- Devel - libvirt List Archives

[libvirt] printf()'s in virsh?
by Justin Clift 15 Jun '10

15 Jun '10

Hi guys, In virsh, in the cmdSchedinfo() function, there's a block of code using printf instead of vshPrintf(): for (i = 0; i < nparams; i++){ switch (params[i].type) { case VIR_DOMAIN_SCHED_FIELD_INT: printf("%-15s: %d\n", params[i].field, params[i].value.i); break; case VIR_DOMAIN_SCHED_FIELD_UINT: printf("%-15s: %u\n", params[i].field, params[i].value.ui); break; case VIR_DOMAIN_SCHED_FIELD_LLONG: printf("%-15s: %lld\n", params[i].field, params[i].value.l); break; case VIR_DOMAIN_SCHED_FIELD_ULLONG: printf("%-15s: %llu\n", params[i].field, params[i].value.ul); break; case VIR_DOMAIN_SCHED_FIELD_DOUBLE: printf("%-15s: %f\n", params[i].field, params[i].value.d); break; case VIR_DOMAIN_SCHED_FIELD_BOOLEAN: printf("%-15s: %d\n", params[i].field, params[i].value.b); break; default: printf("not implemented scheduler parameter type\n"); } } These should be vshPrintf() shouldn't they? Regards and best wishes, Justin Clift -- Salasaga - Open Source eLearning IDE http://www.salasaga.org

1 0

[libvirt] [PATCH] [RFC] nwfilter: fix loadable module support
by Stefan Berger 15 Jun '10

15 Jun '10

Hello! I am trying to fix the loadable modules support by removing all direct calls from qemu_driver.c and qemu_conf.c into nwfilter functions. The below patch extends the nwfilter driver interface with 2 more private functions to instantiate and tear down the filters. I call them 'private' because no client should ever be calling them via RPC and be able to tear down the filters for example -- they should only ever be callable from within libvirtd. I extended the qemudShutdownVMDaemon function with the virConnectPtr type of parameter so that I have the pointer to conn->nwfilterDriver to access the private nwfilter driver functions. The problem with that is that the conn pointer is not always available to be passed (qemuReconnectDomain, qemuHandleMonitorEOF), particularly in tear-down functions, and so I have to pass a NULL pointer, which then prevents the calling of the private interface function to tear down the filters.Does anyone have a suggestion on how I would best change the interface with the nwfilter driver to fix this particular problem? Stefan Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com> --- src/esx/esx_nwfilter_driver.c | 10 ++++++++ src/nwfilter/nwfilter_driver.c | 15 ++++++++++++ src/nwfilter/nwfilter_gentech_driver.h | 26 ++++++++++++++++----- src/qemu/qemu_conf.c | 10 ++++---- src/qemu/qemu_driver.c | 40 ++++++++++++++++++--------------- 5 files changed, 72 insertions(+), 29 deletions(-) Index: libvirt-acl/src/nwfilter/nwfilter_driver.c =================================================================== --- libvirt-acl.orig/src/nwfilter/nwfilter_driver.c +++ libvirt-acl/src/nwfilter/nwfilter_driver.c @@ -410,6 +410,19 @@ cleanup: } +static int +nwfilterInstantiateFilter(virConnectPtr conn, + virDomainNetDefPtr net) { + return virNWFilterInstantiateFilter(conn, net); +} + + +static void +nwfilterTeardownFilter(virDomainNetDefPtr net) { + virNWFilterTeardownFilter(net); +} + + static virNWFilterDriver nwfilterDriver = { .name = "nwfilter", .open = nwfilterOpen, @@ -421,6 +434,8 @@ static virNWFilterDriver nwfilterDriver .defineXML = nwfilterDefine, .undefine = nwfilterUndefine, .getXMLDesc = nwfilterDumpXML, + .privateInstantiateFilter = nwfilterInstantiateFilter, + .privateTeardownFilter = nwfilterTeardownFilter, }; Index: libvirt-acl/src/esx/esx_nwfilter_driver.c =================================================================== --- libvirt-acl.orig/src/esx/esx_nwfilter_driver.c +++ libvirt-acl/src/esx/esx_nwfilter_driver.c @@ -75,6 +75,8 @@ static virNWFilterDriver esxNWFilterDriv NULL, /* defineXML */ NULL, /* undefine */ NULL, /* getXMLDesc */ + NULL, /* privateInstantiateFilter */ + NULL, /* privateTeardownFilter */ }; Index: libvirt-acl/src/nwfilter/nwfilter_gentech_driver.h =================================================================== --- libvirt-acl.orig/src/nwfilter/nwfilter_gentech_driver.h +++ libvirt-acl/src/nwfilter/nwfilter_gentech_driver.h @@ -23,6 +23,8 @@ #ifndef __NWFILTER_GENTECH_DRIVER_H # define __NWFILTER_GENTECH_DRIVER_H +# include "datatypes.h" + virNWFilterTechDriverPtr virNWFilterTechDriverForName(const char *name); int virNWFilterRuleInstAddData(virNWFilterRuleInstPtr res, @@ -37,8 +39,18 @@ enum instCase { }; +static inline +int virNWFilterInstNWFilter(virConnectPtr conn, + const virDomainNetDefPtr net) +{ + if ((conn) && (conn->nwfilterDriver)) + return conn->nwfilterDriver->privateInstantiateFilter(conn, net); + return 0; +} + int virNWFilterInstantiateFilter(virConnectPtr conn, const virDomainNetDefPtr net); + int virNWFilterUpdateInstantiateFilter(virConnectPtr conn, const virDomainNetDefPtr net, bool *skipIface); @@ -70,18 +82,20 @@ void virNWFilterDomainFWUpdateCB(void *p /* tear down an interface's filter before tearing down the interface */ static inline void -virNWFilterTearNWFilter(virDomainNetDefPtr net) { - if ((net->filter) && (net->ifname)) - virNWFilterTeardownFilter(net); +virNWFilterTearNWFilter(virConnectPtr conn, virDomainNetDefPtr net) { + if ((conn) && (conn->nwfilterDriver) && (net->filter) && (net->ifname)) + conn->nwfilterDriver->privateTeardownFilter(net); } static inline void -virNWFilterTearVMNWFilters(virDomainObjPtr vm) { +virNWFilterTearVMNWFilters(virConnectPtr conn, virDomainObjPtr vm) { int i; - for (i = 0; i < vm->def->nnets; i++) - virNWFilterTearNWFilter(vm->def->nets[i]); + if (conn) { + for (i = 0; i < vm->def->nnets; i++) + virNWFilterTearNWFilter(conn, vm->def->nets[i]); + } } #endif Index: libvirt-acl/src/qemu/qemu_conf.c =================================================================== --- libvirt-acl.orig/src/qemu/qemu_conf.c +++ libvirt-acl/src/qemu/qemu_conf.c @@ -1555,7 +1555,7 @@ qemudPhysIfaceConnect(virConnectPtr conn if (rc >= 0) { if ((net->filter) && (net->ifname)) { - err = virNWFilterInstantiateFilter(conn, net); + err = virNWFilterInstNWFilter(conn, net); if (err) { close(rc); rc = -1; @@ -1688,7 +1688,7 @@ qemudNetworkIfaceConnect(virConnectPtr c if (tapfd >= 0) { if ((net->filter) && (net->ifname)) { - err = virNWFilterInstantiateFilter(conn, net); + err = virNWFilterInstNWFilter(conn, net); if (err) { close(tapfd); tapfd = -1; @@ -4207,7 +4207,7 @@ int qemudBuildCommandLine(virConnectPtr goto error; if (VIR_REALLOC_N(*vmfds, (*nvmfds)+1) < 0) { - virNWFilterTearNWFilter(net); + virNWFilterTearNWFilter(conn, net); close(tapfd); goto no_memory; } @@ -4226,7 +4226,7 @@ int qemudBuildCommandLine(virConnectPtr goto error; if (VIR_REALLOC_N(*vmfds, (*nvmfds)+1) < 0) { - virNWFilterTearNWFilter(net); + virNWFilterTearNWFilter(conn, net); close(tapfd); goto no_memory; } @@ -4766,7 +4766,7 @@ int qemudBuildCommandLine(virConnectPtr virReportOOMError(); error: for (i = 0; i <= last_good_net; i++) - virNWFilterTearNWFilter(def->nets[i]); + virNWFilterTearNWFilter(conn, def->nets[i]); if (vmfds && *vmfds) { for (i = 0; i < *nvmfds; i++) Index: libvirt-acl/src/qemu/qemu_driver.c =================================================================== --- libvirt-acl.orig/src/qemu/qemu_driver.c +++ libvirt-acl/src/qemu/qemu_driver.c @@ -158,7 +158,8 @@ static int qemudStartVMDaemon(virConnect int stdin_fd, const char *stdin_path); -static void qemudShutdownVMDaemon(struct qemud_driver *driver, +static void qemudShutdownVMDaemon(virConnectPtr conn, + struct qemud_driver *driver, virDomainObjPtr vm, int migrated); @@ -735,7 +736,7 @@ qemuHandleMonitorEOF(qemuMonitorPtr mon VIR_DOMAIN_EVENT_STOPPED_FAILED : VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN); - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(NULL, driver, vm, 0); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); else @@ -1283,7 +1284,7 @@ error: /* We can't get the monitor back, so must kill the VM * to remove danger of it ending up running twice if * user tries to start it again later */ - qemudShutdownVMDaemon(driver, obj, 0); + qemudShutdownVMDaemon(NULL, driver, obj, 0); if (!obj->persistent) virDomainRemoveInactive(&driver->domains, obj); else @@ -3576,7 +3577,7 @@ static int qemudStartVMDaemon(virConnect VIR_FREE(progenv); if (ret == -1) /* The VM failed to start; tear filters before taps */ - virNWFilterTearVMNWFilters(vm); + virNWFilterTearVMNWFilters(conn, vm); if (vmfds) { for (i = 0 ; i < nvmfds ; i++) { @@ -3642,7 +3643,7 @@ cleanup: /* We jump here if we failed to start the VM for any reason, or * if we failed to initialize the now running VM. kill it off and * pretend we never started it */ - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(conn, driver, vm, 0); if (logfile != -1) close(logfile); @@ -3651,7 +3652,8 @@ cleanup: } -static void qemudShutdownVMDaemon(struct qemud_driver *driver, +static void qemudShutdownVMDaemon(virConnectPtr conn, + struct qemud_driver *driver, virDomainObjPtr vm, int migrated) { int ret; @@ -3668,7 +3670,7 @@ static void qemudShutdownVMDaemon(struct * reporting so we don't squash a legit error. */ orig_err = virSaveLastError(); - virNWFilterTearVMNWFilters(vm); + virNWFilterTearVMNWFilters(conn, vm); if (driver->macFilter) { def = vm->def; @@ -4465,7 +4467,7 @@ static int qemudDomainDestroy(virDomainP goto endjob; } - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(dom->conn, driver, vm, 0); event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_STOPPED, VIR_DOMAIN_EVENT_STOPPED_DESTROYED); @@ -5211,7 +5213,7 @@ static int qemudDomainSaveFlag(virDomain ret = 0; /* Shut it down */ - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(dom->conn, driver, vm, 0); event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_STOPPED, VIR_DOMAIN_EVENT_STOPPED_SAVED); @@ -5528,7 +5530,7 @@ static int qemudDomainCoreDump(virDomain endjob: if ((ret == 0) && (flags & VIR_DUMP_CRASH)) { - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(dom->conn, driver, vm, 0); event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_STOPPED, VIR_DOMAIN_EVENT_STOPPED_CRASHED); @@ -7640,7 +7642,7 @@ cleanup: VIR_WARN0("Unable to release PCI address on NIC"); if (ret != 0) - virNWFilterTearNWFilter(net); + virNWFilterTearNWFilter(conn, net); VIR_FREE(nicstr); VIR_FREE(netstr); @@ -8534,7 +8536,8 @@ cleanup: } static int -qemudDomainDetachNetDevice(struct qemud_driver *driver, +qemudDomainDetachNetDevice(virConnectPtr conn, + struct qemud_driver *driver, virDomainObjPtr vm, virDomainDeviceDefPtr dev, unsigned long long qemuCmdFlags) @@ -8609,7 +8612,7 @@ qemudDomainDetachNetDevice(struct qemud_ } qemuDomainObjExitMonitorWithDriver(driver, vm); - virNWFilterTearNWFilter(detach); + virNWFilterTearNWFilter(conn, detach); #if WITH_MACVTAP if (detach->type == VIR_DOMAIN_NET_TYPE_DIRECT) { @@ -8921,7 +8924,8 @@ static int qemudDomainDetachDevice(virDo _("This type of disk cannot be hot unplugged")); } } else if (dev->type == VIR_DOMAIN_DEVICE_NET) { - ret = qemudDomainDetachNetDevice(driver, vm, dev, qemuCmdFlags); + ret = qemudDomainDetachNetDevice(dom->conn, driver, vm, dev, + qemuCmdFlags); } else if (dev->type == VIR_DOMAIN_DEVICE_CONTROLLER) { if (dev->data.controller->type == VIR_DOMAIN_CONTROLLER_TYPE_SCSI) { ret = qemudDomainDetachPciControllerDevice(driver, vm, dev, @@ -10215,7 +10219,7 @@ qemudDomainMigratePrepareTunnel(virConne qemust = qemuStreamMigOpen(st, unixfile); if (qemust == NULL) { - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(dconn, driver, vm, 0); if (!vm->persistent) { if (qemuDomainObjEndJob(vm) > 0) virDomainRemoveInactive(&driver->domains, vm); @@ -10940,7 +10944,7 @@ qemudDomainMigratePerform (virDomainPtr } /* Clean up the source domain. */ - qemudShutdownVMDaemon(driver, vm, 1); + qemudShutdownVMDaemon(dom->conn, driver, vm, 1); resume = 0; event = virDomainEventNewFromObj(vm, @@ -11100,7 +11104,7 @@ qemudDomainMigrateFinish2 (virConnectPtr goto endjob; } } else { - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(dconn, driver, vm, 0); event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_STOPPED, VIR_DOMAIN_EVENT_STOPPED_FAILED); @@ -11946,7 +11950,7 @@ static int qemuDomainRevertToSnapshot(vi */ if (virDomainObjIsActive(vm)) { - qemudShutdownVMDaemon(driver, vm, 0); + qemudShutdownVMDaemon(snapshot->domain->conn, driver, vm, 0); event = virDomainEventNewFromObj(vm, VIR_DOMAIN_EVENT_STOPPED, VIR_DOMAIN_EVENT_STOPPED_FROM_SNAPSHOT); Index: libvirt-acl/src/driver.h =================================================================== --- libvirt-acl.orig/src/driver.h +++ libvirt-acl/src/driver.h @@ -1098,6 +1098,14 @@ typedef char * (*virDrvNWFilterGetXMLDesc) (virNWFilterPtr pool, unsigned int flags); +# include "conf/domain_conf.h" + +typedef int + (*virDrvNWFilterInstantiateFilter) (virConnectPtr conn, + const virDomainNetDefPtr net); + +typedef void + (*virDrvNWFilterTeardownFilter) (virDomainNetDefPtr net); typedef struct _virNWFilterDriver virNWFilterDriver; typedef virNWFilterDriver *virNWFilterDriverPtr; @@ -1124,6 +1132,8 @@ struct _virNWFilterDriver { virDrvNWFilterDefineXML defineXML; virDrvNWFilterUndefine undefine; virDrvNWFilterGetXMLDesc getXMLDesc; + virDrvNWFilterInstantiateFilter privateInstantiateFilter; + virDrvNWFilterTeardownFilter privateTeardownFilter; };

2 1

[libvirt] [PATCH] Fix enumeration of partitions in disks with a trailing digit in path
by Daniel P. Berrange 15 Jun '10

15 Jun '10

Disks with a trailing digit in their path (eg /dev/loop0 or /dev/dm0) have an extra 'p' appended before the partition number (eg, to form /dev/loop0p1 not /dev/loop01). Fix the partition lookup to append this extra 'p' when required * src/storage/parthelper.c: Add a 'p' before partition number if required --- src/storage/parthelper.c | 15 +++++++++++---- 1 files changed, 11 insertions(+), 4 deletions(-) diff --git a/src/storage/parthelper.c b/src/storage/parthelper.c index 5626cd2..28d88c9 100644 --- a/src/storage/parthelper.c +++ b/src/storage/parthelper.c @@ -36,6 +36,8 @@ #include <stdio.h> #include <string.h> +#include "c-ctype.h" + /* we don't need to include the full internal.h just for this */ #define STREQ(a,b) (strcmp(a,b) == 0) @@ -56,6 +58,8 @@ int main(int argc, char **argv) PedDisk *disk; PedPartition *part; int cmd = DISK_LAYOUT; + const char *path; + const char *partsep; if (argc == 3 && STREQ(argv[2], "-g")) { cmd = DISK_GEOMETRY; @@ -64,8 +68,11 @@ int main(int argc, char **argv) return 1; } - if ((dev = ped_device_get(argv[1])) == NULL) { - fprintf(stderr, "unable to access device %s\n", argv[1]); + path = argv[1]; + partsep = c_isdigit(path[strlen(path)-1]) ? "p" : ""; + + if ((dev = ped_device_get(path)) == NULL) { + fprintf(stderr, "unable to access device %s\n", path); return 2; } @@ -117,8 +124,8 @@ int main(int argc, char **argv) * in bytes, not the last sector number */ if (part->num != -1) { - printf("%s%d%c%s%c%s%c%llu%c%llu%c%llu%c", - part->geom.dev->path, + printf("%s%s%d%c%s%c%s%c%llu%c%llu%c%llu%c", + path, partsep, part->num, '\0', type, '\0', content, '\0', -- 1.6.5.2

3 8

[libvirt] Error-message when creating virtual mashin
by Sebastian Andres 14 Jun '10

14 Jun '10

Hello. When I creating virtual Mashines with virt-install (Version 0.500.3-1) with following options: # virt-install --serial pty --hvm --virt-type kvm -l http://ftp.de.debian.org/debian/dists/lenny/main/installer-i386/ -x console=ttyS0,9600 -w bridge=virt0,mac=52:54:00:12:34:29,model=virtio -k=de --nographics --prompt --disk=test.img,bus=virtio I get the following error messages: Creating domain... 0 B 00:00 error: Unable to get domain status error: operation failed: could not query memory balloon allocation Domain installation does not appear to have been successful. If it was, you can restart your domain by running 'virsh start test'; otherwise, please restart your installation. I have 4 gb Ram and 2 gb free. I am using libvirt-bin 0.8.1-2 and run all Commands as root. Can anywhere help? Best regards Sebastian -- Sebastian Andres mail: <sebastian(a)sebastianandres.de> Olbersstr. 33, 10589 Berlin google maps: http://tinyurl.com/6z3f3 home:+49-30-3010.7852 www: http://www.sebastianandres.de cell:+49-170-3224.773 pgp: http://www.sebastianandres.de/key.txt

2 1

[libvirt] About cgroup mechanism using in libvirt
by Alex Jia 14 Jun '10

14 Jun '10

Hey Daniel, The cgroup mechanism have been integrated into libvirt for LXC and QEMU driver, and the LXC driver uses all of cgroup controllers except for net_cls and cpuset, while the QEMU driver only uses the cpu and devices controllers at present. >From the user point of view, user can use some virsh commands to control some guest resources: 1. Using 'virsh schedinfo' command to get/set CPU scheduler priority for a guest 2. Using 'virsh vcpuin' command to control guest vcpu affinity 3. Using 'virsh setmem' command to change memory allocation 4. Using 'virsh setmaxmem' command to change maximum memory limit 5. Using 'virsh setvcpus' command to change number of virtual CPUs I just make sure the above 1 using CPU scheduler controller, maybe 4 using Memory controller? and maybe 5 using CPU set controller? I am not sure. And I wonder how to control devices access via virsh command or libvirt binding API such as python binding? in addition, for CPU accounting and Freezer controller, how to use them to control guest resource from libvirt application layer? and how to check setting result is valid such as cpuacct? these issues let me confuse at recent. Any comments and suggestions are welcomed, thanks for your help. Best Regards, Alex

3 3

Re: [libvirt] Virtual disk size associated with a Domain
by IKI-サガル　バルウェ 14 Jun '10

14 Jun '10

Thanks Daniel, > In 0.8.1 we added a virDomainGetBlockInfo API that will give you this > data directly. I am currently using libvirt 0.6.3. Is there a way to do it in this version. If it is not at all possible in 0.6.3, and only possible from 0.8.1 onwards, then I will upgrade to 0.8.1. Because for the version upgrade, I will have to clearly explain to my client the reason why we need to upgrade. So, If I can fetch the same info as "virDomainGetBlockInfo" in 0.6.3 in some other way, that would be just awesome. Or if I can calculate the disk size from any other parameters, then please guide me on that also. Please help. Thanks and Regards Sagar Barve ----- Original Message ----- From: "Daniel P. Berrange" <berrange(a)redhat.com> To: "IKI-サガル　バルウェ" <sagar.barve(a)ikic.co.jp> Cc: <libvir-list(a)redhat.com> Sent: Friday, June 11, 2010 6:43 PM Subject: Re: [libvirt] Virtual disk size associated with a Domain > On Fri, Jun 11, 2010 at 06:09:19PM +0900, IKI-サガル　バルウェ wrote: >> Hello, >> >> Is there any way to fetch the size of all the virtual disks associated >> with >> a Domain?? >> >> I could not find one in the libvirt API., I tried using 'virStorageVol' >> and >> 'virStoragePool', but they returned null objects. >> Also I fetched 'virDomainBlockStats', but did not understand the >> information >> returned. i.e. 'rd_bytes' and 'wr_bytes'. Can we use this info for >> calculating the disk sizes? > > In 0.8.1 we added a virDomainGetBlockInfo API that will give you this > data directly. > >> Also, What does the size of the "img" file for a Domain represent? Is it >> the >> disc size? > > virDomainGetBlockInfo gives both several sizes - see the header file > comments for a description of what each means. > > > Daniel > -- > |: Red Hat, Engineering, London -o- > http://people.redhat.com/berrange/ :| > |: http://libvirt.org -o- http://virt-manager.org -o- > http://deltacloud.org :| > |: http://autobuild.org -o- > http://search.cpan.org/~danberr/ :| > |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B > 9505 :|

1 0

[libvirt] Create a copy of a Domain
by IKI-サガル　バルウェ 14 Jun '10

14 Jun '10

Hello All, I am trying to create a Domain/VM copy program. In this program, a completely new domain should be created from a source domain. What I tried so far is this: 1. Get the XML Desc of the source Domain using "virDomainGetXMLDesc". 2. Get the Hypervisor Object for the source Domain using "virDomainGetConnect". 3. Create a new Domain using the "virDomainCreateXML" method. After doing this, I get the following error: libvirt.libvirtError: POST operation failed: xend_post: error from xen daemon: (xend.err "Error creating domain: VM name already in use by domain 4") First of all, Is this the correct way to create a new Domain as a copy of the original?? If not, please tell me how should I go about this?? Thanks and Regards Sagar Barve

3 5

[libvirt] Don't add iptables rules when creating networks
by Felix Schwarz 13 Jun '10

13 Jun '10

Hi, I just found out that libvirt always add some iptables rules if it creates a natted (or routed) network. There were a couple of mailing list posts about this so I'm pretty sure this is not news to you. I don't want to go into the debate if your approach is sensible or not (I guess there are some use cases where I kind of like it). However on my server machine I really need full control over my (rather complicated) firewall settings. Currently the newly added rules really create a lot of problems for me. For example if I manage to have a good configuration after startup and then start a libvirt network afterwards, it will inject its rules at the start of the FORWARD queue (even though the same parameters are already present at the end!). On every net start there will be more duplicated rules and they will take preference over my existing rules. Besides that specific issue I think this is only one tiny problem compared to others (central configuration of firewall rules, auditing requirements, ...). Therefore I would like to have some kind 'power user' flag that prevents libvirt from adding any filter rules. I'm fine with activating it manually as long as I don't have to patch libvirt. fs

3 6

[libvirt] PATCH: support virtual disks
by Andrew Farmer 12 Jun '10

12 Jun '10

Libvirt currently makes the assumption that disk devices are always accessible as files on the local system. However, certain types of virtual storage on qemu (e.g, NBD) may not be. This patch defines a new "virtual" disk device which is treated similarly to existing disk types, but which is exempted from at least one code path (qemuSecurityDACSetSecurityAllLabel) which assumes the disk to be accessible as a file. With this patch in place, a QEMU virtual disk can be declared as: <disk type='virtual' device='disk'> <driver name='qemu' type='nbd' /> <source path='nbd:nbdhost:1234' /> <target dev='vda' bus='virtio' /> </disk> I haven't tested this with any drivers besides QEMU, so there might be some adjustments needed for those as well. Having the infrastructure in place to create these devices makes it easier to add them in the future, though. --- diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 64b5cf3..b753734 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -103,7 +103,8 @@ VIR_ENUM_IMPL(virDomainDeviceAddress, VIR_DOMAIN_DEVICE_ADDRESS_TYPE_LAST, VIR_ENUM_IMPL(virDomainDisk, VIR_DOMAIN_DISK_TYPE_LAST, "block", "file", - "dir") + "dir", + "virtual") VIR_ENUM_IMPL(virDomainDiskDevice, VIR_DOMAIN_DISK_DEVICE_LAST, "disk", @@ -1434,6 +1435,9 @@ virDomainDiskDefParseXML(xmlNodePtr node, case VIR_DOMAIN_DISK_TYPE_DIR: source = virXMLPropString(cur, "dir"); break; + case VIR_DOMAIN_DISK_TYPE_VIRTUAL: + source = virXMLPropString(cur, "path"); + break; default: virDomainReportError(VIR_ERR_INTERNAL_ERROR, _("unexpected disk type %s"), diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index f83de83..9bb5073 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -119,6 +119,7 @@ enum virDomainDiskType { VIR_DOMAIN_DISK_TYPE_BLOCK, VIR_DOMAIN_DISK_TYPE_FILE, VIR_DOMAIN_DISK_TYPE_DIR, + VIR_DOMAIN_DISK_TYPE_VIRTUAL, VIR_DOMAIN_DISK_TYPE_LAST }; diff --git a/src/qemu/qemu_security_dac.c b/src/qemu/qemu_security_dac.c index ffc9b8d..69c7bf1 100644 --- a/src/qemu/qemu_security_dac.c +++ b/src/qemu/qemu_security_dac.c @@ -374,7 +374,7 @@ qemuSecurityDACSetSecurityAllLabel(virDomainObjPtr vm, const char *stdin_path AT for (i = 0 ; i < vm->def->ndisks ; i++) { /* XXX fixme - we need to recursively label the entriy tree :-( */ - if (vm->def->disks[i]->type == VIR_DOMAIN_DISK_TYPE_DIR) + if (vm->def->disks[i]->type == VIR_DOMAIN_DISK_TYPE_DIR || vm->def->disks[i]->type == VIR_DOMAIN_DISK_TYPE_VIRTUAL) continue; if (qemuSecurityDACSetSecurityImageLabel(vm, vm->def->disks[i]) < 0) return -1;

3 4

[libvirt] [PATCH] Improve error message for disabled client-side drivers
by Matthias Bolte 12 Jun '10

12 Jun '10

Report that libvirt was built without that driver instead of trying to connect to a libvirtd, when we know that this is going to fail. --- I had this on my todo list for a while now, because multiple people on the mailing list and on IRC reported problems that can be summarized as: "I just built/installed libvirt and want to connect to an ESX server, but libvirt complains about missing certificates or wants to connect to a non-existing libvirtd on the ESX server. I don't get it..." The problem was always the same: libvirt built without the ESX driver. But people don't get that, because libvirt reported cryptic errors in that situation. I decided to fix this now because the problem was reported again on IRC today. src/libvirt.c | 28 ++++++++++++++++++++++++++++ 1 files changed, 28 insertions(+), 0 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index 2754fd0..2487b82 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -1210,6 +1210,34 @@ do_open (const char *name, ret->flags = flags & VIR_CONNECT_RO; for (i = 0; i < virDriverTabCount; i++) { + /* We're going to probe the remote driver next. So we have already + * probed all other client-side-only driver before, but none of them + * accepted the URI. + * If the scheme corresponds to a known but disabled client-side-only + * driver then report a useful error, instead of a cryptic one about + * not being able to connect to libvirtd or not being able to find + * certificates. */ + if (virDriverTab[i]->no == VIR_DRV_REMOTE && + ret->uri != NULL && ret->uri->scheme != NULL && + ( +# ifndef WITH_PHYP + STRCASEEQ(ret->uri->scheme, "phyp") || +# endif +# ifndef WITH_ESX + STRCASEEQ(ret->uri->scheme, "esx") || + STRCASEEQ(ret->uri->scheme, "gsx") || +# endif +# ifndef WITH_XENAPI + STRCASEEQ(ret->uri->scheme, "xenapi") || +# endif + false)) { + virReportErrorHelper(NULL, VIR_FROM_NONE, VIR_ERR_INVALID_ARG, + __FILE__, __FUNCTION__, __LINE__, + _("libvirt was built without the '%s' driver"), + ret->uri->scheme); + goto failed; + } + DEBUG("trying driver %d (%s) ...", i, virDriverTab[i]->name); res = virDriverTab[i]->open (ret, auth, flags); -- 1.7.0.4

3 6