- Devel - libvirt List Archives

[libvirt] [PATCH] qemu: process: Use enum belonging to correct type
by Peter Krempa 18 Jan '19

18 Jan '19

GCC was unhappy about comparison between two distinct enum types. Use the correct value instead. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- Pushed as a build fix. src/qemu/qemu_process.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index afa960a683..855bd9cb14 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -7835,7 +7835,7 @@ qemuProcessRefreshLegacyBlockjob(void *payload, return 0; } - if (jobtype == VIR_DOMAIN_BLOCK_JOB_TYPE_COMMIT && + if (jobtype == QEMU_BLOCKJOB_TYPE_COMMIT && disk->mirrorJob == VIR_DOMAIN_BLOCK_JOB_TYPE_ACTIVE_COMMIT) jobtype = disk->mirrorJob; -- 2.20.1

2 2

[libvirt] [PATCH] rpc: virNetClientNew: fix socket leak on error path
by Nikolay Shirokovskiy 18 Jan '19

18 Jan '19

if virNetClientNew finishes with error before sock is set to client object then sock does not get unrefed. This is unexpected by function clients like virNetClientNewUNIX. Let's make sure sock gets unrefed on any error path. Next some clients like virNetClientNewLibSSH2 try to unref sock on virNetClientNew errors. This is not correct even before this patch because in some cases virNetClientNew unrefed sock on error path by itself. Let's give up sock managment to virNetClientNew entirely. Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy(a)virtuozzo.com> --- src/rpc/virnetclient.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c index 2aced79..b7ffdcb 100644 --- a/src/rpc/virnetclient.c +++ b/src/rpc/virnetclient.c @@ -299,7 +299,7 @@ static virNetClientPtr virNetClientNew(virNetSocketPtr sock, int wakeupFD[2] = { -1, -1 }; if (virNetClientInitialize() < 0) - return NULL; + goto error; if (pipe2(wakeupFD, O_CLOEXEC) < 0) { virReportSystemError(errno, "%s", @@ -311,6 +311,7 @@ static virNetClientPtr virNetClientNew(virNetSocketPtr sock, goto error; client->sock = sock; + sock = NULL; client->wakeupReadFD = wakeupFD[0]; client->wakeupSendFD = wakeupFD[1]; wakeupFD[0] = wakeupFD[1] = -1; @@ -327,6 +328,7 @@ static virNetClientPtr virNetClientNew(virNetSocketPtr sock, VIR_FORCE_CLOSE(wakeupFD[0]); VIR_FORCE_CLOSE(wakeupFD[1]); virObjectUnref(client); + virObjectUnref(sock); return NULL; } @@ -513,7 +515,6 @@ virNetClientPtr virNetClientNewLibSSH2(const char *host, if (!(ret = virNetClientNew(sock, NULL))) goto cleanup; - sock = NULL; cleanup: VIR_FREE(command); @@ -522,7 +523,6 @@ virNetClientPtr virNetClientNewLibSSH2(const char *host, VIR_FREE(homedir); VIR_FREE(confdir); VIR_FREE(nc); - virObjectUnref(sock); return ret; no_memory: @@ -619,7 +619,6 @@ virNetClientPtr virNetClientNewLibssh(const char *host, if (!(ret = virNetClientNew(sock, NULL))) goto cleanup; - sock = NULL; cleanup: VIR_FREE(command); @@ -628,7 +627,6 @@ virNetClientPtr virNetClientNewLibssh(const char *host, VIR_FREE(homedir); VIR_FREE(confdir); VIR_FREE(nc); - virObjectUnref(sock); return ret; no_memory: -- 1.8.3.1

2 2

[libvirt] [PATCH] storage: More uniquely identify NPIV LUNs
by John Ferlan 18 Jan '19

18 Jan '19

https://bugzilla.redhat.com/show_bug.cgi?id=1657468 Commit be1bb6c95 changed the way volumes were stored from a forward linked list to a hash table. In doing so, it required that each vol object would have 3 unique values as keys into tables - key, name, and path. Due to how vHBA/NPIV LUNs are created/used this resulted in a failure to utilize all the LUN's found during processing. The virStorageBackendSCSINewLun uses virStorageBackendSCSISerial in order to read/return the unique serial number of the LUN to be used as a key for the volume. However, for NPIV based LUNs the logic results in usage of the same serial value for each path to the LUN. For example, $ lsscsi -tg ... [4:0:3:13] disk fc:0x207800c0ffd79b2a0xeb02ef /dev/sde /dev/sg16 [4:0:4:0] disk fc:0x50060169446021980xeb1f00 /dev/sdf /dev/sg17 [4:0:5:0] disk fc:0x50060161446021980xeb2000 /dev/sdg /dev/sg18 ... /lib/udev/scsi_id --replace-whitespace --whitelisted --device /dev/sde 3600c0ff000d7a2965c603e5401000000 /lib/udev/scsi_id --replace-whitespace --whitelisted --device /dev/sdf 350060160c460219850060160c4602198 /lib/udev/scsi_id --replace-whitespace --whitelisted --device /dev/sdg 350060160c460219850060160c4602198 The /dev/sdf and /dev/sdg although separate LUNs would end up with the same serial number used for the vol->key value. When attempting to add the LUN via virStoragePoolObjAddVol, the hash table code will indicate that we have a duplicate: virHashAddOrUpdateEntry:341 : internal error: Duplicate key and thus the LUN is not added to the pool. Digging deeper into the scsi_id output using the --export option one will find that the only difference between the two luns is: ID_TARGET_PORT=1 for /dev/sdf ID_TARGET_PORT=2 for /dev/sdg So, let's use the ID_TARGET_PORT string value along with the serial to generate a more unique key using "@serial_PORT@target_port", where @target_port is just the value of the ID_TARGET_PORT for the LUN. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/storage/storage_util.c | 61 +++++++++++++++++++++++++++++++++++++- 1 file changed, 60 insertions(+), 1 deletion(-) diff --git a/src/storage/storage_util.c b/src/storage/storage_util.c index a84ee5b600..d6d441c06d 100644 --- a/src/storage/storage_util.c +++ b/src/storage/storage_util.c @@ -3755,6 +3755,49 @@ virStorageBackendRefreshLocal(virStoragePoolObjPtr pool) } +static char * +virStorageBackendSCSITargetPort(const char *dev) +{ + char *target_port = NULL; + const char *id; +#ifdef WITH_UDEV + virCommandPtr cmd = virCommandNewArgList( + "/lib/udev/scsi_id", + "--replace-whitespace", + "--whitelisted", + "--export", + "--device", dev, + NULL + ); + + /* Run the program and capture its output */ + virCommandSetOutputBuffer(cmd, &target_port); + if (virCommandRun(cmd, NULL) < 0) + goto cleanup; +#endif + + if (target_port && STRNEQ(target_port, "") && + (id = strstr(target_port, "ID_TARGET_PORT="))) { + char *nl = strchr(id, '\n'); + if (nl) + *nl = '\0'; + id = strrchr(id, '='); + memmove(target_port, id + 1, strlen(id)); + } else { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unable to uniquely identify target port for '%s'"), + dev); + } + +#ifdef WITH_UDEV + cleanup: + virCommandFree(cmd); +#endif + + return target_port; +} + + static char * virStorageBackendSCSISerial(const char *dev) { @@ -3813,6 +3856,8 @@ virStorageBackendSCSINewLun(virStoragePoolObjPtr pool, virStoragePoolDefPtr def = virStoragePoolObjGetDef(pool); virStorageVolDefPtr vol = NULL; char *devpath = NULL; + char *key = NULL; + char *target_port = NULL; int retval = -1; /* Check if the pool is using a stable target path. The call to @@ -3877,9 +3922,21 @@ virStorageBackendSCSINewLun(virStoragePoolObjPtr pool, VIR_STORAGE_VOL_READ_NOERROR)) < 0) goto cleanup; - if (!(vol->key = virStorageBackendSCSISerial(vol->target.path))) + if (!(key = virStorageBackendSCSISerial(vol->target.path))) goto cleanup; + if (def->source.adapter.type == VIR_STORAGE_ADAPTER_TYPE_FC_HOST && + STRNEQ(key, vol->target.path)) { + /* NPIV based LUNs use the same "serial" key. In order to distinguish + * we need to append a port value */ + if (!(target_port = virStorageBackendSCSITargetPort(vol->target.path))) + goto cleanup; + if (virAsprintf(&vol->key, "%s_PORT%s", key, target_port) < 0) + goto cleanup; + } else { + VIR_STEAL_PTR(vol->key, key); + } + def->capacity += vol->target.capacity; def->allocation += vol->target.allocation; @@ -3892,6 +3949,8 @@ virStorageBackendSCSINewLun(virStoragePoolObjPtr pool, cleanup: virStorageVolDefFree(vol); VIR_FREE(devpath); + VIR_FREE(target_port); + VIR_FREE(key); return retval; } -- 2.17.2

3 7

[libvirt] [PATCH v2] qemu: Assign device address before qemuDomainSupportsNicdev
by Wang Yechao 17 Jan '19

17 Jan '19

qemuDomainSupportsNicdev will check the device address type on aarch64. If it is invoked before device address assigned, hotadd vhostuser interface with no address specified will get error. Let qemuDomainEnsurePCIAddress run before qemuDomainSupportsNicdev. Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> --- v1 patch: https://www.redhat.com/archives/libvir-list/2018-December/msg00435.html Changes in v2: - do not modify the address type, let qemuDomainEnsurePCIAddress do it. --- src/qemu/qemu_hotplug.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 8da0233..f25c8db 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1369,6 +1369,25 @@ qemuDomainAttachNetDevice(virQEMUDriverPtr driver, if (qemuAssignDeviceNetAlias(vm->def, net, -1) < 0) goto cleanup; + if (qemuDomainIsS390CCW(vm->def) && + net->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI && + virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_CCW)) { + net->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCW; + if (!(ccwaddrs = virDomainCCWAddressSetCreateFromDomain(vm->def))) + goto cleanup; + if (virDomainCCWAddressAssign(&net->info, ccwaddrs, + !net->info.addr.ccw.assigned) < 0) + goto cleanup; + } else if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_VIRTIO_S390)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("virtio-s390 net device cannot be hotplugged.")); + goto cleanup; + } else if (qemuDomainEnsurePCIAddress(vm, &dev, driver) < 0) { + goto cleanup; + } + + releaseaddr = true; + switch (actualType) { case VIR_DOMAIN_NET_TYPE_BRIDGE: case VIR_DOMAIN_NET_TYPE_NETWORK: @@ -1503,25 +1522,6 @@ qemuDomainAttachNetDevice(virQEMUDriverPtr driver, goto cleanup; } - if (qemuDomainIsS390CCW(vm->def) && - net->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI && - virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_CCW)) { - net->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCW; - if (!(ccwaddrs = virDomainCCWAddressSetCreateFromDomain(vm->def))) - goto cleanup; - if (virDomainCCWAddressAssign(&net->info, ccwaddrs, - !net->info.addr.ccw.assigned) < 0) - goto cleanup; - } else if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_VIRTIO_S390)) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("virtio-s390 net device cannot be hotplugged.")); - goto cleanup; - } else if (qemuDomainEnsurePCIAddress(vm, &dev, driver) < 0) { - goto cleanup; - } - - releaseaddr = true; - if (VIR_ALLOC_N(tapfdName, tapfdSize) < 0 || VIR_ALLOC_N(vhostfdName, vhostfdSize) < 0) goto cleanup; -- 1.8.3.1

3 4

[libvirt] ANNOUNCE: libguestfs 1.40 released
by Richard W.M. Jones 17 Jan '19

17 Jan '19

I'm pleased to announce libguestfs 1.40, a library and a set of tools for accessing and modifying virtual machine disk images. This release represents about a year of work by many contributors. I'd like to call out in particular substantial contributions from: Pino Toscano, Tomáš Golembiovský, Nir Soffer and Nikolay Ivanets. See the release notes below for full details. You can get libguestfs 1.40 from here: Main website: http://libguestfs.org/ [not updated yet] Source: http://libguestfs.org/download/1.40-stable/ Fedora: https://koji.fedoraproject.org/koji/packageinfo?packageID=8391 Debian/experimental: https://packages.debian.org/libguestfs0 Rich. ---------------------------------------------------------------------- Release notes for libguestfs 1.40 These are also available online at: http://libguestfs.org/guestfs-release-notes.1.html New features New features in existing tools Virt-inspector now displays the libosinfo short ID for guests (Pino Toscano). Guestfish -N will now generate 1G disks instead of 100M disks by default. Virt-resize supports f2fs filesystems (Pino Toscano). libguestfs-test-tool now supports bash tab completion (Pino Toscano). The --machine-readable option now supports sending output to files or stdout/stderr. This works uniformly across most OCaml-based virt tools, specifically: virt-builder, virt-builder-repository, virt-dib, virt-get-kernel, virt-resize, virt-sparsify, and virt-v2v (Pino Toscano). Virt-builder --uninstall option now works on SUSE (Sebastian Meyer). Virt-builder now supports Windows. We are not able to ship Windows templates in the public respository for obvious licensing reasons, but if you are an MSDN subscriber you may build your own. See https://rwmj.wordpress.com/2018/09/13/creating-windows-templates-for-virt-b… Many tools now support a --key option allowing you to pass in decryption keys for filesystems stored in local files on the host rather than having to feed them in over stdin (Pino Toscano). virt-v2v and virt-p2v New -o rhv-upload mode for directly uploading images to RHV, bypassing the Export Storage Domain (Tomáš Golembiovský, Nir Soffer, Daniel Erez, Pino Toscano). New -o openstack mode for directly uploading images to OpenStack and Cinder using OpenStack APIs. Virt-v2v now has a general mechanism for input and output options: -io and -oo. Virt-v2v can now install the RHV tools or QEMU GA in guests (Tomáš Golembiovský). The huge manual has now been split into several smaller documents and is more focused on helping users to accomplish specific v2v tasks. Multiple improvements to the OVF metadata when converting to RHV (Tomáš Golembiovský, Pino Toscano, Arik Hadas). Virt-v2v can now convert Linux guests with split kernel packages, especially Ubuntu 18.04 (Pino Toscano). Virt-v2v old --password-file option has been replaced by -ip (the old option remains for backwards compatibility). Virt-v2v now preserves the VM Generation ID. Virt-v2v has a new --mac option allowing specific NICs to be mapped precisely to networks or bridges on the target. New virt-v2v --print-estimate option to estimate the size of data that virt-v2v will copy. Virt-v2v is now usually able to remove open-vm-tools and VMware Tools from the Linux guest during conversion (Pino Toscano). Virt-v2v can now support UEFI conversions to RHV (Tomáš Golembiovský). Virt-p2v now prefers you to shut down instead of rebooting the machine after conversion has finished (Pino Toscano). Virt-v2v now writes the libosinfo short ID to the libvirt metadata when using -o local and -o libvirt output modes (Martin Kletzander). Language bindings OpenJDK 10+ is supported (Pino Toscano). Java bindings fixed for Gentoo host (Martin Kletzander). Inspection Support Kali Linux (Pino Toscano). When inspecting mountpoints, look for /etc/mdadm/mdadm.conf as well as /etc/mdadm.conf (Nikolay Ivanets). Improved support for OpenSUSE Tumbleweed (Pino Toscano). Architectures and platforms Miscellaneous macOS build fixes (Adam Robinson). Other Multiple documentation typos fixed (Yuri Chornoivan). Security See also guestfs-security(1). CVE-2018-11806 Qemu's slirp (userspace networking) had several buffer overflows which could be triggered from the guest or network side. API New APIs "f2fs_expand" Expand an f2fs filesystem (Pino Toscano). "inspect_get_osinfo" Get the libosinfo short ID for the inspected guest (Pino Toscano). "lvm_scan" This rescans all PVs, VGs and LVs, optionally activating them. Other API changes "part_get_mbr_part_type" was reimplemented in OCaml (Nikolay Ivanets). "list_filesystems" now filters out MBR extended partitions (Nikolay Ivanets). "vgscan" has been deprecated. Use the new "lvm_scan" API in preference. Build changes The JSON library switched from yajl to Jansson ≥ 2.7 (Pino Toscano). Support for The Sleuth Kit 4.5 (Pino Toscano). Multiple improvements in the "./configure" script (Lin Ma). More out-of-tree build fixes (Hilko Bengen). Some work on reproducible builds (Bernhard M. Wiedemann). Internals lvmetad is now used unconditionally, and is started much earlier in the appliance boot process. This fixes support for certain PV types. On Debian, dash is installed inside the appliance and in virt-rescue (Pino Toscano). The amount of RAM assigned to the appliance has been increased from 500M to 768M (on x86_64, other architectures have different limits). This enables use of up to 255 disks when using recent Linux kernels. Virt-v2v now models the source machine type (eg. i440FX, Q35 or virt). Virt-p2v now generates configuration code automatically, also code for parsing the kernel command line. Use "LT_INIT" instead of "AC_PROG_LIBTOOL". The OCaml "JSON" code for parsing and serializing has been unified into a single abstract data type. Most OCaml tools are now built using the PIC runtime. This has a performance penalty on i686, but is relatively free on other architectures and has security benefits. The direct backend now queries the qemu binary for availability of KVM, instead of using a heuristic based on /dev/kvm (Andrea Bolognani). Our "clever" libxml2 writer macros are now used consistently across all parts of the code base. Qemu dropped -nodefconfig without telling us. The equivalent -no-user-config option is now used instead. Qemu also moved the "serial" option from -drive to -device, again without any notice, and this is also fixed. Bugs fixed https://bugzilla.redhat.com/1664310 [RHEL 7.6 LP] openstack output leaks passwords https://bugzilla.redhat.com/1661038 virt-inspector fails with "error: int_of_string" on a Linux image when /etc/fstab contains a partionless device https://bugzilla.redhat.com/1651432 v2v - support for -insecure option to support OSP SSL connection for VM migrations https://bugzilla.redhat.com/1642044 [RFE] virt-p2v UI should give the option to shutdown after conversion is finished https://bugzilla.redhat.com/1634248 virt-v2v cannot convert opensuse15 guest https://bugzilla.redhat.com/1624902 Fix rhev-apt command that virt-v2v runs in Windows guests on first boot https://bugzilla.redhat.com/1624878 Update v2v docs to describe support for SHA 2 certs required for converting Windows 7 and 2008 R2 guests https://bugzilla.redhat.com/1615885 Minor log issue of virt-v2v https://bugzilla.redhat.com/1614276 VixDiskLib: VixDiskLib_Read: Read 4096 sectors at 57516160 failed. Error 2 (Memory allocation failed. Out of memory.) (DiskLib error 802: NBD_ERR_INSUFFICIENT_RESOURCES) at 5240. https://bugzilla.redhat.com/1612785 "warning: <target dev='sr0'> was ignored because the device name could not be recognized" should be hidden since convert cdrom is supported by virt-v2v https://bugzilla.redhat.com/1611690 part_to_dev "/dev/sdp1" returns "/dev/sd" instead of "/dev/sdp" https://bugzilla.redhat.com/1608131 inspection fails when swap partition in /etc/fstab has incorrect UUID compared to actual swap partition https://bugzilla.redhat.com/1605071 On machines where /dev/kvm exists but KVM doesn't work, libguestfs will not fall back to TCG https://bugzilla.redhat.com/1602353 virt-inspector can't inspect LUKS-encrypted RHEL7 guest image https://bugzilla.redhat.com/1601943 Improve error "No module named ovirtsdk4" in v2v rhv-upload conversion https://bugzilla.redhat.com/1598715 Adding QXL device in OVF causes Cirrus Logic device to be added to guest https://bugzilla.redhat.com/1598350 RFE: virt-v2v should preserve <genid> https://bugzilla.redhat.com/1596851 Transfer fails if local host is in maintenance mode https://bugzilla.redhat.com/1596810 Transfer fails if local host belongs to another DC https://bugzilla.redhat.com/1592468 v2v to RHV transfer fails with: error: [empty name]: cannot read '//*/disksection' with value: null https://bugzilla.redhat.com/1592061 virt-v2v convert rhel5.3 failure when executing rpm -ql kernel-2.6.18-128.el5 command https://bugzilla.redhat.com/1591789 p2v: error: XML error: CPU vendor specified without CPU model https://bugzilla.redhat.com/1590220 Hide rhv-upload option in virt-p2v client https://bugzilla.redhat.com/1588451 Don't use relative socket paths for NBD https://bugzilla.redhat.com/1588088 RFE: Support for -o rhv-upload via Unix domain socket https://bugzilla.redhat.com/1586198 ovirt-imageio-daemon times out during migration https://bugzilla.redhat.com/1580309 virt-v2v rhel7.6 build can't convert guest to null with qemu-kvm rhel7.5.z build https://bugzilla.redhat.com/1580292 Virt-v2v rhel7.6 build can't convert guest with qemu-kvm rhel7.5.z build https://bugzilla.redhat.com/1579047 virt-builder: ubuntu-18.04 template has stray HTTP proxy configured for apt https://bugzilla.redhat.com/1575640 virt-builder: error: host cpu (x86_64) and guest arch (unknown) are not compatible https://bugzilla.redhat.com/1570407 virt-v2v -i ova gives checksum error when missing files are mentioned in .mf https://bugzilla.redhat.com/1567763 guestfsd: error: /Windows/Drivers/VirtIO: Read-only file system https://bugzilla.redhat.com/1564983 The mount-vfs failed to mount the logical volume https://bugzilla.redhat.com/1561828 RHV tools not properly installed in migrated Windows VM's causing boot failure https://bugzilla.redhat.com/1559027 virt-v2v does not honor --network with -o rhv-upload https://bugzilla.redhat.com/1557273 [RFE] Upload images directly to oVirt (virt-v2v -o rhv-upload) https://bugzilla.redhat.com/1544842 [RFE] Add libosinfo short-id to virt-inspector output https://bugzilla.redhat.com/1544674 virt-inspector --help differs from manpage https://bugzilla.redhat.com/1544227 RFE: recognize the Kali Linux distribution https://bugzilla.redhat.com/1532224 Resume=/dev/sdaX can't be updated to resume=/dev/vdaX in rhel7 guest's grub file after v2v conversion https://bugzilla.redhat.com/1530967 Change categories of "Guest Operating system management by VIX API" in v2v man page https://bugzilla.redhat.com/1527334 Various -it vddk -io vddk* options should be documented, removed or fixed https://bugzilla.redhat.com/1148012 Reverted: RFE: Allow qemu-bridge-helper to be used to implement guestfs_set_network -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com Fedora Windows cross-compiler. Compile Windows programs, test, and build Windows installers. Over 100 libraries supported. http://fedoraproject.org/wiki/MinGW

1 0

[libvirt] [PATCH RFC 00/51] refactor block job handling and add support for new event/commands (blockdev-add saga)
by Peter Krempa 17 Jan '19

17 Jan '19

This series refactors how libvirt stores block job data and handles block job events in general. This is done in preparation for -blockdev support where we'll need to add support for the 'blockdev-create' job which is not associated to any frontend 'disk' in some cases. This series introduces a new structure and table of blockjobs currently running and implements handlers for the new even "JOB_STATUS_CHANGE". RFC status excuses: - the series needs much more testing than I did - the last few patches are true RFC - I didn't port/refactor the rest of the blockjob finishing code that I have for the individual job types after the big refactor so there aren't any examples for the -blockdev case yet. I'll post the link to a repo containing the patches some time later. Peter Krempa (51): qemu: blockjob: Extract emitting of libvirt events qemu: blockjob: Emit VIR_DOMAIN_EVENT_ID_BLOCK_JOB only for local disks qemu: processBlockJobEvent: Use qemuBlockJobUpdate to process block job events qemu: blockjob: Unexport qemuBlockJobEventProcess qemu: blockjob: Rename public APIs qemu: blockjob: Remove header dependency on qemu_domain.h qemu: Consolidate disk blockjob variables into a structure qemu: process: Consolidate error paths in qemuProcessHandleBlockJob qemu: blockjob: Rename qemuBlockJobEventProcess to qemuBlockJobEventProcessLegacy qemu: blockjob: Split out handling of comlpleted jobs qemu: migration: Properly note that non-shared-storage migration uses a blockjob qemu: process: refresh block jobs on reconnect qemu: driver: Remove block job status reprobing from qemuDomainBlockPivot qemu: migration: Simplify cancellation of migration blockjobs qemu: blockjob: Turn struct qemuBlockJobData into a virObject qemu: blockjob: Clarify that job 'status' field contains new state qemu: migration: Separate startup of disk mirror from migration logic qemu: blockjob: Add functions for block job state control qemu: blockjob: Drop unnecessary calls to qemuBlockJobSyncEndDisk qemu: blockjob: Add reference to disk into struct qemuBlockJobData qemu: blockjob: Record job type when starting the job qemu: blockjob: Pass in job to qemuBlockJobEventProcessLegacy qemu: blockjob: Convert qemuBlockJobSyncBeginDisk to work with the job qemu: blockjob: Track current state of blockjob qemu: migration: Extract reporting of disk migration error qemu: blockjob: Remove error propagation from qemuBlockJobUpdateDisk qemu: blockjob: Consume new block job state in the processing function qemu: blockjob: Pass job into qemuBlockJobUpdateDisk and rename it qemu: Allocate diskPriv->blockjob only when there's a blockjob qemu: migration: Don't call qemuBlockJobSyncEndDisk when block job has terminated qemu: blockjob: Convert qemuBlockJobSyncEndDisk to take job instead of disk qemu: blockjob: Add job name into the data qemu: domain: Add global table of blockjobs qemu: blockjob: Register new and running blockjobs in the global table qemu: blockjob: Add string convertors for blockjob type and state enums qemu: domain: Store blockjob data in the status XML tests: qemustatusxml2xml: Add test case for block job tracking qemu: monitor: Add new fields for 'block-stream' command qemu: monitor: Add new fields for 'block-commit' command qemu: monitor: Add new fields for 'blockdev-mirror' command qemu: monitor: Add support for 'job-dismiss' command qemu: monitor: Add support for 'job-cancel' command qemu: monitor: Add support for 'job-complete' command qemu: monitor: Add infrastructure for 'query-jobs' qemu: blockjob: Add 'concluded' state for a block job qemu: monitor: Implement support for 'JOB_STATUS_CHANGE' event qemu: process: Don't trigger BLOCK_JOB* events with -blockdev qemu: blockjob: Add helper to convert monitor job status to internal state qemu: Add handler for job state change event qemu: blockjob: Add modern block job event handler qemu: process: Refresh -blockdev based blockjobs on reconnect to qemu src/qemu/qemu_blockjob.c | 687 ++++++++++++++---- src/qemu/qemu_blockjob.h | 111 ++- src/qemu/qemu_domain.c | 123 +++- src/qemu/qemu_domain.h | 13 +- src/qemu/qemu_driver.c | 114 ++- src/qemu/qemu_migration.c | 233 +++--- src/qemu/qemu_monitor.c | 120 ++- src/qemu/qemu_monitor.h | 82 ++- src/qemu/qemu_monitor_json.c | 239 +++++- src/qemu/qemu_monitor_json.h | 28 +- src/qemu/qemu_process.c | 179 ++++- .../query-jobs-create.json | 20 + .../query-jobs-create.result | 11 + .../qemumonitorjsondata/query-jobs-empty.json | 1 + .../query-jobs-empty.result | 0 tests/qemumonitorjsontest.c | 101 ++- .../blockjob-blockdev-in.xml | 364 ++++++++++ .../blockjob-blockdev-out.xml | 1 + tests/qemuxml2xmltest.c | 2 + 19 files changed, 2097 insertions(+), 332 deletions(-) create mode 100644 tests/qemumonitorjsondata/query-jobs-create.json create mode 100644 tests/qemumonitorjsondata/query-jobs-create.result create mode 100644 tests/qemumonitorjsondata/query-jobs-empty.json create mode 100644 tests/qemumonitorjsondata/query-jobs-empty.result create mode 100644 tests/qemustatusxml2xmldata/blockjob-blockdev-in.xml create mode 120000 tests/qemustatusxml2xmldata/blockjob-blockdev-out.xml -- 2.19.2

2 105

[libvirt] [PATCH 0/5] network: fix networking for firewalld+nftables
by Laine Stump 17 Jan '19

17 Jan '19

The detailed explanation of this is in Patch 4/5. Basically, when firewalld enables their new nftables backend, libvirt virtual networks lose all ability to forward packets from guests out to the physical network, and can only communicate with the host itself as much as firewalld's "public" zone will allow (which isn't much, and doesn't include DHCP or DNS). Laine Stump (5): docs: add forgotten mentions of forward mode "open" util: move all firewalld-specific stuff into its own file util: new function virFirewallDInterfaceSetZone() network: regain guest network connectivity after firewalld switch to nftables network: allow configuring firewalld zone for virtual network bridge device docs/formatnetwork.html.in | 21 ++- docs/news.xml | 40 ++++++ docs/schemas/basictypes.rng | 6 + docs/schemas/network.rng | 6 + include/libvirt/virterror.h | 1 + libvirt.spec.in | 16 +++ src/conf/network_conf.c | 14 +- src/conf/network_conf.h | 1 + src/libvirt_private.syms | 4 + src/network/Makefile.inc.am | 10 +- src/network/bridge_driver_linux.c | 25 ++++ src/network/libvirt.zone | 14 ++ src/util/Makefile.inc.am | 2 + src/util/virerror.c | 1 + src/util/virfirewall.c | 86 +----------- src/util/virfirewalld.c | 151 +++++++++++++++++++++ src/util/virfirewalld.h | 36 +++++ src/util/virfirewallpriv.h | 2 - tests/networkxml2xmlin/routed-network.xml | 2 +- tests/networkxml2xmlout/routed-network.xml | 2 +- tests/virfirewalltest.c | 1 + 21 files changed, 350 insertions(+), 91 deletions(-) create mode 100644 src/network/libvirt.zone create mode 100644 src/util/virfirewalld.c create mode 100644 src/util/virfirewalld.h -- 2.20.1

4 21

[libvirt] [PATCH] qemu: Update a domain's persistent definition file if it's defined
by Wang Yechao 17 Jan '19

17 Jan '19

From: Li XueLei <li.xuelei1(a)zte.com.cn> During making disk snapshot in an active domain, sometimes we should update the domain's persistent definition.We must check if the domain is defined, before we update the persistent definition file. First,we create a vm.Then,define the vm and undefine the vm.Last create a --memspec snapshot for the vm.Now,we will find a persistent definition file in the config directory if we don't use this patch. Signed-off-by: Li XueLei <li.xuelei1(a)zte.com.cn> --- src/qemu/qemu_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index e32257f..94dcb97 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -15330,7 +15330,7 @@ qemuDomainSnapshotCreateDiskActive(virQEMUDriverPtr driver, if (ret == 0 || !do_transaction) { if (virDomainSaveStatus(driver->xmlopt, cfg->stateDir, vm, driver->caps) < 0 || - (persist && virDomainSaveConfig(cfg->configDir, driver->caps, + (vm->persistent && persist && virDomainSaveConfig(cfg->configDir, driver->caps, vm->newDef) < 0)) ret = -1; } -- 1.8.3.1

2 1

[libvirt] [PATCH 0/6] RFC: qemu: virtio-{non-}transitional support
by Cole Robinson 17 Jan '19

17 Jan '19

This series adds the beginnings of support for virtio-transitional and virtio-non-transitional qemu devices. qemu patches, queued for qemu 4.0.0: https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg00923.html Previous libvirt discussion around this: https://www.redhat.com/archives/libvir-list/2018-August/msg01073.html Long story short we need to expose these options so apps have a usable way to support rhel6 + virtio + q35. This series only covers exposing the associated device models for disk and rng devices to make sure I'm on the right track. serial, net, scsi, input-host, balloon 9p, vsock, vhost-scsi still need to be implemented. Those should follow the rng example, except vhost-scsi may need to be a different approach. The main RFC bits here are: * The disk approach. danpb and I briefly discussed on IRC adding new bus= values vs a new model= attribute. We decided model= is the lesser of two evils, since bus= handling in apps is tied with target= generation, so adding new virtio-X bus= values will cause more work for apps. These patches add a <disk model=X/> attribute * The XML and naming. Previous discussions seemed to favor adding new model-style values rather than a 'transitional' attribute or similar. So these patches add model='virtio-transitional' and model='virtio-non-transitional' * The PCI address handling. I just mapped virtio-non-transitional to imply plain PCI addressing. I think that's all we need but I'm not positive so I'd appreciate a review of that approach. Cole Robinson (6): tests: Add capabilities data for QEMU 4.0.0 x86_64 tests: qemuxml2xml: Add basic DO_TEST_CAPS impl conf: Add <disk model='virtio-{non-}transitional'/> qemu: Wire up disk model=virtio-{non-}transitional qemu: domcaps: Report disk <enum name="model"> qemu: Support rng model=virtio-{non-}transitional docs/formatdomain.html.in | 10 + docs/schemas/domaincommon.rng | 14 +- src/conf/domain_capabilities.c | 1 + src/conf/domain_capabilities.h | 1 + src/conf/domain_conf.c | 32 +- src/conf/domain_conf.h | 12 + src/libvirt_private.syms | 2 + src/qemu/qemu_capabilities.c | 17 + src/qemu/qemu_capabilities.h | 6 + src/qemu/qemu_command.c | 43 +- src/qemu/qemu_domain_address.c | 8 +- .../bhyve_basic.x86_64.xml | 1 + .../bhyve_fbuf.x86_64.xml | 1 + .../bhyve_uefi.x86_64.xml | 1 + tests/domaincapsschemadata/full.xml | 5 + .../domaincapsschemadata/libxl-xenfv-usb.xml | 1 + .../domaincapsschemadata/libxl-xenpv-usb.xml | 1 + .../qemu_1.7.0.x86_64.xml | 1 + .../qemu_2.12.0-virt.aarch64.xml | 1 + .../qemu_2.12.0.ppc64.xml | 1 + .../qemu_2.12.0.s390x.xml | 1 + .../qemu_2.12.0.x86_64.xml | 1 + .../qemu_2.6.0-virt.aarch64.xml | 1 + .../qemu_2.6.0.aarch64.xml | 1 + .../domaincapsschemadata/qemu_2.6.0.ppc64.xml | 1 + .../qemu_2.6.0.x86_64.xml | 1 + .../domaincapsschemadata/qemu_2.7.0.s390x.xml | 1 + .../qemu_2.8.0-tcg.x86_64.xml | 1 + .../domaincapsschemadata/qemu_2.8.0.s390x.xml | 1 + .../qemu_2.8.0.x86_64.xml | 1 + .../qemu_2.9.0-q35.x86_64.xml | 1 + .../qemu_2.9.0-tcg.x86_64.xml | 1 + .../qemu_2.9.0.x86_64.xml | 1 + .../domaincapsschemadata/qemu_3.0.0.s390x.xml | 1 + .../qemu_4.0.0.x86_64.xml | 153 + tests/domaincapstest.c | 4 + .../caps_4.0.0.x86_64.replies | 23180 ++++++++++++++++ .../caps_4.0.0.x86_64.xml | 1388 + tests/qemucapabilitiestest.c | 1 + ...virtio-non-transitional.x86_64-latest.args | 37 + .../virtio-non-transitional.xml | 29 + .../virtio-transitional.x86_64-latest.args | 37 + .../qemuxml2argvdata/virtio-transitional.xml | 29 + tests/qemuxml2argvtest.c | 3 + .../virtio-non-transitional.xml | 50 + .../virtio-transitional.xml | 51 + tests/qemuxml2xmltest.c | 60 +- 47 files changed, 25172 insertions(+), 23 deletions(-) create mode 100644 tests/domaincapsschemadata/qemu_4.0.0.x86_64.xml create mode 100644 tests/qemucapabilitiesdata/caps_4.0.0.x86_64.replies create mode 100644 tests/qemucapabilitiesdata/caps_4.0.0.x86_64.xml create mode 100644 tests/qemuxml2argvdata/virtio-non-transitional.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/virtio-non-transitional.xml create mode 100644 tests/qemuxml2argvdata/virtio-transitional.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/virtio-transitional.xml create mode 100644 tests/qemuxml2xmloutdata/virtio-non-transitional.xml create mode 100644 tests/qemuxml2xmloutdata/virtio-transitional.xml -- 2.20.1

5 39

[libvirt] [PATCH v2] util: audit: Fix logging an error when kernel lacks audit support
by Erik Skultety 16 Jan '19

16 Jan '19

Based on an upstream discussion, reporting the errno is useful for the user to know why audit isn't supported. Even though having an error in the logs might look concerning when 'audit_log=1', it also denotes that audit is only going to be used if it's available, continuing normally if it's unavailable for whatever reason. Partially reverts commit 4199c2f221c. https://bugzilla.redhat.com/show_bug.cgi?id=1596119 Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- src/remote/remote_daemon.c | 2 +- src/util/viraudit.c | 16 ++-------------- src/util/viraudit.h | 2 +- 3 files changed, 4 insertions(+), 16 deletions(-) diff --git a/src/remote/remote_daemon.c b/src/remote/remote_daemon.c index 3be3ad02fc..ededef97b4 100644 --- a/src/remote/remote_daemon.c +++ b/src/remote/remote_daemon.c @@ -1380,7 +1380,7 @@ int main(int argc, char **argv) { if (config->audit_level) { VIR_DEBUG("Attempting to configure auditing subsystem"); - if (virAuditOpen(config->audit_level) < 0) { + if (virAuditOpen() < 0) { if (config->audit_level > 1) { ret = VIR_DAEMON_ERR_AUDIT; goto cleanup; diff --git a/src/util/viraudit.c b/src/util/viraudit.c index a02e5b36fd..135d0e626a 100644 --- a/src/util/viraudit.c +++ b/src/util/viraudit.c @@ -54,23 +54,11 @@ static int auditfd = -1; #endif static bool auditlog; -int virAuditOpen(unsigned int audit_level ATTRIBUTE_UNUSED) +int virAuditOpen(void) { #if WITH_AUDIT if ((auditfd = audit_open()) < 0) { - /* You get these error codes only when the kernel does not - * have audit compiled in or it's disabled (e.g. by the kernel - * cmdline) */ - if (errno == EINVAL || errno == EPROTONOSUPPORT || - errno == EAFNOSUPPORT) { - if (audit_level < 2) - VIR_INFO("Audit is not supported by the kernel"); - else - virReportError(VIR_FROM_THIS, "%s", _("Audit is not supported by the kernel")); - } else { - virReportSystemError(errno, "%s", _("Unable to initialize audit layer")); - } - + virReportSystemError(errno, "%s", _("Unable to initialize audit layer")); return -1; } diff --git a/src/util/viraudit.h b/src/util/viraudit.h index 66605b16b5..7fbc28ba9b 100644 --- a/src/util/viraudit.h +++ b/src/util/viraudit.h @@ -31,7 +31,7 @@ typedef enum { VIR_AUDIT_RECORD_RESOURCE, } virAuditRecordType; -int virAuditOpen(unsigned int audit_level); +int virAuditOpen(void); void virAuditLog(bool enabled); -- 2.20.1

2 2