolte 26 May '09

26 May '09

Hello, I'm participate in a project of the Paderborn Center for Parallel Computing, an institute of the University of Paderborn: http://pc2.uni-paderborn.de The project's goal is to use virtualization in a supercomputer environment. We've decided to use libvirt to manage different hypervisor. A subgoal is to extend the driver base of libvirt. We've started an VMware ESX driver and are investigating Hyper-V support (see next mail). The ESX driver isn't complete yet, currently it supports: - domain lookup by ID, UUID and name - domain listing - domain suspend and resume - domain reboot, if the VMware tools are installed inside the domain - domain start and shutdown - domain migration The driver uses the SOAP based VMware VI API. We've tried to generate SOAP client code with gSOAP and looked at other SOAP libraries, but the VI API uses inheritance in a way that gSOAP fails to generate C code for. Because of this we wrote a minimal SOAP client based on libcurl and libxml2 that can handle this inheritance problem in C. The next item on the todo list is domain creation, because currently the driver can only deal with prior existing domains. We think this code might be useful for others and would be glad if the driver could be merged into libvirt in medium term. All of our work is licensed under LGPLv2+.

4 7

[libvirt] Koan, KVM and PXE
by Fabien Dupont 25 May '09

25 May '09

Hi all. I use Cobbler to manage my PXE boot. I would like to extend my use to manage VM deployment with Koan. But I have a problem related to network and my VM can't boot. I have already posted on Cobbler mailing-list but it's probably on kvm/libvirt side... I am on Fedora 10 with Cobbler 1.6.4 and Koan 1.6.3 (latest available from stable repos), and libvirt 0.5.1-2.fc10. I would like to create VMs using KVM thus I have imported FC10 DVD, added Everything and Updates repos and created a host using the profile created during import. The system has the following properties : root# cobbler system report vmtest system : vmtest profile : fedora-core-10-x86_64 comment : created : Fri May 8 13:43:40 2009 gateway : 192.168.1.254 hostname : vmtest image : kernel options : {} kernel options post : {} kickstart : <<inherit>> ks metadata : {} mgmt classes : [] modified : Sat May 23 18:42:59 2009 name servers : [] name servers search : [] netboot enabled? : True owners : ['admin'] redhat mgmt key : <<inherit>> redhat mgmt server : <<inherit>> server : <<inherit>> template files : {} virt cpus : <<inherit>> virt file size : <<inherit>> virt path : <<inherit>> virt ram : <<inherit>> virt type : <<inherit>> power type : ipmitool power address : power user : power password : power id : interface : eth0 mac address : ff:00:00:01:00:00 bonding : bonding_master : bonding_opts : is static? : True ip address : 192.168.1.201 subnet : 255.255.255.0 static routes : [] dns name : vmtest.local dhcp tag : virt bridge : br0 As you can see, I use a bridged network with a real bridge create with the method described on the wiki (eth0 is the real interface and br0 the bridge). Then I create the VM using : koan --server=cobbler.evenit.info --virt --system=vmtest The VM is correctly created and started : kernel and initrd are correctly set, virtual hard drive is right and network interface seems right. root# ps -edf | grep kvm root 19484 23655 67 21:54 ? 00:00:06 /usr/bin/qemu-kvm -S -M pc -m 256 -smp 1 -name vmtest -monitor pty -no-reboot -boot c -kernel /var/lib/libvirt/boot/virtinst-vmlinuz.omv9yr -initrd /var/lib/libvirt/boot/virtinst-initrd.img.Ufj-qc -append ks= http://cobbler.evenit.info/cblr/svc/op/ks/system/vmtest ksdevice=link kssendmac lang= text method= http://cobbler.evenit.info:80/cblr/links/fedora-core-10-x86_64/<http://cobbler.evenit.info/cblr/links/fedora-core-10-x86_64/>-drive file=/var/lib/libvirt/images/vmtest-disk0,if=virtio,index=0,boot=on -net nic,macaddr=ff:00:00:01:00:00,vlan=0,model=virtio -net tap,fd=14,script=,vlan=0,ifname=vnet0 -serial pty -parallel none -usb -vnc 127.0.0.1:0 -k en-us But, when I open the VM display with virt-manager, I see that installation can't start as DHCP is not answering... There's no trace of the DHCP requests on DHCP server and vnet0 seems to have no trafic : [root@lilith ~]# netstat -i Table d'interfaces noyau Iface MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR Flg br0 1500 0 173524 0 0 0 261005 0 0 0 BMRU eth0 1500 0 173589 0 0 0 260909 0 0 0 BMPRU lo 16436 0 29105 0 0 0 29105 0 0 0 LRU vnet0 1500 0 0 0 0 0 1 0 0 0 BMRU As far as I can see, it seems vnet0 doesn't forward its trafic to br0 like there's no link. I have also seen that the vnet0 interface seen from host (opposed to guest) doesn't have the MAC address asked for vnet0, but this may be normal. -- Fabien

1 0

[libvirt] PATCH: add support for -snapshot
by Geert Jansen 25 May '09

25 May '09

Hi, i'm working on some infrastructure to run unit tests in a virtual environment. For this, I need to be able to pass a "snapshot=on" argument to a qemu virtual disk so that I know I start with the same state all the time. The attached patch implements this. It adds a boolean <snapshot/> to the <disk> section in the XML. Please consider for inclusion. Regards, Geert Jansen

2 1

[libvirt] [PATCH 1/1] fixed the event-test.c example to check for return values
by Pritesh Kothari 25 May '09

25 May '09

Hi All, The event-test.c example doesn't check for the return values while registering for the callbacks, this patch just checks for them along with a correcting a small typo. (Unpaused => Paused) Regards, Pritesh

3 2

[libvirt] migration
by Zvi Dubitzky 25 May '09

25 May '09

Does anybody know if migration within a host from one domain to another works today It did not work for me for libvirt 0.6.1 thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

1 0

[libvirt] save/restore for KVM
by Christian Weyermann 24 May '09

24 May '09

Hello everybody, I try to save and restore a virtual machine, but I always get this error message: ~$ virsh save test1 test1.vm Domain test1 nach test1.vm gespeichert [german for test1 successfully saved to test1.vm] ~$ virsh restore test1.vm 17:53:10.341: error : this function is not supported by the hypervisor: STDIO migration is not supported with this QEMU binary libvir: QEMU error : this function is not supported by the hypervisor: STDIO migration is not supported with this QEMU binary 17:53:10.341: error : operation failed: failed to start VM libvir: QEMU error : operation failed: failed to start VM Fehler: Wiederherstellen der Domain aus test1.vm gescheitert [german: Restoring test1.vm failed!] Fehler: operation failed: failed to start VM I am using libvirt 0.6.3, KVM-84 with qemu 0.9.1. Does anyone have an idea, how to solve this issue? Thanks in advance and kind regards, Christian

1 0

[libvirt] libvirt native parameters
by Zvi Dubitzky 24 May '09

24 May '09

can anybody remind me where is the native qemu parameters that result from libvirt translation of an XML file stored (file path). thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

3 2

[libvirt] error: failed to connect to the hypervisor
by Erkan Unal 24 May '09

24 May '09

Hi, I installed the libvirt with the following configure options as an unprivileged user, I am planning to use it just for qemu/kvm: ./configure --without-xen --without-lxc --without-uml --without-openvz --without-vbox --without-numactl --prefix=/some/nonstandard/dir make make install --without-numactl is added because of a compilation error. I am not sure whether it is completely necessary or not. If you think it has effect on the error, please state the reason. After installation I got the following error when I tried to run the virsh without any options: error: failed to connect to the hypervisor With the following command line error was "Segmentation Fault": virsh -c qemu:///session. Qemu executable path: /usr/local/bin/qemu-system-x86_64 and it is defined on the path. Libvirt daemon is failing with the same error too. System ---> Scientific Linux SL release 4.7 (Beryllium), Kernel version: 2.6.28.2, libvirt version: 0.6.3 Any ideas? P.S.: Is there a way to compile static virsh executable? I tried altering make files, configure options, linker flags and compiler flags but no luck. "ldd virsh" command should say "not a dynamic executable" after compilation. Thanks, Erkan Unal

2 1

[libvirt] SetMaxMemory vs. SetMemory
by Matthias Bolte 24 May '09

24 May '09

Hello, I just took a look at the driver functions for SetMaxMemory and SetMemory, as they are not implemented yet for the ESX driver and Daniel Veillard was a bit surprised that they are missing, as he expects them to be simple to implement. The problem is that I'm not sure how the memory model of ESX maps to SetMaxMemory vs. SetMemory. An ESX virtual machine has a defined memory size. That's the size reported to the guest OS as the available "physical" memory size. Beside this ESX allows the user to control how the hypervisor satisfies this "physical" memory size. You can define a reservation and an upper limit. The hypervisor will at least use the reserved amount of real physical memory to satisfy the "physical" memory size, but will not use more than the upper limit. How does this map to SetMaxMemory and SetMemory? My first assumption was, SetMaxMemory defines the "physical" memory size and SetMemory defines the upper limit of real physical memory to satisfy the "physical" memory size. This assumption seems to be in sync with the QEMU driver from just looking at the code. But with Xen it seems to be different. If I call SetMaxMemory and SetMemory with 2GB than free inside the domain reports 2GB total memory. After I call SetMemory with 1GB, free reports 1GB of total memory. I'm confused. So, what is the intended semantic for SetMaxMemory and SetMemory? Regards, Matthias

4 6

[libvirt] Re: BUG REPORT: -net tap, fd=X does not work correctly in kvm-86
by Anthony Liguori 24 May '09

24 May '09

Josh Wilsdon wrote: > We are doing work with libvirt 0.6.3 through which we have started > virtual machines. Using the same "create" command with the same > storage, the same everything else this configuration works with both > kvm-84 and kvm-85. When I built a fresh version of kvm-86 this > stopped working with an error: > > qemu: invalid parameter '15' in 'fd=15,script=,vlan=0,ifname=vnet0' > It's a libvirt bug. script and ifname have never been valid parameters for -net tap. Historically, -net has ignored unknown options but recently, it now throws an error when encountering an unsupported option. Regards, Anthony Liguori

4 3

[libvirt] PATCH: Ensure filename/funcname/linenr gets passed to error func
by Daniel P. Berrange 22 May '09

22 May '09

The current virRaiseError() function does not accept any information about these source location of the error. The various virReportSystemError and virReportOOMError wrappers all collect this info, but then discard it. This patch turns virRaiseError into virRaiseErrorFull() and adds params for linenr, filename & funcname, and then adds a macro providing the original virRaiseError() API contract, and collecting source location info. The other wrappers are changed to call virRaiseErrorFull() directly preserving their source location data. Finally the source location is passed into the virLogMessage call. This should address the issue that Mattias mentioned in the VMWare driver thread. Regards, Daniel diff -r 625ffe1918a4 src/virterror.c --- a/src/virterror.c Thu May 21 15:56:27 2009 +0100 +++ b/src/virterror.c Thu May 21 16:13:49 2009 +0100 @@ -587,10 +587,11 @@ virSetConnError(virConnectPtr conn) /** - * virRaiseError: + * virRaiseErrorFull: * @conn: the connection to the hypervisor if available - * @dom: the domain if available - * @net: the network if available + * @filename: filename where error was raised + * @funcname: function name where error was raised + * @linenr: line number where error was raised * @domain: the virErrorDomain indicating where it's coming from * @code: the virErrorNumber code for the error * @level: the virErrorLevel for the error @@ -606,12 +607,19 @@ virSetConnError(virConnectPtr conn) * immediately if a callback is found and store it for later handling. */ void -virRaiseError(virConnectPtr conn, - virDomainPtr dom ATTRIBUTE_UNUSED, - virNetworkPtr net ATTRIBUTE_UNUSED, - int domain, int code, virErrorLevel level, - const char *str1, const char *str2, const char *str3, - int int1, int int2, const char *msg, ...) +virRaiseErrorFull(virConnectPtr conn, + const char *filename ATTRIBUTE_UNUSED, + const char *funcname, + size_t linenr, + int domain, + int code, + virErrorLevel level, + const char *str1, + const char *str2, + const char *str3, + int int1, + int int2, + const char *fmt, ...) { virErrorPtr to; void *userData = virUserData; @@ -647,18 +655,18 @@ virRaiseError(virConnectPtr conn, /* * formats the message */ - if (msg == NULL) { + if (fmt == NULL) { str = strdup(_("No error message provided")); } else { - VIR_GET_VAR_STR(msg, str); + VIR_GET_VAR_STR(fmt, str); } /* * Hook up the error or warning to the logging facility - * TODO: pass function name and lineno + * XXXX should we include filename as 'category' instead of domain name ? */ virLogMessage(virErrorDomainName(domain), virErrorLevelPriority(level), - NULL, 0, 1, "%s", str); + funcname, linenr, 1, "%s", str); /* * Save the information about the error @@ -1064,10 +1072,12 @@ virErrorMsg(virErrorNumber error, const * Helper function to do most of the grunt work for individual driver * ReportError */ -void virReportErrorHelper(virConnectPtr conn, int domcode, int errcode, - const char *filename ATTRIBUTE_UNUSED, - const char *funcname ATTRIBUTE_UNUSED, - size_t linenr ATTRIBUTE_UNUSED, +void virReportErrorHelper(virConnectPtr conn, + int domcode, + int errcode, + const char *filename, + const char *funcname, + size_t linenr, const char *fmt, ...) { va_list args; @@ -1083,8 +1093,10 @@ void virReportErrorHelper(virConnectPtr } virerr = virErrorMsg(errcode, (errorMessage[0] ? errorMessage : NULL)); - virRaiseError(conn, NULL, NULL, domcode, errcode, VIR_ERR_ERROR, - virerr, errorMessage, NULL, -1, -1, virerr, errorMessage); + virRaiseErrorFull(conn, filename, funcname, linenr, + domcode, errcode, VIR_ERR_ERROR, + virerr, errorMessage, NULL, + -1, -1, virerr, errorMessage); } @@ -1113,9 +1125,9 @@ const char *virStrerror(int theerrno, ch void virReportSystemErrorFull(virConnectPtr conn, int domcode, int theerrno, - const char *filename ATTRIBUTE_UNUSED, - const char *funcname ATTRIBUTE_UNUSED, - size_t linenr ATTRIBUTE_UNUSED, + const char *filename, + const char *funcname, + size_t linenr, const char *fmt, ...) { char strerror_buf[1024]; @@ -1145,19 +1157,21 @@ void virReportSystemErrorFull(virConnect if (!msgDetail) msgDetail = errnoDetail; - virRaiseError(conn, NULL, NULL, domcode, VIR_ERR_SYSTEM_ERROR, VIR_ERR_ERROR, - msg, msgDetail, NULL, -1, -1, msg, msgDetail); + virRaiseErrorFull(conn, filename, funcname, linenr, + domcode, VIR_ERR_SYSTEM_ERROR, VIR_ERR_ERROR, + msg, msgDetail, NULL, -1, -1, msg, msgDetail); } void virReportOOMErrorFull(virConnectPtr conn, int domcode, - const char *filename ATTRIBUTE_UNUSED, - const char *funcname ATTRIBUTE_UNUSED, - size_t linenr ATTRIBUTE_UNUSED) + const char *filename, + const char *funcname, + size_t linenr) { const char *virerr; virerr = virErrorMsg(VIR_ERR_NO_MEMORY, NULL); - virRaiseError(conn, NULL, NULL, domcode, VIR_ERR_NO_MEMORY, VIR_ERR_ERROR, - virerr, NULL, NULL, -1, -1, virerr, NULL); + virRaiseErrorFull(conn, filename, funcname, linenr, + domcode, VIR_ERR_NO_MEMORY, VIR_ERR_ERROR, + virerr, NULL, NULL, -1, -1, virerr, NULL); } diff -r 625ffe1918a4 src/virterror_internal.h --- a/src/virterror_internal.h Thu May 21 15:56:27 2009 +0100 +++ b/src/virterror_internal.h Thu May 21 16:13:49 2009 +0100 @@ -33,17 +33,28 @@ extern void *virUserData; * * ************************************************************************/ int virErrorInitialize(void); -void virRaiseError(virConnectPtr conn, - virDomainPtr dom, - virNetworkPtr net, - int domain, - int code, - virErrorLevel level, - const char *str1, - const char *str2, - const char *str3, - int int1, int int2, const char *msg, ...) - ATTRIBUTE_FORMAT(printf, 12, 13); +void virRaiseErrorFull(virConnectPtr conn, + const char *filename, + const char *funcname, + size_t linenr, + int domain, + int code, + virErrorLevel level, + const char *str1, + const char *str2, + const char *str3, + int int1, + int int2, + const char *fmt, ...) + ATTRIBUTE_FORMAT(printf, 13, 14); + +/* Includes 'dom' and 'net' for compatbility, but they're ignored */ +#define virRaiseError(conn, dom, net, domain, code, level, \ + str1, str2, str3, int1, int2, msg, ...) \ + virRaiseErrorFull(conn, __FILE__, __FUNCTION__, __LINE__, \ + domain, code, level, str1, str2, str3, int1, int2, \ + msg, __VA_ARGS__) + const char *virErrorMsg(virErrorNumber error, const char *info); void virReportErrorHelper(virConnectPtr conn, int domcode, int errcode, const char *filename ATTRIBUTE_UNUSED, -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] storage: inconsistent in 'format' scheme between document and implementation
by Ryota Ozaki 22 May '09

22 May '09

Hi, I've found a problem that a 'format' element in storage-{pool,vol} XML doesn't behave as described in the document. The document says that a format of {pool,vol} is specified as the value of a format element, like: <target> <path>/var/lib/virt/images/sparse.img</path> <format>qcow2</format> <permissions> However, the implementation doesn't follow this: if (options->formatFromString) { char *format = virXPathString(conn, "string(/volume/target/format/@type)", ctxt); if (format == NULL) ret->target.format = options->defaultFormat; else ret->target.format = (options->formatFromString)(format); if (ret->target.format < 0) { virStorageReportError(conn, VIR_ERR_XML_ERROR, _("unknown volume format type %s"), format); VIR_FREE(format); goto cleanup; } VIR_FREE(format); } The implementation assumes that a format of {pool,vol} is specified as the attribute of a format element, like: <target> <path>/var/lib/virt/images/sparse.img</path> <format type='qcow2' /> <permissions> Thus, we need to fix either the document or the implementation. (I guess the implementation is correct, right?) Thanks, ozaki-r

3 10

[libvirt] [PATH 0/5] CORRECTED - Interface Config Public APIs and remote stubs, with MAC addresses in string format
by Laine Stump 21 May '09

21 May '09

MY APOLOGIES! Yesterday I accidentally sent the old patchset rather than the corrected set. This time it's right. These patches are based on those I sent last week (the public virInterface*() API definition, the local plumbing, and the RPC glue), but with suggestions incorporated: 1) MAC address is always used in null-terminated ASCII string format. This eliminates any potential problems with extra long addresses. 2) no comments in libvirt.h 3) flags arguments are all unsigned.

3 17

[libvirt] PATCH: Impl python bindings for virInterface
by Daniel P. Berrange 21 May '09

21 May '09

We forgot that adding the virInterface APIs would break the build of the python bindings. Here is the patch to wire up the code generator for the new APIs. This is actually surprisingly easy, since we ditched the raw unsigned char MAC address APIs, no special cases were needed. Daniel Index: generator.py =================================================================== RCS file: /data/cvs/libvirt/python/generator.py,v retrieving revision 1.42 diff -u -p -r1.42 generator.py --- generator.py 1 Apr 2009 10:39:12 -0000 1.42 +++ generator.py 20 May 2009 16:43:58 -0000 @@ -246,6 +246,11 @@ py_types = { 'virNetwork *': ('O', "virNetwork", "virNetworkPtr", "virNetworkPtr"), 'const virNetwork *': ('O', "virNetwork", "virNetworkPtr", "virNetworkPtr"), + 'virInterfacePtr': ('O', "virInterface", "virInterfacePtr", "virInterfacePtr"), + 'const virInterfacePtr': ('O', "virInterface", "virInterfacePtr", "virInterfacePtr"), + 'virInterface *': ('O', "virInterface", "virInterfacePtr", "virInterfacePtr"), + 'const virInterface *': ('O', "virInterface", "virInterfacePtr", "virInterfacePtr"), + 'virStoragePoolPtr': ('O', "virStoragePool", "virStoragePoolPtr", "virStoragePoolPtr"), 'const virStoragePoolPtr': ('O', "virStoragePool", "virStoragePoolPtr", "virStoragePoolPtr"), 'virStoragePool *': ('O', "virStoragePool", "virStoragePoolPtr", "virStoragePoolPtr"), @@ -289,6 +294,8 @@ skip_impl = ( 'virConnectListDefinedDomains', 'virConnectListNetworks', 'virConnectListDefinedNetworks', + 'virConnectListInterfaces', + 'virConnectListDefinedInterfaces', 'virConnectListStoragePools', 'virConnectListDefinedStoragePools', 'virConnectListStorageVols', @@ -608,6 +615,8 @@ classes_type = { "virDomain *": ("._o", "virDomain(self, _obj=%s)", "virDomain"), "virNetworkPtr": ("._o", "virNetwork(self, _obj=%s)", "virNetwork"), "virNetwork *": ("._o", "virNetwork(self, _obj=%s)", "virNetwork"), + "virInterfacePtr": ("._o", "virInterface(self, _obj=%s)", "virInterface"), + "virInterface *": ("._o", "virInterface(self, _obj=%s)", "virInterface"), "virStoragePoolPtr": ("._o", "virStoragePool(self, _obj=%s)", "virStoragePool"), "virStoragePool *": ("._o", "virStoragePool(self, _obj=%s)", "virStoragePool"), "virStorageVolPtr": ("._o", "virStorageVol(self, _obj=%s)", "virStorageVol"), @@ -621,7 +630,8 @@ classes_type = { converter_type = { } -primary_classes = ["virDomain", "virNetwork", "virStoragePool", "virStorageVol", +primary_classes = ["virDomain", "virNetwork", "virInterface", + "virStoragePool", "virStorageVol", "virConnect", "virNodeDevice" ] classes_ancestor = { @@ -629,6 +639,7 @@ classes_ancestor = { classes_destructors = { "virDomain": "virDomainFree", "virNetwork": "virNetworkFree", + "virInterface": "virInterfaceFree", "virStoragePool": "virStoragePoolFree", "virStorageVol": "virStorageVolFree", "virNodeDevice" : "virNodeDeviceFree" @@ -638,6 +649,7 @@ functions_noexcept = { 'virDomainGetID': True, 'virDomainGetName': True, 'virNetworkGetName': True, + 'virInterfaceGetName': True, 'virStoragePoolGetName': True, 'virStorageVolGetName': True, 'virStorageVolGetkey': True, @@ -690,6 +702,15 @@ def nameFixup(name, classe, type, file): elif name[0:16] == "virNetworkLookup": func = name[3:] func = string.lower(func[0:1]) + func[1:] + elif name[0:18] == "virInterfaceDefine": + func = name[3:] + func = string.lower(func[0:1]) + func[1:] + elif name[0:21] == "virInterfaceCreateXML": + func = name[3:] + func = string.lower(func[0:1]) + func[1:] + elif name[0:18] == "virInterfaceLookup": + func = name[3:] + func = string.lower(func[0:1]) + func[1:] elif name[0:20] == "virStoragePoolDefine": func = name[3:] func = string.lower(func[0:1]) + func[1:] @@ -717,6 +738,12 @@ def nameFixup(name, classe, type, file): elif name[0:10] == "virNetwork": func = name[10:] func = string.lower(func[0:1]) + func[1:] + elif name[0:15] == "virInterfaceGet": + func = name[13:] + func = string.lower(func[0:1]) + func[1:] + elif name[0:12] == "virInterface": + func = name[10:] + func = string.lower(func[0:1]) + func[1:] elif name[0:17] == "virStoragePoolGet": func = name[17:] func = string.lower(func[0:1]) + func[1:] @@ -988,7 +1015,7 @@ def buildWrappers(): else: txt.write("Class %s()\n" % (classname)) classes.write("class %s:\n" % (classname)) - if classname in [ "virDomain", "virNetwork", "virStoragePool", "virStorageVol", "virNodeDevice" ]: + if classname in [ "virDomain", "virNetwork", "virInterface", "virStoragePool", "virStorageVol", "virNodeDevice" ]: classes.write(" def __init__(self, conn, _obj=None):\n") else: classes.write(" def __init__(self, _obj=None):\n") @@ -996,7 +1023,7 @@ def buildWrappers(): list = reference_keepers[classname] for ref in list: classes.write(" self.%s = None\n" % ref[1]) - if classname in [ "virDomain", "virNetwork", "virNodeDevice" ]: + if classname in [ "virDomain", "virNetwork", "virInterface", "virNodeDevice" ]: classes.write(" self._conn = conn\n") elif classname in [ "virStorageVol", "virStoragePool" ]: classes.write(" self._conn = conn\n" + \ @@ -1097,6 +1124,10 @@ def buildWrappers(): classes.write( " if ret is None:raise libvirtError('%s() failed', net=self)\n" % (name)) + elif classname == "virInterface": + classes.write( + " if ret is None:raise libvirtError('%s() failed', net=self)\n" % + (name)) elif classname == "virStoragePool": classes.write( " if ret is None:raise libvirtError('%s() failed', pool=self)\n" % @@ -1177,6 +1208,10 @@ def buildWrappers(): classes.write ((" if " + test + ": raise libvirtError ('%s() failed', net=self)\n") % ("ret", name)) + elif classname == "virInterface": + classes.write ((" if " + test + + ": raise libvirtError ('%s() failed', net=self)\n") % + ("ret", name)) elif classname == "virStoragePool": classes.write ((" if " + test + ": raise libvirtError ('%s() failed', pool=self)\n") % @@ -1215,6 +1250,10 @@ def buildWrappers(): classes.write ((" if " + test + ": raise libvirtError ('%s() failed', net=self)\n") % ("ret", name)) + elif classname == "virInterface": + classes.write ((" if " + test + + ": raise libvirtError ('%s() failed', net=self)\n") % + ("ret", name)) elif classname == "virStoragePool": classes.write ((" if " + test + ": raise libvirtError ('%s() failed', pool=self)\n") % Index: libvirt_wrap.h =================================================================== RCS file: /data/cvs/libvirt/python/libvirt_wrap.h,v retrieving revision 1.15 diff -u -p -r1.15 libvirt_wrap.h --- libvirt_wrap.h 24 Nov 2008 19:28:12 -0000 1.15 +++ libvirt_wrap.h 20 May 2009 16:43:58 -0000 @@ -48,6 +48,15 @@ typedef struct { } PyvirNetwork_Object; +#define PyvirInterface_Get(v) (((v) == Py_None) ? NULL : \ + (((PyvirInterface_Object *)(v))->obj)) + +typedef struct { + PyObject_HEAD + virInterfacePtr obj; +} PyvirInterface_Object; + + #define PyvirStoragePool_Get(v) (((v) == Py_None) ? NULL : \ (((PyvirStoragePool_Object *)(v))->obj)) @@ -118,6 +127,7 @@ PyObject * libvirt_charPtrConstWrap(cons PyObject * libvirt_virConnectPtrWrap(virConnectPtr node); PyObject * libvirt_virDomainPtrWrap(virDomainPtr node); PyObject * libvirt_virNetworkPtrWrap(virNetworkPtr node); +PyObject * libvirt_virInterfacePtrWrap(virInterfacePtr node); PyObject * libvirt_virStoragePoolPtrWrap(virStoragePoolPtr node); PyObject * libvirt_virStorageVolPtrWrap(virStorageVolPtr node); PyObject * libvirt_virEventHandleCallbackWrap(virEventHandleCallback node); Index: types.c =================================================================== RCS file: /data/cvs/libvirt/python/types.c,v retrieving revision 1.17 diff -u -p -r1.17 types.c --- types.c 18 Dec 2008 12:25:11 -0000 1.17 +++ types.c 20 May 2009 16:43:58 -0000 @@ -119,6 +119,21 @@ libvirt_virNetworkPtrWrap(virNetworkPtr } PyObject * +libvirt_virInterfacePtrWrap(virInterfacePtr node) +{ + PyObject *ret; + + if (node == NULL) { + Py_INCREF(Py_None); + return (Py_None); + } + ret = + PyCObject_FromVoidPtrAndDesc((void *) node, (char *) "virInterfacePtr", + NULL); + return (ret); +} + +PyObject * libvirt_virStoragePoolPtrWrap(virStoragePoolPtr node) { PyObject *ret; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 3

[libvirt] PATCH: Work in progress suport for video device config
by Daniel P. Berrange 21 May '09

21 May '09

We discussed adding a new XML element for representing video devices a few weeks back. This is my work in progress patch for the XML parsing routines supporting <video> <model type='vga|cirrus|vmvga|xen' vram='64' heads='1'/> </video> For compatability, if an existing guest has a <graphics> tag, but no <video> tag, then the parser automatically adds a <video> tag for type=cirrus. Still todo - Tweak the addition of default <video> tag a little so it uses the correct type for the type of guest - eg it should use type=xen in some cases. - Add RNG XML schemas & website docs - Make QEMU driver use this info for setting -vga argument - Make Xen drivers use this info for setting stdvga=1|0 config arg - Make VirtualBox use this info in whatever way it needs Daniel domain_conf.c | 168 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ domain_conf.h | 26 ++++++++ 2 files changed, 194 insertions(+) diff -r 66fa9bfc797c src/domain_conf.c --- a/src/domain_conf.c Mon May 18 11:28:46 2009 +0100 +++ b/src/domain_conf.c Mon May 18 11:29:00 2009 +0100 @@ -80,6 +80,7 @@ VIR_ENUM_IMPL(virDomainDevice, VIR_DOMAI "interface", "input", "sound", + "video", "hostdev") VIR_ENUM_IMPL(virDomainDisk, VIR_DOMAIN_DISK_TYPE_LAST, @@ -140,6 +141,12 @@ VIR_ENUM_IMPL(virDomainSoundModel, VIR_D "pcspk", "ac97") +VIR_ENUM_IMPL(virDomainVideo, VIR_DOMAIN_VIDEO_TYPE_LAST, + "vga", + "cirrus", + "vmvga", + "xen") + VIR_ENUM_IMPL(virDomainInput, VIR_DOMAIN_INPUT_TYPE_LAST, "mouse", "tablet") @@ -372,6 +379,14 @@ void virDomainSoundDefFree(virDomainSoun VIR_FREE(def); } +void virDomainVideoDefFree(virDomainVideoDefPtr def) +{ + if (!def) + return; + + VIR_FREE(def); +} + void virDomainHostdevDefFree(virDomainHostdevDefPtr def) { if (!def) @@ -399,6 +414,9 @@ void virDomainDeviceDefFree(virDomainDev case VIR_DOMAIN_DEVICE_SOUND: virDomainSoundDefFree(def->data.sound); break; + case VIR_DOMAIN_DEVICE_VIDEO: + virDomainVideoDefFree(def->data.video); + break; case VIR_DOMAIN_DEVICE_HOSTDEV: virDomainHostdevDefFree(def->data.hostdev); break; @@ -456,6 +474,10 @@ void virDomainDefFree(virDomainDefPtr de virDomainSoundDefFree(def->sounds[i]); VIR_FREE(def->sounds); + for (i = 0 ; i < def->nvideos ; i++) + virDomainVideoDefFree(def->videos[i]); + VIR_FREE(def->videos); + for (i = 0 ; i < def->nhostdevs ; i++) virDomainHostdevDefFree(def->hostdevs[i]); VIR_FREE(def->hostdevs); @@ -1638,6 +1660,84 @@ error: goto cleanup; } +static virDomainVideoDefPtr +virDomainVideoDefParseXML(virConnectPtr conn, + const xmlNodePtr node, + int flags ATTRIBUTE_UNUSED) { + virDomainVideoDefPtr def; + xmlNodePtr cur; + char *type = NULL; + char *heads = NULL; + char *vram = NULL; + + if (VIR_ALLOC(def) < 0) { + virReportOOMError(conn); + return NULL; + } + + cur = node->children; + while (cur != NULL) { + if (cur->type == XML_ELEMENT_NODE) { + if ((type == NULL) && (vram == NULL) && (heads == NULL) && + xmlStrEqual(cur->name, BAD_CAST "model")) { + type = virXMLPropString(cur, "type"); + vram = virXMLPropString(cur, "vram"); + heads = virXMLPropString(cur, "heads"); + } + } + cur = cur->next; + } + + if (type && + (def->type = virDomainVideoTypeFromString(type)) < 0) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown video model '%s'"), type); + goto error; + } + + if (vram && + virStrToLong_ui(vram, NULL, 10, &def->vram) < 0) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse video ram '%s'"), vram); + goto error; + } else { + switch (def->type) { + /* Wierd, QEMU defaults to 9 MB ??! */ + case VIR_DOMAIN_VIDEO_TYPE_VGA: + case VIR_DOMAIN_VIDEO_TYPE_CIRRUS: + case VIR_DOMAIN_VIDEO_TYPE_VMVGA: + def->vram = 9 * 1024; + break; + case VIR_DOMAIN_VIDEO_TYPE_XEN: + /* Original PVFB hardcoded to 4 MB */ + def->vram = 4 * 1024; + break; + } + } + + if (heads && + virStrToLong_ui(heads, NULL, 10, &def->heads) < 0) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse video heads '%s'"), heads); + goto error; + } else { + def->heads = 1; + } + + VIR_FREE(type); + VIR_FREE(vram); + VIR_FREE(heads); + + return def; + +error: + virDomainVideoDefFree(def); + VIR_FREE(type); + VIR_FREE(vram); + VIR_FREE(heads); + return NULL; +} + static int virDomainHostdevSubsysUsbDefParseXML(virConnectPtr conn, const xmlNodePtr node, @@ -2064,6 +2164,10 @@ virDomainDeviceDefPtr virDomainDeviceDef dev->type = VIR_DOMAIN_DEVICE_SOUND; if (!(dev->data.sound = virDomainSoundDefParseXML(conn, node, flags))) goto error; + } else if (xmlStrEqual(node->name, BAD_CAST "video")) { + dev->type = VIR_DOMAIN_DEVICE_VIDEO; + if (!(dev->data.video = virDomainVideoDefParseXML(conn, node, flags))) + goto error; } else if (xmlStrEqual(node->name, BAD_CAST "hostdev")) { dev->type = VIR_DOMAIN_DEVICE_HOSTDEV; if (!(dev->data.hostdev = virDomainHostdevDefParseXML(conn, node, flags))) @@ -2584,6 +2688,40 @@ static virDomainDefPtr virDomainDefParse } VIR_FREE(nodes); + /* analysis of the video devices */ + if ((n = virXPathNodeSet(conn, "./devices/video", ctxt, &nodes)) < 0) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("cannot extract video devices")); + goto error; + } + if (n && VIR_ALLOC_N(def->videos, n) < 0) + goto no_memory; + for (i = 0 ; i < n ; i++) { + virDomainVideoDefPtr video = virDomainVideoDefParseXML(conn, + nodes[i], + flags); + if (!video) + goto error; + def->videos[def->nvideos++] = video; + } + VIR_FREE(nodes); + + /* For backwards compatability, if no <video> tag is set but there + * is a <graphics> tag, then we add a single video tag */ + if (def->ngraphics && !def->nvideos) { + virDomainVideoDefPtr video; + if (VIR_ALLOC(video) < 0) + goto no_memory; + video->type = VIR_DOMAIN_VIDEO_TYPE_CIRRUS; + video->vram = 9 * 1024; + video->heads = 1; + if (VIR_ALLOC_N(def->videos, 1) < 0) { + virDomainVideoDefFree(video); + goto no_memory; + } + def->videos[def->nvideos++] = video; + } + /* analysis of the host devices */ if ((n = virXPathNodeSet(conn, "./devices/hostdev", ctxt, &nodes)) < 0) { virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, @@ -3286,6 +3424,32 @@ virDomainSoundDefFormat(virConnectPtr co } static int +virDomainVideoDefFormat(virConnectPtr conn, + virBufferPtr buf, + virDomainVideoDefPtr def) +{ + const char *model = virDomainVideoTypeToString(def->type); + + if (!model) { + virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("unexpected video model %d"), def->type); + return -1; + } + + virBufferAddLit(buf, " <video>\n"); + virBufferVSprintf(buf, " <model type='%s'", + model); + if (def->vram) + virBufferVSprintf(buf, " vram='%u'", def->vram); + if (def->heads) + virBufferVSprintf(buf, " heads='%u'", def->heads); + virBufferAddLit(buf, "/>\n"); + virBufferAddLit(buf, " </video>\n"); + + return 0; +} + +static int virDomainInputDefFormat(virConnectPtr conn, virBufferPtr buf, virDomainInputDefPtr def) @@ -3660,6 +3824,10 @@ char *virDomainDefFormat(virConnectPtr c if (virDomainSoundDefFormat(conn, &buf, def->sounds[n]) < 0) goto cleanup; + for (n = 0 ; n < def->nvideos ; n++) + if (virDomainVideoDefFormat(conn, &buf, def->videos[n]) < 0) + goto cleanup; + for (n = 0 ; n < def->nhostdevs ; n++) if (virDomainHostdevDefFormat(conn, &buf, def->hostdevs[n]) < 0) goto cleanup; diff -r 66fa9bfc797c src/domain_conf.h --- a/src/domain_conf.h Mon May 18 11:28:46 2009 +0100 +++ b/src/domain_conf.h Mon May 18 11:29:00 2009 +0100 @@ -264,6 +264,25 @@ struct _virDomainSoundDef { int model; }; + +enum virDomainVideoType { + VIR_DOMAIN_VIDEO_TYPE_VGA, + VIR_DOMAIN_VIDEO_TYPE_CIRRUS, + VIR_DOMAIN_VIDEO_TYPE_VMVGA, + VIR_DOMAIN_VIDEO_TYPE_XEN, + + VIR_DOMAIN_VIDEO_TYPE_LAST +}; + + +typedef struct _virDomainVideoDef virDomainVideoDef; +typedef virDomainVideoDef *virDomainVideoDefPtr; +struct _virDomainVideoDef { + int type; + unsigned int vram; + unsigned int heads; +}; + /* 3 possible graphics console modes */ enum virDomainGraphicsType { VIR_DOMAIN_GRAPHICS_TYPE_SDL, @@ -360,6 +379,7 @@ enum virDomainDeviceType { VIR_DOMAIN_DEVICE_NET, VIR_DOMAIN_DEVICE_INPUT, VIR_DOMAIN_DEVICE_SOUND, + VIR_DOMAIN_DEVICE_VIDEO, VIR_DOMAIN_DEVICE_HOSTDEV, VIR_DOMAIN_DEVICE_LAST, @@ -375,6 +395,7 @@ struct _virDomainDeviceDef { virDomainNetDefPtr net; virDomainInputDefPtr input; virDomainSoundDefPtr sound; + virDomainVideoDefPtr video; virDomainHostdevDefPtr hostdev; } data; }; @@ -491,6 +512,9 @@ struct _virDomainDef { int nsounds; virDomainSoundDefPtr *sounds; + int nvideos; + virDomainVideoDefPtr *videos; + int nhostdevs; virDomainHostdevDefPtr *hostdevs; @@ -557,6 +581,7 @@ void virDomainFSDefFree(virDomainFSDefPt void virDomainNetDefFree(virDomainNetDefPtr def); void virDomainChrDefFree(virDomainChrDefPtr def); void virDomainSoundDefFree(virDomainSoundDefPtr def); +void virDomainVideoDefFree(virDomainVideoDefPtr def); void virDomainHostdevDefFree(virDomainHostdevDefPtr def); void virDomainDeviceDefFree(virDomainDeviceDefPtr def); void virDomainDefFree(virDomainDefPtr vm); @@ -671,6 +696,7 @@ VIR_ENUM_DECL(virDomainFS) VIR_ENUM_DECL(virDomainNet) VIR_ENUM_DECL(virDomainChr) VIR_ENUM_DECL(virDomainSoundModel) +VIR_ENUM_DECL(virDomainVideo) VIR_ENUM_DECL(virDomainHostdevMode) VIR_ENUM_DECL(virDomainHostdevSubsys) VIR_ENUM_DECL(virDomainInput) -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 3

[libvirt] [PATCH]: Secure migration support for KVM
by Chris Lalancette 21 May '09

21 May '09

All, Attached is the secure migration patch for libvirt. What this patch implements is a new remote RPC call for secure migration. On the source of the migration, we do a migration from the qemu process to the libvirtd on localhost. As each read() in libvirtd completes, it issues an RPC message to the remote libvirtd, using the standard libvirt RPC mechanisms. On the destination, we do essentially the mirror; the libvirtd accepts the data from RPC, and then writes it to a qemu container process listening on localhost. In order to actually use this, the command-line is pretty complex. If you want to use standard live migration, the command-line looks something like: # virsh -c qemu+tls://source.example.org/system migrate --live guest qemu+tls://dest.example.org/system This says to a live migration of "guest" from "source.example.org" to "dest.example.org", connecting to each of the remote libvirtd via TLS. Note that in this model, the virsh process connects to the remote libvirtd's via the -c argument (source) and the destination argument (dest). To do secure live migration, this becomes: # virsh -c qemu+tls://source.example.org/system migrate --live --secure guest qemu+tls://dest.example.org/system qemu+tls://dest.example.org/system This says to do a secure live migration of "guest" from "source.example.org" to "dest.example.org". The virsh process figures out how to connect to the remote libvirtd's via the -c argument (source) and the destination argument (dest). The second "qemu+tls://dest.example.org/system" is the secure channel from the source libvirtd to the destination libvirtd. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> docs/apibuild.py | 1 include/libvirt/libvirt.h | 1 include/libvirt/libvirt.h.in | 1 qemud/remote.c | 34 +++ qemud/remote_dispatch_args.h | 1 qemud/remote_dispatch_prototypes.h | 7 qemud/remote_dispatch_table.h | 5 qemud/remote_protocol.c | 13 + qemud/remote_protocol.h | 17 + qemud/remote_protocol.x | 12 + src/driver.h | 10 + src/libvirt.c | 55 ++++- src/libvirt_internal.h | 5 src/libvirt_private.syms | 1 src/lxc_driver.c | 1 src/openvz_driver.c | 1 src/qemu_driver.c | 367 ++++++++++++++++++++++++++++++++----- src/remote_internal.c | 33 +++ src/test.c | 1 src/uml_driver.c | 1 src/virsh.c | 4 src/xen_unified.c | 1 22 files changed, 518 insertions(+), 54 deletions(-)

2 1

[libvirt] Hyper-V driver advise request
by Matthias Bolte 21 May '09

21 May '09

Hello, I mentioned Hyper-V in my last mail. We are currently investigating Hyper-V support for libvirt. Basically there seem to be two options: - use the DCE/RPC library from the Samba project for Hyper-V's WMI protocol - write a Windows based Hyper-V driver and use it from a Windows based libvirt client only or port libvirtd to Windows We're currently not sure which way to go and if any of this will actually work. So, we just want to ask for advise.

2 2

[libvirt] Incomplete cleanup in storage backend for physical disks?
by Henrik Persson E 20 May '09

20 May '09

Hi all, I have some questions related to the storage backend for physical disks. If I try to create a new storage pool connected to a physical disk that doesn't have a partition table I get the following error messages: error: Failed to create pool from /root/physical_disk_pool.xml error: internal error non-zero exit status from command 2 However, the pool is still exists in inactive state. If I check the code some part of the pool is freed: cleanup: virStoragePoolDefFree(def); if (pool) { virStoragePoolObjUnlock(pool); } storageDriverUnlock(driver); return ret; but not the pool object itself. It feels like I end up with an non-complete pool. What is the story here, shouldn't the pool object also be freed in this case? And what is the suggested to create and start up a storage pool connected to a physical disk that doesn't have a partition table yet? Regards, Henrik

2 1

[libvirt] [PATCH] cleanup of direct stderr logging
by Daniel Veillard 20 May '09

20 May '09

In a number of places we still had direct output of logs, warnings or errors, instead of using the existing error and logging infrastructure. This patches tries to clean this up, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] Works: libvirt client on OS X 10.5.6
by Schley Andrew Kutz 20 May '09

20 May '09

Getting libvirt-0.6.3 (client) to compile on OS X - Use MacPorts to install gnutls (and its several dependencies) - Set environment variables: export LDFLAGS="-L/opt/local/lib" export CPPFLAGS="-I/opt/local/include" export MACOSX_DEPLOYMENT_TARGET=10.4 - Configure --prefix=/opt/libvirt/ --without-sasl --without-avahi --without-polkit --without-python --without-xen --without-qemu --without-lxc --without- openvz --without-libvirtd --without-uml - Apply patches src/pci.c #ifndef MODPROBE #define MODPROBE 0 #endif src/virsh.c:5665 if (command_ret != 0 /* WEXITSTATUS (0) */) { - Compile The MACOSX_DEPLOYMENT_TARGET variable is very important, otherwise you will get symbol errors when linking. -- -a "Ideally, a code library must be immediately usable by naive developers, easily customized by more sophisticated developers, and readily extensible by experts." -- L. Stein

3 7

[libvirt] [PATCH]: Add a flag to let live migration be persistent
by Chris Lalancette 20 May '09

20 May '09

As pointed out by Łukasz Mierzwa <l.mierzwa(a)gmail.com>, it would be nice if there was an option to automatically make a domain persistent on the destination during a live migration. The attached patch adds this simple capability. Note that this has to be applied on top of my previous secure migration patch, otherwise you'll have conflicts. Signed-off-by: Chris Lalancette <clalance(a)redhat.com>

3 5

[libvirt] PATCH: Test the QEMU argv -> XML convertor
by Daniel P. Berrange 20 May '09

20 May '09

This patch adds a test case which validates the argv -> XML convertor for QEMU. This re-uses the existing XML data files from the XML -> argv test. This has a few complications though because we can't roundtrip all the data. So I have a couple really seriously evil regexes which splice out the XML lines we know won't match - name, uuid, memory and disk sharing modes. This isn't ideal, but I figure it is better todo this and get coverage over 90% of the code, than to have no test coverage. Thanks to Cole for this testsuite suggestion Daniel diff -r 453b896e0e3f .hgignore --- a/.hgignore Tue May 19 18:43:32 2009 +0100 +++ b/.hgignore Tue May 19 18:43:50 2009 +0100 @@ -270,6 +270,7 @@ tests/object-locking tests/object-locking-files.txt tests/object-locking.cmi tests/object-locking.cmx +tests/qemuargv2xmltest tests/qemuxml2argvtest tests/qemuxml2xmltest tests/qparamtest diff -r 453b896e0e3f tests/.cvsignore --- a/tests/.cvsignore Tue May 19 18:43:32 2009 +0100 +++ b/tests/.cvsignore Tue May 19 18:43:50 2009 +0100 @@ -11,6 +11,7 @@ xmconfigtest xencapstest qemuxml2xmltest qemuxml2argvtest +qemuargv2xmltest nodedevxml2xmltest nodeinfotest statstest diff -r 453b896e0e3f tests/.gitignore --- a/tests/.gitignore Tue May 19 18:43:32 2009 +0100 +++ b/tests/.gitignore Tue May 19 18:43:50 2009 +0100 @@ -11,6 +11,7 @@ xmconfigtest xencapstest qemuxml2xmltest qemuxml2argvtest +qemuargv2xmltest nodedevxml2xmltest nodeinfotest statstest diff -r 453b896e0e3f tests/Makefile.am --- a/tests/Makefile.am Tue May 19 18:43:32 2009 +0100 +++ b/tests/Makefile.am Tue May 19 18:43:50 2009 +0100 @@ -61,7 +61,7 @@ noinst_PROGRAMS += xml2sexprtest sexpr2x reconnect xmconfigtest xencapstest endif if WITH_QEMU -noinst_PROGRAMS += qemuxml2argvtest qemuxml2xmltest +noinst_PROGRAMS += qemuxml2argvtest qemuxml2xmltest qemuargv2xmltest endif if WITH_SECDRIVER_SELINUX @@ -119,7 +119,7 @@ endif endif if WITH_QEMU -TESTS += qemuxml2argvtest qemuxml2xmltest +TESTS += qemuxml2argvtest qemuxml2xmltest qemuargv2xmltest endif if WITH_SECDRIVER_SELINUX @@ -194,8 +194,13 @@ qemuxml2xmltest_SOURCES = \ qemuxml2xmltest.c testutilsqemu.c testutilsqemu.h \ testutils.c testutils.h qemuxml2xmltest_LDADD = ../src/libvirt_driver_qemu.la $(LDADDS) + +qemuargv2xmltest_SOURCES = \ + qemuargv2xmltest.c testutilsqemu.c testutilsqemu.h \ + testutils.c testutils.h +qemuargv2xmltest_LDADD = ../src/libvirt_driver_qemu.la $(LDADDS) else -EXTRA_DIST += qemuxml2argvtest.c qemuxml2xmltest.c testutilsqemu.c testutilsqemu.h +EXTRA_DIST += qemuxml2argvtest.c qemuxml2xmltest.c qemuargv2xmltest.c testutilsqemu.c testutilsqemu.h endif nodedevxml2xmltest_SOURCES = \ diff -r 453b896e0e3f tests/qemuargv2xmltest.c --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/qemuargv2xmltest.c Tue May 19 18:43:50 2009 +0100 @@ -0,0 +1,235 @@ +#include <config.h> + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <string.h> + +#include <sys/types.h> +#include <fcntl.h> + +#ifdef WITH_QEMU + +#include "internal.h" +#include "testutils.h" +#include "qemu_conf.h" + +#include "testutilsqemu.h" + +static char *progname; +static char *abs_srcdir; +static struct qemud_driver driver; + +#define MAX_FILE 4096 + +static int blankProblemElements(char *data) +{ + if (virtTestClearLineRegex("<name>[[:alnum:]]+</name>", data) < 0 || + virtTestClearLineRegex("<uuid>([[:alnum:]]|-)+</uuid>", data) < 0 || + virtTestClearLineRegex("<memory>[[:digit:]]+</memory>", data) < 0 || + virtTestClearLineRegex("<currentMemory>[[:digit:]]+</currentMemory>", data) < 0 || + virtTestClearLineRegex("<readonly/>", data) < 0 || + virtTestClearLineRegex("<sharable/>", data) < 0) + return -1; + return 0; +} + +static int testCompareXMLToArgvFiles(const char *xml, + const char *cmdfile) { + char xmlData[MAX_FILE]; + char cmdData[MAX_FILE]; + char *expectxml = &(xmlData[0]); + char *actualxml = NULL; + char *cmd = &(cmdData[0]); + int ret = -1; + virDomainDefPtr vmdef = NULL; + + if (virtTestLoadFile(cmdfile, &cmd, MAX_FILE) < 0) + goto fail; + if (virtTestLoadFile(xml, &expectxml, MAX_FILE) < 0) + goto fail; + + if (!(vmdef = qemuParseCommandLineString(NULL, cmd))) + goto fail; + + if (!(actualxml = virDomainDefFormat(NULL, vmdef, 0))) + goto fail; + + if (blankProblemElements(expectxml) < 0 || + blankProblemElements(actualxml) < 0) + goto fail; + + if (STRNEQ(expectxml, actualxml)) { + virtTestDifference(stderr, expectxml, actualxml); + goto fail; + } + + ret = 0; + + fail: + free(actualxml); + virDomainDefFree(vmdef); + return ret; +} + + +struct testInfo { + const char *name; + int extraFlags; + const char *migrateFrom; +}; + +static int testCompareXMLToArgvHelper(const void *data) { + const struct testInfo *info = data; + char xml[PATH_MAX]; + char args[PATH_MAX]; + snprintf(xml, PATH_MAX, "%s/qemuxml2argvdata/qemuxml2argv-%s.xml", + abs_srcdir, info->name); + snprintf(args, PATH_MAX, "%s/qemuxml2argvdata/qemuxml2argv-%s.args", + abs_srcdir, info->name); + return testCompareXMLToArgvFiles(xml, args); +} + + + +static int +mymain(int argc, char **argv) +{ + int ret = 0; + char cwd[PATH_MAX]; + + progname = argv[0]; + + if (argc > 1) { + fprintf(stderr, "Usage: %s\n", progname); + return (EXIT_FAILURE); + } + + abs_srcdir = getenv("abs_srcdir"); + if (!abs_srcdir) + abs_srcdir = getcwd(cwd, sizeof(cwd)); + + if ((driver.caps = testQemuCapsInit()) == NULL) + return EXIT_FAILURE; + if((driver.stateDir = strdup("/nowhere")) == NULL) + return EXIT_FAILURE; + +#define DO_TEST_FULL(name, extraFlags, migrateFrom) \ + do { \ + const struct testInfo info = { name, extraFlags, migrateFrom }; \ + if (virtTestRun("QEMU ARGV-2-XML " name, \ + 1, testCompareXMLToArgvHelper, &info) < 0) \ + ret = -1; \ + } while (0) + +#define DO_TEST(name, extraFlags) \ + DO_TEST_FULL(name, extraFlags, NULL) + + setenv("PATH", "/bin", 1); + setenv("USER", "test", 1); + setenv("LOGNAME", "test", 1); + setenv("HOME", "/home/test", 1); + unsetenv("TMPDIR"); + unsetenv("LD_PRELOAD"); + unsetenv("LD_LIBRARY_PATH"); + + /* Can't roundtrip vcpu cpuset attribute */ + /*DO_TEST("minimal", QEMUD_CMD_FLAG_NAME);*/ + DO_TEST("boot-cdrom", 0); + DO_TEST("boot-network", 0); + DO_TEST("boot-floppy", 0); + /* Can't roundtrip xenner arch */ + /*DO_TEST("bootloader", 0);*/ + DO_TEST("clock-utc", 0); + DO_TEST("clock-localtime", 0); + DO_TEST("disk-cdrom", 0); + DO_TEST("disk-cdrom-empty", QEMUD_CMD_FLAG_DRIVE); + DO_TEST("disk-floppy", 0); + DO_TEST("disk-many", 0); + DO_TEST("disk-virtio", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_BOOT); + DO_TEST("disk-xenvbd", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_BOOT); + DO_TEST("disk-drive-boot-disk", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_BOOT); + DO_TEST("disk-drive-boot-cdrom", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_BOOT); + DO_TEST("disk-drive-fmt-qcow", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_BOOT); + /* Can't roundtrip shareable+cache mode option */ + /*DO_TEST("disk-drive-shared", QEMUD_CMD_FLAG_DRIVE);*/ + /* Can't roundtrip v1 writethrough option */ + /*DO_TEST("disk-drive-cache-v1-wt", QEMUD_CMD_FLAG_DRIVE);*/ + DO_TEST("disk-drive-cache-v1-wb", QEMUD_CMD_FLAG_DRIVE); + DO_TEST("disk-drive-cache-v1-none", QEMUD_CMD_FLAG_DRIVE); + DO_TEST("disk-drive-cache-v2-wt", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_CACHE_V2); + DO_TEST("disk-drive-cache-v2-wb", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_CACHE_V2); + DO_TEST("disk-drive-cache-v2-none", QEMUD_CMD_FLAG_DRIVE | + QEMUD_CMD_FLAG_DRIVE_CACHE_V2); + DO_TEST("disk-usb", 0); + DO_TEST("graphics-vnc", 0); + + driver.vncSASL = 1; + driver.vncSASLdir = strdup("/root/.sasl2"); + DO_TEST("graphics-vnc-sasl", 0); + driver.vncTLS = 1; + driver.vncTLSx509verify = 1; + driver.vncTLSx509certdir = strdup("/etc/pki/tls/qemu"); + DO_TEST("graphics-vnc-tls", 0); + driver.vncSASL = driver.vncTLSx509verify = driver.vncTLS = 0; + free(driver.vncSASLdir); + free(driver.vncTLSx509certdir); + driver.vncSASLdir = driver.vncTLSx509certdir = NULL; + + DO_TEST("graphics-sdl", 0); + DO_TEST("graphics-sdl-fullscreen", 0); + DO_TEST("input-usbmouse", 0); + DO_TEST("input-usbtablet", 0); + /* Can't rountrip xenner arch */ + /*DO_TEST("input-xen", 0);*/ + DO_TEST("misc-acpi", 0); + DO_TEST("misc-no-reboot", 0); + DO_TEST("misc-uuid", QEMUD_CMD_FLAG_NAME | + QEMUD_CMD_FLAG_UUID | QEMUD_CMD_FLAG_DOMID); + DO_TEST("net-user", 0); + DO_TEST("net-virtio", 0); + DO_TEST("net-eth", 0); + DO_TEST("net-eth-ifname", 0); + + DO_TEST("serial-vc", 0); + DO_TEST("serial-pty", 0); + DO_TEST("serial-dev", 0); + DO_TEST("serial-file", 0); + DO_TEST("serial-unix", 0); + DO_TEST("serial-tcp", 0); + DO_TEST("serial-udp", 0); + DO_TEST("serial-tcp-telnet", 0); + DO_TEST("serial-many", 0); + DO_TEST("parallel-tcp", 0); + DO_TEST("console-compat", 0); + DO_TEST("sound", 0); + + DO_TEST("hostdev-usb-product", 0); + DO_TEST("hostdev-usb-address", 0); + + DO_TEST("hostdev-pci-address", 0); + + DO_TEST_FULL("restore-v1", QEMUD_CMD_FLAG_MIGRATE_KVM_STDIO, "stdio"); + DO_TEST_FULL("restore-v2", QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC, "stdio"); + DO_TEST_FULL("restore-v2", QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC, "exec:cat"); + DO_TEST_FULL("migrate", QEMUD_CMD_FLAG_MIGRATE_QEMU_TCP, "tcp:10.0.0.1:5000"); + + virCapabilitiesFree(driver.caps); + + return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); +} + +VIRT_TEST_MAIN(mymain) + +#else + +int main (void) { return (77); /* means 'test skipped' for automake */ } + +#endif /* WITH_QEMU */ diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-bootloader.args --- a/tests/qemuxml2argvdata/qemuxml2argv-bootloader.args Tue May 19 18:43:32 2009 +0100 +++ b/tests/qemuxml2argvdata/qemuxml2argv-bootloader.args Tue May 19 18:43:50 2009 +0100 @@ -1,1 +1,1 @@ -LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu-kvm -S -M xenner -m 214 -smp 1 -nographic -monitor pty -no-acpi -bootloader /usr/bin/pygrub -cdrom /dev/cdrom -net none -serial none -parallel none -usb +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/xenner -S -M xenner -m 214 -smp 1 -nographic -monitor pty -no-acpi -bootloader /usr/bin/pygrub -cdrom /dev/cdrom -net none -serial none -parallel none -usb diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml --- a/tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml Tue May 19 18:43:32 2009 +0100 +++ b/tests/qemuxml2argvdata/qemuxml2argv-bootloader.xml Tue May 19 18:43:50 2009 +0100 @@ -13,7 +13,7 @@ <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> - <emulator>/usr/bin/qemu-kvm</emulator> + <emulator>/usr/bin/xenner</emulator> <disk type='block' device='cdrom'> <source dev='/dev/cdrom'/> <target dev='hdc' bus='ide'/> diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml --- a/tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml Tue May 19 18:43:32 2009 +0100 +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-cdrom-empty.xml Tue May 19 18:43:50 2009 +0100 @@ -24,4 +24,3 @@ </disk> </devices> </domain> - diff -r 453b896e0e3f tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml --- a/tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml Tue May 19 18:43:32 2009 +0100 +++ b/tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml Tue May 19 18:43:50 2009 +0100 @@ -18,7 +18,7 @@ <source dev='/dev/HostVG/QEMUGuest2'/> <target dev='hda' bus='ide'/> </disk> - <hostdev mode='subsystem' type='pci' managed='no'> + <hostdev mode='subsystem' type='pci' managed='yes'> <source> <address domain='0x0000' bus='0x06' slot='0x12' function='0x5'/> </source> diff -r 453b896e0e3f tests/testutils.c --- a/tests/testutils.c Tue May 19 18:43:32 2009 +0100 +++ b/tests/testutils.c Tue May 19 18:43:50 2009 +0100 @@ -18,6 +18,7 @@ #ifndef WIN32 #include <sys/wait.h> #endif +#include <regex.h> #include <unistd.h> #include <string.h> #include <fcntl.h> @@ -457,3 +458,48 @@ cleanup: virResetLastError(); return ret; } + + +int virtTestClearLineRegex(const char *pattern, + char *str) +{ + regex_t reg; + char *lineStart = str; + char *lineEnd = strchr(str, '\n'); + + if (regcomp(&reg, pattern, REG_EXTENDED | REG_NOSUB) != 0) + return -1; + + while (lineStart) { + int ret; + if (lineEnd) + *lineEnd = '\0'; + + + ret = regexec(&reg, lineStart, 0, NULL, 0); + //fprintf(stderr, "Match %d '%s' '%s'\n", ret, lineStart, pattern); + if (ret == 0) { + if (lineEnd) { + memmove(lineStart, lineEnd + 1, strlen(lineEnd+1) + 1); + /* Don't update lineStart - just iterate again on this + location */ + lineEnd = strchr(lineStart, '\n'); + } else { + *lineStart = '\0'; + lineStart = NULL; + } + } else { + if (lineEnd) { + *lineEnd = '\n'; + lineStart = lineEnd + 1; + lineEnd = strchr(lineStart, '\n'); + } else { + lineStart = NULL; + } + } + } + + regfree(&reg); + + return 0; +} diff -r 453b896e0e3f tests/testutils.h --- a/tests/testutils.h Tue May 19 18:43:32 2009 +0100 +++ b/tests/testutils.h Tue May 19 18:43:50 2009 +0100 @@ -27,6 +27,8 @@ int virtTestCaptureProgramOutput(const c char **buf, int buflen); +int virtTestClearLineRegex(const char *pattern, + char *string); int virtTestDifference(FILE *stream, const char *expect, -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] kvm migration
by Łukasz Mierzwa 20 May '09

20 May '09

Hi, can someone explain me how migration works in libvirt? I got 2 machines with ubuntu 9.04 (libvirt 0.6.1 and kvm 84), disk images are stored on nfs share so both machines can access them. When I run live migration (virsh migrate --live domain uri) my domain is migrated to second host but: 1. domain is defined on both hosts after migration, it does not disappear from original host 2. when I shutdown this domain on second host it gets undefined so I guess that migration is only temporary but this is not what I want. When I migrate domain to other host I want it to stay there, how can I do that? I can't find anything about it in documentation on libvirt website, google does not seem to know anything either. Łukasz Mierzwa

3 5