May 2014 - Devel - libvirt List Archives

[libvirt] Release of libvirt-1.2.4
by Daniel Veillard 04 May '14

04 May '14

As planned I tagged the release in git and pushed the tarball and signed rpms to the usual place: ftp://libvirt.org/libvirt/ I also generated those for python too even if the only change is a spec file update, they are in the python subdir. This is a smaller than average release, with most of the changes on code refactoring, portability and bug fixes, so no big features here but general improvements: Documentation: - Device{Attach,Detach}: Document S4 limitations (Michal Privoznik) - Add a new example to illustrate domain migration (Sahid Orentino Ferdjaoui) - update docs for setting the QEMU BIOS path (Chen Hanxiao) - document nmdm type console (Roman Bogorodskiy) - Fix typos in src/* (Nehal J Wani) - document that vfio is default for hostdev networks too (Laine Stump) - cpu: Add documentation for CPU driver APIs (Jiri Denemark) - virsh: Fix comment of vshCmdInfo (Li Yang) Portability: - Explicitly link virfirewalltest and virsystemdtest against dbus (Guido Günther) - qemuxml2argvtest: Don't use privileged mode upfront (Guido Günther) - tests: skip virfirewalltest on non-Linux systems (Roman Bogorodskiy) - tests: don't fail with newer gnutls (Martin Kletzander) - fix build with older gcc (Ján Tomko) - storage: reject negative indices (Eric Blake) - networkxml2firewalltest: fix build failure on freebsd (Pavel Hrdina) - virfirewall: fix build on freebsd (Pavel Hrdina) - Disable libvirtd by default when building on Win32 (Daniel P. Berrange) - Don't use SO_REUSEADDR on Win32 platforms (Daniel P. Berrange) - Conditionalize include of dlfcn.h in virmock.h (Daniel P. Berrange) - build: avoid 'index' as variable name (Eric Blake) - build: Don't use code with dbus_message_unref when built without dbus (Martin Kletzander) - tests: Fix systemd test with --without-driver-modules (Jiri Denemark) - Fix build on mingw32 (Ján Tomko) - build: avoid compiler warning on shadowed name (Jean-Baptiste Rouault) - tests: link against libxml2 (Guido Günther) - tests: build viridentitytest only WITH_ATTR. (Jincheng Miao) - maint: Correctly detect wether "gluster" cli tool is accessible (Peter Krempa) - libvirt-guests: avoid bashism (Guido Günther) - Use the force flag for mkfs -t xfs (Ján Tomko) Bug Fixes: - Restore skipping of setting capacity (John Ferlan) - qemu: fix crash when removing <filterref> from interface with update-device (Laine Stump) - storage: Clear all data allocated about backing store before reparsing (Peter Krempa) - nwfilter: Tear down temp. filters when tearing all filters (Stefan Berger) - Set mknod permission in device ACL for LXC USB devices (Daniel P. Berrange) - conf: avoid null deref during storage probe (Eric Blake) - qemu: properly quit migration with abort_on_error (Martin Kletzander) - qemu: don't call virFileExists() for network type disks (Martin Kletzander) - storage_backend_rbd: Correct argument order to rbd_create3 (Steven McDonald) - xen: ensure /usr/sbin/xend exists before checking status (Jim Fehlig) - Remove bogus ATTRIBUTE_NONNULL from virFirewallAddRuleFull (Daniel P. Berrange) - Make autostart of virtlockd actually work (Daniel P. Berrange) - Fix leak on OOM in virNWFilterVarValueCreateSimpleCopyValue (Daniel P. Berrange) - qemu: Avoid overflow when setting migration speed on inactive domains (Jiri Denemark) - qemu: don't check for backing chains for formats w/o snapshot support (Martin Kletzander) - Fix pci bus naming for PPC (Daniel P. Berrange) - Document behavior of setvcpus during guest boot (Ján Tomko) - Save domain status after cpu hotplug (Ján Tomko) - Fix error for out of range vcpu in qemuDomainPinVcpuFlags (Ján Tomko) - Properly free vcpupin info for unplugged CPUs (Ján Tomko) - Only set QEMU_CAPS_NO_HPET on x86 (Ján Tomko) - Fix Memory Leak in virStorageFileGetMetadataRecurse() (Nehal J Wani) - qemu: Unlock the NWFilter update lock by leaving via the cleanup label (Stefan Berger) - storage: netfs: Handle backend errors (John Ferlan) - conf: fix omission of <driver> in domain dumpxml (Eric Blake) - Fix virsystemdtest without SYSTEMD_DAEMON (Ján Tomko) - qemu: Avoid overflow when setting migration speed (Jiri Denemark) - bhyve: fix domain management (Wojciech Macek) - Check maximum startcpu value correctly (Ján Tomko) - storage: Don't update pool available/allocation if buildVol fails (John Ferlan) - LXC: Fix return code evaulation in lxcCheckNetNsSupport() (Richard Weinberger) - Fix incorrect values in redirdev ABI check error (Ján Tomko) - virNetDev{Replace,Restore}MacAddress: Fix memory leak (Wangrui K) - bhyveConnectGetCapabilities: Fix double caps unref (Michal Privoznik) - Simplify bhyveDriverGetCapabilities() (Michal Privoznik) - bhyve_capabilities: Add Semihalf to Copyright (Michal Privoznik) - tests: Don't crash when creating the config object fails (Guido Günther) - conf: avoid memleak on NULL path (Eric Blake) - lxc conf2xml: don't let current vcpus at 0: define won't like it (Cédric Bosdonnat) - QoS: make tc filters match all traffic (Antoni S. Puimedon) - NFS storage pool: Fix libvirtd crash due to refactor edit (John Ferlan) - Define CPUINFO_FILE_LEN and fix maxlen of cpuinfo file for all uses (Olivia Yin) - Fix Memory Leak in daemon/libvirtd.c (Nehal J Wani) - qemu: make sure agent returns error when required data are missing (Martin Kletzander) - Fix coverity-reported leak in virSecurityManagerGenLabel (Ján Tomko) - phyp: fix logic error on volume creation (Eric Blake) - qemu: cleanup error checking on agent replies (Martin Kletzander) Improvements - util: new stricter unsigned int parsing (Eric Blake) - util: fix uint parsing on 64-bit platforms (Eric Blake) - Misc error reporting bugs in QEMU cli builder (Daniel P. Berrange) - nwfilter: Validate rule after parsing (Stefan Berger) - Add support for QEMU migration to use SASL authentication (Sahid Orentino Ferdjaoui) - enforce sane readdir usage (Eric Blake) - network: use virDirRead in networkMigrateStateFiles (Laine Stump) - storage: use virDirRead API (Eric Blake) - drivers: use virDirRead API (Eric Blake) - util: use virDirRead API (Eric Blake) - conf: use virDirRead API (Eric Blake) - nodeinfo: use virDirRead API (Natanael Copa) - util: introduce virDirRead wrapper for readdir (Natanael Copa) - tests: remove hostdevmgr directory on cleanup (Martin Kletzander) - Use virFileFindResource to locate virtlockd daemon (Daniel P. Berrange) - Use virFileFindResource to locate libvirtd daemon (Daniel P. Berrange) - Recheck disk backing chains after snapshot (Jiri Denemark) - network: centralize check for active network during interface attach (Laine Stump) - network: set macvtap/hostdev networks active if their state file exists (Laine Stump) - network: change location of network state xml files (Laine Stump) - network: create statedir during driver initialization (Laine Stump) - network: fix virNetworkObjAssignDef and persistence (Laine Stump) - build: -avoid-version on libvirt_driver_nwfilter (Dwight Engen) - libxl: Support PV consoles (Ian Campbell) - build: add nwfilterxml2firewalldata to dist (Dwight Engen) - Add a test suite for nwfilter ebiptables tech driver (Daniel P. Berrange) - Remove last trace of direct firewall command exection (Daniel P. Berrange) - Convert ebiptablesDriverProbeStateMatch to virFirewall (Daniel P. Berrange) - Convert nwfilter ebiptablesApplyNewRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebtablesApplyDropAllRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebtablesApplyDHCPOnlyRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebtablesApplyBasicRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebiptablesTearNewRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebtablesRemoveBasicRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebiptablesTearOldRules to virFirewall (Daniel P. Berrange) - Convert nwfilter ebiptablesAllTeardown to virFirewall (Daniel P. Berrange) - Convert ebtables code over to use firewall APIs (Daniel P. Berrange) - Add test for converting network XML to iptables rules (Daniel P. Berrange) - Replace virNetworkObjPtr with virNetworkDefPtr in network platform APIs (Daniel P. Berrange) - Convert bridge driver over to use new firewall APIs (Daniel P. Berrange) - Introduce an object for managing firewall rulesets (Daniel P. Berrange) - Preserve error when tearing down nwfilter rules (Daniel P. Berrange) - Remove two-stage construction of commands in nwfilter (Daniel P. Berrange) - Merge nwfilter createRuleInstance driver into applyNewRules (Daniel P. Berrange) - Push virNWFilterRuleInstPtr out of (eb|ip)tablesCreateRuleInstance (Daniel P. Berrange) - Add helper methods for determining what protocol layer is used (Daniel P. Berrange) - Remove nwfilter tech driver 'displayRuleInstance' callback (Daniel P. Berrange) - Remove nwfilter tech driver 'removeRules' callback (Daniel P. Berrange) - Remove pointless storage of var names in virNWFilterHashTable (Daniel P. Berrange) - Remove virDomainNetType parameter from nwfilter drivers (Daniel P. Berrange) - Move virNWFilterTechDriver struct out of nwfilter_conf.h (Daniel P. Berrange) - Use virFileFindResource to locate CPU map XML (Daniel P. Berrange) - Use virFileFindResource to locate driver plugins (Daniel P. Berrange) - Use virFileFindResource to locate lock manager plugins (Daniel P. Berrange) - Use virFileFindResource to locate iohelper for fdstream (Nehal J Wani) - Use virFileFindResource to locate parthelper for storage backend (Nehal J Wani) - Use virFileFindResource to locate libvirt_lxc for capabilities (Nehal J Wani) - Use virFileFindResource to locate iohelper for virFileWrapperFdNew (Nehal J Wani) - Activate build dir overrides in libvirtd, virtlockd, virsh & tests (Daniel P. Berrange) - Add helpers for resolving path to resources in build tree (Daniel P. Berrange) - Add test suite for viralloc APIs (Daniel P. Berrange) - Add support for addressing backing stores by index (Jiri Denemark) - virStorageFileChainLookup: Return virStorageSourcePtr (Jiri Denemark) - qemuDomainBlockCommit: Track virStorageSourcePtr for base (Jiri Denemark) - qemuDomainBlockCommit: Don't track top_canon path separately (Jiri Denemark) - tests: Test backing store XML formatting and parsing (Jiri Denemark) - tests: More output options for xml2xml tests (Jiri Denemark) - conf: Format and parse backing chains in domain XML (Jiri Denemark) - conf: Output disk backing store details in domain XML (Jiri Denemark) - util: storage: Invert the way recursive metadata retrieval works (Peter Krempa) - util: virstoragefile: Don't mangle data stored about directories (Peter Krempa) - storage: Move disk->backingChain to the recursive disk->src.backingStore (Peter Krempa) - util: virstoragefile: Rename backingMeta to backingStore (Peter Krempa) - util: virstorage: Kill struct virStorageFileMetadata (Peter Krempa) - maint: Switch over from struct virStorageFileMetadata to virStorageSource (Peter Krempa) - util: storagefile: Add fields from virStorageMetadata to virStorageSource (Peter Krempa) - util: storagefile: Add function to free a virStorageSourcePtr (Peter Krempa) - virstoragefile: Kill "backingStore" field from virStorageFileMetadata (Peter Krempa) - util: virstoragefile: Don't use "backingStore" directly (Peter Krempa) - util: storagefile: Rename "canonPath" to "path" in virStorageFileMetadata (Peter Krempa) - util: storage: Rename "path" to "relPath" in virStorageFileMetadata (Peter Krempa) - storage: util: Clean up arguments of virStorageFileGetMetadataInternal (Peter Krempa) - util: storage: Move checking of the actual backing image to the worker (Peter Krempa) - util: storage: Remove obsolete argument virStorageFileGetMetadataInternal (Peter Krempa) - util: storagefile: Always store raw backing name in the metadata (Peter Krempa) - qemu: unexport qemuDiskChainCheckBroken (Peter Krempa) - bhyve: bhyveDomainDefineXML fixes (Roman Bogorodskiy) - PPC64 prefers to set pci-ohci controller as default USB controller (Li Zhang) - Make virDomainVcpuPinDel return void (Ján Tomko) - maint: update to latest gnulib (Eric Blake) - bhyve: domainCreateXML (Wojciech Macek) - Remove QEMU_CAPS_MACHINE_USB_OPT from ComputeCmdFlags (Ján Tomko) - conf: split <disk> schema into more pieces (Eric Blake) - conf: set up for per-grammar overrides in schemas (Eric Blake) - conf: restrict external snapshots to backing store formats (Eric Blake) - conf: move storage formats to common RNG file (Eric Blake) - conf: better <disk> interleaving in schema (Eric Blake) - conf: create common storage RNG grammar file (Eric Blake) - conf: delete internal directory field (Eric Blake) - conf: tweak chain lookup internals (Eric Blake) - conf: drop redundant parameter to chain lookup (Eric Blake) - conf: report error on chain lookup failure (Eric Blake) - util: new virFileRelLinkPointsTo function (Eric Blake) - conf: test backing chain lookup (Eric Blake) - Introduce --without-pm-utils to get rid of pm-is-supported dependency (Cédric Bosdonnat) - conf: delete useless backingStoreFormat field (Eric Blake) - conf: return backing information separately from metadata (Eric Blake) - conf: delete useless backingStoreIsFile field (Eric Blake) - conf: expose probe for non-local storage (Eric Blake) - conf: provide details on network backing store (Eric Blake) - conf: make virstoragetest debug easier (Eric Blake) - cpu: Properly check input parameters (Jiri Denemark) - Clean up virCgroupGetPercpuStats (Ján Tomko) - Rename id, max_id to need_cpus, total_cpus (Ján Tomko) - Extend virCgroupGetPercpuStats to fill in vcputime too (Ján Tomko) - Fix return value of virCgroupGetPercpuStats (Ján Tomko) - Don't require domain obj in qemuDomainGetPercpuStats (Ján Tomko) - conf: test for more fields (Eric Blake) - conf: start testing contents of the new backing chain fields (Eric Blake) - conf: track more fields in backing chain metadata (Eric Blake) - conf: rename some test fields (Eric Blake) - conf: earlier allocation during backing chain crawl (Eric Blake) - conf: track user vs. canonical name through full chain lookup (Eric Blake) - qemu: Unexport qemuBuildNetworkDriveURI() (Peter Krempa) - qemu: Refactor qemuGetDriveSourceString to take virStorageSourcePtr (Peter Krempa) - storage: Refactor location of metadata for storage drive access to files (Peter Krempa) - storage: Refactor storage file initialization to use virStorageSourcePtr (Peter Krempa) - conf: Refactor helpers to retrieve actual storage type (Peter Krempa) - tests: use virBhyveCapsBuild in bhyvexml2argv test (Roman Bogorodskiy) - conf: another refactor of virstoragetest (Eric Blake) - conf: interleave virstoragetest structs (Eric Blake) - conf: test for more scenarios (Eric Blake) - conf: fix detection of infinite backing loop (Eric Blake) - vmware: set the driver version (Jean-Baptiste Rouault) - tests: add bhyve xml2xml test (Roman Bogorodskiy) - bhyve: add domain metadata support (Roman Bogorodskiy) - bhyve: fix ATTRIBUTE_NONNULL usage (Roman Bogorodskiy) - Use a static initializer for static mutexes (Daniel P. Berrange) - Add syntax check to validate capitalization of abbreviations (Daniel P. Berrange) - Replace Pci with PCI throughout (Daniel P. Berrange) - Replace Usb with USB throughout (Daniel P. Berrange) - Replace Scsi with SCSI throughout (Daniel P. Berrange) - Switch systemd test to use generic dbus mock (Daniel P. Berrange) - Create a re-usable DBus LD_PRELOAD mock library (Daniel P. Berrange) - Introduce a new set of helper macros for mocking symbols (Daniel P. Berrange) - bhyve: connectCompareCPU support (Wojciech Macek) - bhyve: create capabilities submodule (Wojciech Macek) - bhyve: support for connectBaselineCPU (Wojciech Macek) - interface: dump inactive xml when interface isn't active (Laine Stump) - hash: add common utility functions (Eric Blake) - bhyve: add xml2argv tests for console (Roman Bogorodskiy) - bhyve: add console support through nmdm device (Roman Bogorodskiy) - bhyve: domain autostart support (David Shane Holden) - conf: track when storage type is still undetermined (Eric Blake) - tests: refactor virstoragetest for less stack space (Eric Blake) - tests: use C99 initialization for storage test (Eric Blake) - libxl: Set disk format for empty cdrom device (Stefan Bader) - libxl: Use id from virDomainObj inside the driver (Stefan Bader) - Add redirdevs to ABI stability check (Ján Tomko) - virsh: Make 'exit' action same as 'quit' (Li Yang) - Include PCI address in the error in virDomainNetFindIdx (Ján Tomko) - Move error reporting into virDomainNetFindIdx (Ján Tomko) - tests: simplify storage test cleanup (Eric Blake) - storage: Report error from VolOpen by default (Cole Robinson) - conf: modify tracking of encrypted images (Eric Blake) - conf: drop redundant parameters during probe (Eric Blake) - conf: track sizes directly in source struct (Eric Blake) - conf: use common struct in storage volumes (Eric Blake) - conf: move volume structs to util/ (Eric Blake) - conf: tweak volume target struct details (Eric Blake) - conf: manage disk source by struct instead of pieces (Eric Blake) - virsh: man: delete the unexpected character in snapshot-list (Shanzhi Yu) - maint: fix spelling errors in disk pools (Eric Blake) - conf: let snapshots share disk source struct (Eric Blake) - conf: move common disk source functions (Eric Blake) - util: don't support loopback and nbd when setuid (Eric Blake) - util: move detection of shared filesystems (Eric Blake) - conf: move storage source type to util/ (Eric Blake) - conf: move storage secret type to util/ (Eric Blake) - conf: move source pool type to util/ (Eric Blake) - conf: move storage encryption type to util/ (Eric Blake) - conf: move network disk protocol type to util/ (Eric Blake) - conf: move host disk type to util/ (Eric Blake) - conf: split network host structs to util/ (Eric Blake) - conf: split security label structs to util/ (Eric Blake) - maint: ensure src/ directory includes are clean (Eric Blake) - storage: gluster: Implement storage pool lookup (Peter Krempa) - storage: netfs: Support lookup of glusterfs pool sources (Peter Krempa) - storage: netfs: Split up and tidy up NFS storage pool source function (Peter Krempa) Cleanups: - tests: drop dead code from argv2xml and xml2xml (Eric Blake) - qemu: remove unneeded forward declaration (Martin Kletzander) Thanks everybody for your contributions to this release be it with ideas, report, patches, documentation or localizations ! Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] ANNOUNCE: libvirt 1.1.3.5 maintenance release
by Cole Robinson 03 May '14

03 May '14

libvirt 1.1.3.5 maintenance release is now available. This is libvirt 1.1.3 with additional bugfixes that have accumulated upstream since the initial release. This release can be downloaded at: http://libvirt.org/sources/stable_updates/libvirt-1.1.3.5.tar.gz Changes in this version: * qemu: Introduce qemuDomainDefCheckABIStability * interface: dump inactive xml when interface isn't active * interface: Introduce netcfInterfaceObjIsActive * Ignore additional fields in iscsiadm output * qemu: fix crash when removing <filterref> from interface with update- device * Only set QEMU_CAPS_NO_HPET on x86 * Fix journald PRIORITY values * qemu: make sure agent returns error when required data are missing * qemu: remove unneeded forward declaration * qemu: cleanup error checking on agent replies * Ignore char devices in storage pools by default * Ignore missing files on pool refresh * storage: reduce number of stat calls * Fix explicit usage of default video PCI slots * virNetClientSetTLSSession: Restore original signal mask * storage: use valid XML for awkward volume names * maint: fix comma style issues: conf * virNetServerRun: Notify systemd that we're accepting clients * libvirt-guests: Wait for libvirtd to initialize * virSystemdCreateMachine: Set dependencies for slices * Add Documentation fields to systemd service files * Add a mutex to serialize updates to firewall * virt-login-shell: also build virAtomic.h * Fix conflicting types of virInitctlSetRunLevel For info about past maintenance releases, see: http://wiki.libvirt.org/page/Maintenance_Releases Thanks, Cole

1 0

[libvirt] [PATCH 1/3] tests: Don't crash when creating the config object fails
by Guido Günther 03 May '14

03 May '14

As observed when building in a chroot and QEMU_USER doesn't exist --- tests/qemuargv2xmltest.c | 3 +++ tests/qemuxml2argvtest.c | 3 +++ 2 files changed, 6 insertions(+) diff --git a/tests/qemuargv2xmltest.c b/tests/qemuargv2xmltest.c index 6d7e23e..4cc3749 100644 --- a/tests/qemuargv2xmltest.c +++ b/tests/qemuargv2xmltest.c @@ -128,6 +128,9 @@ mymain(void) int ret = 0; driver.config = virQEMUDriverConfigNew(false); + if (driver.config == NULL) + return EXIT_FAILURE; + if ((driver.caps = testQemuCapsInit()) == NULL) return EXIT_FAILURE; diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 56854dc..13ed4f6 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -501,6 +501,9 @@ mymain(void) } driver.config = virQEMUDriverConfigNew(true); + if (driver.config == NULL) + return EXIT_FAILURE; + VIR_FREE(driver.config->spiceListen); VIR_FREE(driver.config->vncListen); -- 1.9.1

4 17

[libvirt] [PATCH] Explicitly link virfirewalltest and virsystemdtest against dbus
by Guido Günther 03 May '14

03 May '14

This fixes link failures like: CCLD virfirewalltest /usr/bin/ld: virfirewalltest-virfirewalltest.o: undefined reference to symbol 'dbus_message_iter_init_append' --- tests/Makefile.am | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index 7ffb44e..5ef8940 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -963,7 +963,7 @@ virmockdbus_la_LDFLAGS = -module -avoid-version \ virsystemdtest_SOURCES = \ virsystemdtest.c testutils.h testutils.c virsystemdtest_CFLAGS = $(AM_CFLAGS) $(DBUS_CFLAGS) -virsystemdtest_LDADD = $(LDADDS) +virsystemdtest_LDADD = $(LDADDS) $(DBUS_LIBS) else ! WITH_DBUS EXTRA_DIST += virdbustest.c virmockdbus.c virsystemdtest.c @@ -1047,7 +1047,7 @@ virfiletest_LDADD = $(LDADDS) virfirewalltest_SOURCES = \ virfirewalltest.c testutils.h testutils.c -virfirewalltest_LDADD = $(LDADDS) +virfirewalltest_LDADD = $(LDADDS) $(DBUS_LIBS) virfirewalltest_CFLAGS = $(AM_CFLAGS) $(DBUS_CFLAGS) jsontest_SOURCES = \ -- 1.9.2

2 2

[libvirt] CPU type/flags when converting a physical machine to run on libvirt
by Richard W.M. Jones 03 May '14

03 May '14

I'm in the process of rewriting virt-p2v which is our program for converting physical machines to become virtual machines, running on top of libvirt + KVM. The physical machine has certain characteristics -- eg: - number of physical cores - amount of RAM - CPU type (eg. AMD Opteron, Intel Core i7) - CPU flags (eg. ACPI, SSE4) which it might make sense to reflect in the libvirt XML of the virtual machine we create. The old version of virt-p2v is pretty simplistic about this. It generates *only* an i386 or x86-64 VM, and the only flags it considers are 'apic', 'acpi', 'pae' and 'lm' [the latter to distinguish between 32 and 64 bit x86]. What should the new version do? Particular questions: - Should we try to reflect the CPU type of the physical machine in the virtual machine? eg. If it's an Opteron, we generate an Opteron target machine. (I believe the answer is *no*, because this is not live migration, and most guests can boot on any compatible CPU). - How can I ask libvirt to give me the best possible CPU, and not some baseline? Normally I use host-model, but I think that prevents migration. - What CPU flags should be reflected in the target libvirt XML? - Is it worth modelling CPU thread layout? (I suspect this will be a lot of work with the potential to break things rather than provide any measurable benefits.) - Is there anything else I haven't thought about? However the overriding rule is: - We *must not* end up with a target virtual machine which doesn't work! Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into KVM guests. http://libguestfs.org/virt-v2v

2 4

[libvirt] Reg : failing to attach iscsi disk from virsh to running VM on xen
by Prapulla Kumar 02 May '14

02 May '14

Hi , For attaching iSCSI disk I have tried like this *iscsi target side :* * 1 created an iqn number* #tgtadm --lld iscsi --op new --mode target --tid 1 -T iqn.2009-06.com.eucalyptus.cluster1:store1 * 2. added one lun to it* #tgtadm --lld iscsi --op new --mode logicalunit --tid 1 --lun 1 -b /fs.iscsi.disk * 3.To enable the target to accept any initiators, enter:* #tgtadm --lld iscsi --op bind --mode target --tid 1 -I ALL * 4.To view the current configuration, enter:* #tgtadm --lld iscsi --op show --mode target *iscsi initaitor side (where libvirt and xen is running)* * 1. to** scan for the newly created target* * #*iscsiadm -m discovery -t sendtargets -p 192.168.53.14 * 2. created pool file * <pool type="iscsi"> <name>sample3</name> <source> <host name="192.168.53.14"/> <device path="iqn.2009-06.com.eucalyptus.cluster1:store1"/> </source> <target> <path>/dev/disk/by-path</path> </target> </pool> * 3.*#virsh pool-define sample_pool.xml #virsh pool-start sample3 then it created one link like this at */dev/disk/by-path/* *ip-192.168.53.14:3260-iscsi-iqn.2009-06.com.eucalyptus.cluster1:store1-lun-1 -> ../../sdb* 4. I have attached using virsh attach-disk #virsh attach-disk 1 /dev/sdb xvdb It is attached successfully , I console in vm I have checked . But while using with eucalyptus,It is not attching. I'm using eucalyptus 3.4.2 , libvirt 1.2.3 with xen4.3.1 to attach volume to vm . eucalyptus attaching volume is done through iscsi disk , intermediate files are created while attaching volume in xml format as below. *File 1:* <?xml version="1.0" encoding="UTF-8"?> <disk type="block"> <driver cache="none" name="phy"/> <source dev="/dev/disk/by-id/scsi-1IET_00230001"/> <target dev="sdf" bus="scsi"/> <serial>vol-759A3CDB-dev-sdf</serial> </disk> *File 2:* <?xml version="1.0" encoding="UTF-8"?> <volume> <hypervisor type="xen" capability="xen+hw" bitness="64"/> <id>vol-759A3CDB</id> <user>AIDOMCCNSP7YNR7IEZEUX</user> <instancePath>/usr/local/eucalyptus/work/AIDOMCCNSP7YNR7IEZEUX/i-ED153F84</instancePath> <os platform="linux" virtioRoot="false" virtioDisk="false" virtioNetwork="false"/> <backing> <root type="image"/> </backing> <diskPath targetDeviceType="disk" targetDeviceName="sdf" targetDeviceBus="scsi" sourceType="block" serial="vol-759A3CDB-dev-sdf">/dev/disk/by-id/scsi-1IET_00230001</diskPath> </volume> *in libvirtd.log i'm getting* *error : libxlDomainAttachDeviceDiskLive:2567 : unsupported configuration: disk bus 'scsi' cannot be hotplugged.* Can u suggest how to enable or patch to scsi hotplug from libvirt which support for xen. Can u say the meaning of PATCH 11/12 what is 11/12 means. eg ; [libvirt] [PATCH 11/12] Properly support SCSI drive hotplug With Regards Prapulla Kumar R

1 0

[libvirt] qemu: managedsave vs. save
by Richard Weinberger 02 May '14

02 May '14

Hi! My KVM hosts share the same filesystem and I'm facing an issue using managedsave. If I save vmX using managedsave on hostA and restore it later using "virsh restore" in hostB the qemu process consumes 100% CPU and makes no progress. On the other hand, if I save vmX using save the restore works fine on hostB. Why can't I restore vms saved by managedsave using restore? Or is this a known issue on libvirt-0.10.X (CentOS 6)? -- Thanks, //richard

3 9

[libvirt] [PATCHv2] storageVolCreateXMLFrom: Allow multiple accesses to origvol
by Michal Privoznik 02 May '14

02 May '14

When creating a new volume, it is possible to copy data into it from another already existing volume (referred to as @origvol). Obviously, the read-only access to @origvol is required, which is thread safe (probably not performance-wise though). However, with current code both @newvol and @origvol are marked as building for the time of copying data from the @origvol to @newvol. The rationale behind is to disallow some operations on both @origvol and @newvol, e.g. vol-wipe, vol-delete, vol-download. While it makes sense to not allow such operations on partly copied mirror, but it doesn't make sense to disallow the operations on the source (@origvol). Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Diff to v1: -introduced a counter in addition to not marking origvol as building src/conf/storage_conf.h | 1 + src/storage/storage_driver.c | 14 +++++++------- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/src/conf/storage_conf.h b/src/conf/storage_conf.h index 9ad38e1..eae959c 100644 --- a/src/conf/storage_conf.h +++ b/src/conf/storage_conf.h @@ -64,6 +64,7 @@ struct _virStorageVolDef { int type; /* enum virStorageVolType */ unsigned int building; + unsigned int in_use; virStorageVolSource source; virStorageSource target; diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c index 2cb8347..a953dfd 100644 --- a/src/storage/storage_driver.c +++ b/src/storage/storage_driver.c @@ -1640,7 +1640,7 @@ storageVolDelete(virStorageVolPtr obj, if (virStorageVolDeleteEnsureACL(obj->conn, pool->def, vol) < 0) goto cleanup; - if (vol->building) { + if (vol->building || vol->in_use) { virReportError(VIR_ERR_OPERATION_INVALID, _("volume '%s' is still being allocated."), vol->name); @@ -1912,8 +1912,8 @@ storageVolCreateXMLFrom(virStoragePoolPtr obj, /* Drop the pool lock during volume allocation */ pool->asyncjobs++; - origvol->building = 1; newvol->building = 1; + origvol->in_use++; virStoragePoolObjUnlock(pool); if (origpool) { @@ -1929,7 +1929,7 @@ storageVolCreateXMLFrom(virStoragePoolPtr obj, virStoragePoolObjLock(origpool); storageDriverUnlock(driver); - origvol->building = 0; + origvol->in_use--; newvol->building = 0; allocation = newvol->target.allocation; pool->asyncjobs--; @@ -2010,7 +2010,7 @@ storageVolDownload(virStorageVolPtr obj, if (virStorageVolDownloadEnsureACL(obj->conn, pool->def, vol) < 0) goto cleanup; - if (vol->building) { + if (vol->building || vol->in_use) { virReportError(VIR_ERR_OPERATION_INVALID, _("volume '%s' is still being allocated."), vol->name); @@ -2076,7 +2076,7 @@ storageVolUpload(virStorageVolPtr obj, if (virStorageVolUploadEnsureACL(obj->conn, pool->def, vol) < 0) goto cleanup; - if (vol->building) { + if (vol->building || vol->in_use) { virReportError(VIR_ERR_OPERATION_INVALID, _("volume '%s' is still being allocated."), vol->name); @@ -2167,7 +2167,7 @@ storageVolResize(virStorageVolPtr obj, if (virStorageVolResizeEnsureACL(obj->conn, pool->def, vol) < 0) goto cleanup; - if (vol->building) { + if (vol->building || vol->in_use) { virReportError(VIR_ERR_OPERATION_INVALID, _("volume '%s' is still being allocated."), vol->name); @@ -2474,7 +2474,7 @@ storageVolWipePattern(virStorageVolPtr obj, if (virStorageVolWipePatternEnsureACL(obj->conn, pool->def, vol) < 0) goto cleanup; - if (vol->building) { + if (vol->building || vol->in_use) { virReportError(VIR_ERR_OPERATION_INVALID, _("volume '%s' is still being allocated."), vol->name); -- 1.9.0

2 1

[libvirt] [PATCH] tests: skip virfirewalltest on non-Linux systems
by Roman Bogorodskiy 02 May '14

02 May '14

Currently firewalling is supported on Linux only, so skip the virfirewalltest on other platforms. --- tests/virfirewalltest.c | 58 +++++++++++++++++++++++++++++-------------------- 1 file changed, 35 insertions(+), 23 deletions(-) diff --git a/tests/virfirewalltest.c b/tests/virfirewalltest.c index 805fa44..af74d57 100644 --- a/tests/virfirewalltest.c +++ b/tests/virfirewalltest.c @@ -24,23 +24,26 @@ #define __VIR_COMMAND_PRIV_H_ALLOW__ #include "testutils.h" -#include "virbuffer.h" -#include "vircommandpriv.h" -#include "virfirewallpriv.h" -#include "virmock.h" -#include "virdbuspriv.h" -#define VIR_FROM_THIS VIR_FROM_FIREWALL +#if defined(__linux__) -#if WITH_DBUS -# include <dbus/dbus.h> -#endif +# include "virbuffer.h" +# include "vircommandpriv.h" +# include "virfirewallpriv.h" +# include "virmock.h" +# include "virdbuspriv.h" + +# define VIR_FROM_THIS VIR_FROM_FIREWALL + +# if WITH_DBUS +# include <dbus/dbus.h> +# endif static bool fwDisabled = true; static virBufferPtr fwBuf; static bool fwError; -#define TEST_FILTER_TABLE_LIST \ +# define TEST_FILTER_TABLE_LIST \ "Chain INPUT (policy ACCEPT)\n" \ "target prot opt source destination\n" \ "\n" \ @@ -50,7 +53,7 @@ static bool fwError; "Chain OUTPUT (policy ACCEPT)\n" \ "target prot opt source destination\n" -#define TEST_NAT_TABLE_LIST \ +# define TEST_NAT_TABLE_LIST \ "Chain PREROUTING (policy ACCEPT)\n" \ "target prot opt source destination\n" \ "\n" \ @@ -63,7 +66,7 @@ static bool fwError; "Chain POSTROUTING (policy ACCEPT)\n" \ "target prot opt source destination\n" -#if WITH_DBUS +# if WITH_DBUS VIR_MOCK_IMPL_RET_ARGS(dbus_connection_send_with_reply_and_block, DBusMessage *, DBusConnection *, connection, @@ -186,7 +189,7 @@ VIR_MOCK_IMPL_RET_ARGS(dbus_connection_send_with_reply_and_block, goto cleanup; } -#endif +# endif struct testFirewallData { virFirewallBackend tryBackend; @@ -1126,7 +1129,7 @@ mymain(void) { int ret = 0; -#define RUN_TEST_DIRECT(name, method) \ +# define RUN_TEST_DIRECT(name, method) \ do { \ struct testFirewallData data; \ data.tryBackend = VIR_FIREWALL_BACKEND_AUTOMATIC; \ @@ -1141,8 +1144,8 @@ mymain(void) ret = -1; \ } while (0) -#if WITH_DBUS -# define RUN_TEST_FIREWALLD(name, method) \ +# if WITH_DBUS +# define RUN_TEST_FIREWALLD(name, method) \ do { \ struct testFirewallData data; \ data.tryBackend = VIR_FIREWALL_BACKEND_AUTOMATIC; \ @@ -1157,13 +1160,13 @@ mymain(void) ret = -1; \ } while (0) -# define RUN_TEST(name, method) \ +# define RUN_TEST(name, method) \ RUN_TEST_DIRECT(name, method); \ RUN_TEST_FIREWALLD(name, method) -#else /* ! WITH_DBUS */ -# define RUN_TEST(name, method) \ +# else /* ! WITH_DBUS */ +# define RUN_TEST(name, method) \ RUN_TEST_DIRECT(name, method) -#endif /* ! WITH_DBUS */ +# endif /* ! WITH_DBUS */ RUN_TEST("single group", testFirewallSingleGroup); RUN_TEST("remove rule", testFirewallRemoveRule); @@ -1179,8 +1182,17 @@ mymain(void) return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; } -#if WITH_DBUS +# if WITH_DBUS VIRT_TEST_MAIN_PRELOAD(mymain, abs_builddir "/.libs/virmockdbus.so") -#else +# else VIRT_TEST_MAIN(mymain) -#endif +# endif + +#else /* ! defined (__linux__) */ + +int main(void) +{ + return EXIT_AM_SKIP; +} + +#endif /* ! defined(__linux__) */ -- 1.9.0

5 8

[libvirt] [PATCH] Restore skipping of setting capacity
by John Ferlan 02 May '14

02 May '14

Commit id 'ac9a0963' refactored out the 'withCapacity' for the virStorageBackendUpdateVolInfo() API. See: http://www.redhat.com/archives/libvir-list/2014-April/msg00043.html This resulted in a difference in how 'virsh vol-info --pool <poolName> <volume>' or 'virsh vol-list vol-list --pool <poolName> --details' outputs the capacity information for a directory pool with a qcow2 sparse file. For example, using the following XML mkdir /home/TestPool cat testpool.xml <pool type='dir'> <name>TestPool</name> <uuid>6bf80895-10b6-75a6-6059-89fdea2aefb7</uuid> <source> </source> <target> <path>/home/TestPool</path> <permissions> <mode>0755</mode> <owner>0</owner> <group>0</group> </permissions> </target> </pool> virsh pool-create testpool.xml virsh vol-create-as --pool TestPool temp_vol_1 \ --capacity 1048576 --allocation 1048576 --format qcow2 virsh vol-info --pool TestPool temp_vol_1 Results in listing a Capacity value. Prior to the commit, the value would be '1.0 MiB' (1048576 bytes). However, after the commit the output would be (for example) '192.50 KiB', which for my system was the size of the volume in my file system (eg 'ls -l TestPool/temp_vol_1' results in '197120' bytes or 192.50 KiB). While perhaps technically correct, it's not necessarily what the user expected (certainly virt-test didn't expect it). This patch restores the code to not update the target capacity for this path Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/storage/storage_backend.c | 22 +++++++++++++++------- src/storage/storage_backend.h | 5 ++++- src/storage/storage_backend_disk.c | 2 +- src/storage/storage_backend_fs.c | 11 +++++++---- src/storage/storage_backend_gluster.c | 2 +- src/storage/storage_backend_logical.c | 2 +- src/storage/storage_backend_mpath.c | 2 +- src/storage/storage_backend_scsi.c | 2 +- 8 files changed, 31 insertions(+), 17 deletions(-) diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index 946196b..afedbf5 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -1403,6 +1403,7 @@ virStorageBackendVolOpen(const char *path, struct stat *sb, int virStorageBackendUpdateVolTargetInfo(virStorageSourcePtr target, + bool updateCapacity, bool withBlockVolFormat, unsigned int openflags) { @@ -1413,7 +1414,8 @@ virStorageBackendUpdateVolTargetInfo(virStorageSourcePtr target, goto cleanup; fd = ret; - if ((ret = virStorageBackendUpdateVolTargetInfoFD(target, fd, &sb)) < 0) + if ((ret = virStorageBackendUpdateVolTargetInfoFD(target, fd, &sb, + updateCapacity)) < 0) goto cleanup; if (withBlockVolFormat) { @@ -1429,18 +1431,21 @@ virStorageBackendUpdateVolTargetInfo(virStorageSourcePtr target, int virStorageBackendUpdateVolInfo(virStorageVolDefPtr vol, + bool updateCapacity, bool withBlockVolFormat, unsigned int openflags) { int ret; if ((ret = virStorageBackendUpdateVolTargetInfo(&vol->target, + updateCapacity, withBlockVolFormat, openflags)) < 0) return ret; if (vol->backingStore.path && (ret = virStorageBackendUpdateVolTargetInfo(&vol->backingStore, + updateCapacity, withBlockVolFormat, VIR_STORAGE_VOL_OPEN_DEFAULT)) < 0) return ret; @@ -1453,15 +1458,15 @@ virStorageBackendUpdateVolInfo(virStorageVolDefPtr vol, * @target: target definition ptr of volume to update * @fd: fd of storage volume to update, via virStorageBackendOpenVol*, or -1 * @sb: details about file (must match @fd, if that is provided) - * @allocation: If not NULL, updated allocation information will be stored - * @capacity: If not NULL, updated capacity info will be stored + * @updateCapacity: If true, updated capacity info will be stored * * Returns 0 for success, -1 on a legitimate error condition. */ int virStorageBackendUpdateVolTargetInfoFD(virStorageSourcePtr target, int fd, - struct stat *sb) + struct stat *sb, + bool updateCapacity) { #if WITH_SELINUX security_context_t filecon = NULL; @@ -1477,10 +1482,12 @@ virStorageBackendUpdateVolTargetInfoFD(virStorageSourcePtr target, /* Regular files may be sparse, so logical size (capacity) is not same * as actual allocation above */ - target->capacity = sb->st_size; + if (updateCapacity) + target->capacity = sb->st_size; } else if (S_ISDIR(sb->st_mode)) { target->allocation = 0; - target->capacity = 0; + if (updateCapacity) + target->capacity = 0; } else if (fd >= 0) { off_t end; /* XXX this is POSIX compliant, but doesn't work for CHAR files, @@ -1496,7 +1503,8 @@ virStorageBackendUpdateVolTargetInfoFD(virStorageSourcePtr target, return -1; } target->allocation = end; - target->capacity = end; + if (updateCapacity) + target->capacity = end; } if (!target->perms && VIR_ALLOC(target->perms) < 0) diff --git a/src/storage/storage_backend.h b/src/storage/storage_backend.h index 5997077..456b9d7 100644 --- a/src/storage/storage_backend.h +++ b/src/storage/storage_backend.h @@ -137,14 +137,17 @@ int virStorageBackendVolOpen(const char *path, struct stat *sb, ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2); int virStorageBackendUpdateVolInfo(virStorageVolDefPtr vol, + bool updateCapacity, bool withBlockVolFormat, unsigned int openflags); int virStorageBackendUpdateVolTargetInfo(virStorageSourcePtr target, + bool updateCapacity, bool withBlockVolFormat, unsigned int openflags); int virStorageBackendUpdateVolTargetInfoFD(virStorageSourcePtr target, int fd, - struct stat *sb); + struct stat *sb, + bool updateCapacity); char *virStorageBackendStablePath(virStoragePoolObjPtr pool, const char *devpath, diff --git a/src/storage/storage_backend_disk.c b/src/storage/storage_backend_disk.c index 9cebcca..13336fc 100644 --- a/src/storage/storage_backend_disk.c +++ b/src/storage/storage_backend_disk.c @@ -113,7 +113,7 @@ virStorageBackendDiskMakeDataVol(virStoragePoolObjPtr pool, } /* Refresh allocation/capacity/perms */ - if (virStorageBackendUpdateVolInfo(vol, false, + if (virStorageBackendUpdateVolInfo(vol, true, false, VIR_STORAGE_VOL_OPEN_DEFAULT) < 0) return -1; diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 3694c26..5a2add8 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -84,7 +84,8 @@ virStorageBackendProbeTarget(virStorageSourcePtr target, goto error; /* Take care to propagate ret, it is not always -1 */ fd = ret; - if ((ret = virStorageBackendUpdateVolTargetInfoFD(target, fd, &sb)) < 0) { + if ((ret = virStorageBackendUpdateVolTargetInfoFD(target, fd, + &sb, true)) < 0) { goto error; } @@ -913,7 +914,7 @@ virStorageBackendFileSystemRefresh(virConnectPtr conn ATTRIBUTE_UNUSED, vol->backingStore.format = backingStoreFormat; ignore_value(virStorageBackendUpdateVolTargetInfo( - &vol->backingStore, false, + &vol->backingStore, true, false, VIR_STORAGE_VOL_OPEN_DEFAULT)); /* If this failed, the backing file is currently unavailable, * the capacity, allocation, owner, group and mode are unknown. @@ -1190,8 +1191,10 @@ virStorageBackendFileSystemVolRefresh(virConnectPtr conn, { int ret; - /* Refresh allocation / permissions info in case its changed */ - ret = virStorageBackendUpdateVolInfo(vol, false, + /* Refresh allocation / permissions info in case its changed + * don't update the capacity value for this pass + */ + ret = virStorageBackendUpdateVolInfo(vol, false, false, VIR_STORAGE_VOL_FS_OPEN_FLAGS); if (ret < 0) return ret; diff --git a/src/storage/storage_backend_gluster.c b/src/storage/storage_backend_gluster.c index e0a25df..28db909 100644 --- a/src/storage/storage_backend_gluster.c +++ b/src/storage/storage_backend_gluster.c @@ -267,7 +267,7 @@ virStorageBackendGlusterRefreshVol(virStorageBackendGlusterStatePtr state, if (VIR_ALLOC(vol) < 0) goto cleanup; - if (virStorageBackendUpdateVolTargetInfoFD(&vol->target, -1, st) < 0) + if (virStorageBackendUpdateVolTargetInfoFD(&vol->target, -1, st, true) < 0) goto cleanup; if (virStorageBackendGlusterSetMetadata(state, vol, name) < 0) diff --git a/src/storage/storage_backend_logical.c b/src/storage/storage_backend_logical.c index ed3a012..a597e67 100644 --- a/src/storage/storage_backend_logical.c +++ b/src/storage/storage_backend_logical.c @@ -149,7 +149,7 @@ virStorageBackendLogicalMakeVol(char **const groups, if (!vol->key && VIR_STRDUP(vol->key, groups[2]) < 0) goto cleanup; - if (virStorageBackendUpdateVolInfo(vol, false, + if (virStorageBackendUpdateVolInfo(vol, true, false, VIR_STORAGE_VOL_OPEN_DEFAULT) < 0) goto cleanup; diff --git a/src/storage/storage_backend_mpath.c b/src/storage/storage_backend_mpath.c index f0ed189..8c3b0df 100644 --- a/src/storage/storage_backend_mpath.c +++ b/src/storage/storage_backend_mpath.c @@ -60,7 +60,7 @@ virStorageBackendMpathNewVol(virStoragePoolObjPtr pool, if (virAsprintf(&vol->target.path, "/dev/%s", dev) < 0) goto cleanup; - if (virStorageBackendUpdateVolInfo(vol, true, + if (virStorageBackendUpdateVolInfo(vol, true, true, VIR_STORAGE_VOL_OPEN_DEFAULT) < 0) { goto cleanup; } diff --git a/src/storage/storage_backend_scsi.c b/src/storage/storage_backend_scsi.c index c448d7f..71bcf56 100644 --- a/src/storage/storage_backend_scsi.c +++ b/src/storage/storage_backend_scsi.c @@ -199,7 +199,7 @@ virStorageBackendSCSINewLun(virStoragePoolObjPtr pool, goto free_vol; } - if (virStorageBackendUpdateVolInfo(vol, true, + if (virStorageBackendUpdateVolInfo(vol, true, true, VIR_STORAGE_VOL_OPEN_DEFAULT) < 0) { retval = -1; goto free_vol; -- 1.9.0

3 5