- Devel - libvirt List Archives

[libvirt] [PATCH 00/10] qemu: Refactor disk hotplug code for better reuse with blockdev (blockdev-add saga)
by Peter Krempa 25 Jun '19

25 Jun '19

Move around few of the helpers which deal with disk addition and removal and fix a potential bug in the blockdev media change routine. The refactors will allow simplify a corner case of using blockdev such as if the disk frontend is unplugged the backends can be unplugged only after blockjobs terminate. This will allow simpler handling of the backing chain data in conjunction with blockjobs. To test the blockdev part you can use the new XML namespace element for qemu: <domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'> [...] <qemu:capabilities> <qemu:add capability='blockdev'/> </qemu:capabilities> if you have a recent enough qemu. Peter Krempa (10): qemu: block: Move and rename qemuHotplugRemoveStorageSourcePrepareData qemu: hotplug: Handle copy-on-read filter separate from rest of backing chain qemu: Introduce new set of helpers for attaching and detaching storage chains qemu: command: Use storage chain helpers in commandline generator qemu: command: Use VIR_AUTO infrastructure in qemuBuildDiskSourceCommandLine qemu: command: get rid of 'cleanup' in qemuBuildDiskSourceCommandLine qemu: hotplug: Use storage chain helpers in qemuDomainAttachDiskGeneric qemu: hotplug: Use storage chain helpers in qemuDomainRemoveDiskDevice qemu: hotplug: qemu: hotplug: Use storage chain helpers in qemuDomainChangeMediaBlockdev qemu: hotplug: Remove rest of source backend if hotplug fails src/qemu/qemu_block.c | 174 +++++++++++++++++++++ src/qemu/qemu_block.h | 33 ++++ src/qemu/qemu_command.c | 132 ++++++++++------ src/qemu/qemu_command.h | 11 ++ src/qemu/qemu_hotplug.c | 332 +++++++--------------------------------- 5 files changed, 361 insertions(+), 321 deletions(-) -- 2.21.0

2 21

[libvirt] [PATCH] conf: storage: also sanitize source dir
by Ján Tomko 25 Jun '19

25 Jun '19

Commit a7fb2258 added sanitization of storage pool target paths, however source dir paths were left unsanitized. A netfs pool with: <source> <host name='10.20.30.40'/> <dir path='/nfs/'/> </source> will not be correctly detected as mounted by virStorageBackendFileSystemIsMounted, because it shows up in the mount list without the trailing slash. Sanitize the source dir as well. https://bugzilla.redhat.com/show_bug.cgi?id=1723247 Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- src/conf/storage_conf.c | 5 ++++- .../storagepoolxml2xmlin/pool-netfs-slash.xml | 20 +++++++++++++++++++ .../pool-netfs-slash.xml | 20 +++++++++++++++++++ tests/storagepoolxml2xmltest.c | 1 + 4 files changed, 45 insertions(+), 1 deletion(-) create mode 100644 tests/storagepoolxml2xmlin/pool-netfs-slash.xml create mode 100644 tests/storagepoolxml2xmlout/pool-netfs-slash.xml diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 29988c36c4..397bd66870 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -533,6 +533,7 @@ virStoragePoolDefParseSource(xmlXPathContextPtr ctxt, VIR_AUTOFREE(char *) port = NULL; VIR_AUTOFREE(char *) ver = NULL; VIR_AUTOFREE(xmlNodePtr *) nodeset = NULL; + VIR_AUTOFREE(char *) sourcedir = NULL; relnode = ctxt->node; ctxt->node = node; @@ -630,7 +631,9 @@ virStoragePoolDefParseSource(xmlXPathContextPtr ctxt, } - source->dir = virXPathString("string(./dir/@path)", ctxt); + sourcedir = virXPathString("string(./dir/@path)", ctxt); + if (sourcedir) + source->dir = virFileSanitizePath(sourcedir); /* In gluster, a missing dir defaults to "/" */ if (!source->dir && pool_type == VIR_STORAGE_POOL_GLUSTER && VIR_STRDUP(source->dir, "/") < 0) diff --git a/tests/storagepoolxml2xmlin/pool-netfs-slash.xml b/tests/storagepoolxml2xmlin/pool-netfs-slash.xml new file mode 100644 index 0000000000..63f5d401ba --- /dev/null +++ b/tests/storagepoolxml2xmlin/pool-netfs-slash.xml @@ -0,0 +1,20 @@ +<pool type='netfs'> + <name>nfsimages</name> + <uuid>7641d5a8-af11-f730-a34e-0a7dfcede71f</uuid> + <capacity>0</capacity> + <allocation>0</allocation> + <available>0</available> + <source> + <host name='localhost'/> + <dir path='/var/lib/libvirt/images/'/> + <format type='nfs'/> + </source> + <target> + <path>/mnt</path> + <permissions> + <mode>0700</mode> + <owner>0</owner> + <group>0</group> + </permissions> + </target> +</pool> diff --git a/tests/storagepoolxml2xmlout/pool-netfs-slash.xml b/tests/storagepoolxml2xmlout/pool-netfs-slash.xml new file mode 100644 index 0000000000..8a5f272137 --- /dev/null +++ b/tests/storagepoolxml2xmlout/pool-netfs-slash.xml @@ -0,0 +1,20 @@ +<pool type='netfs'> + <name>nfsimages</name> + <uuid>7641d5a8-af11-f730-a34e-0a7dfcede71f</uuid> + <capacity unit='bytes'>0</capacity> + <allocation unit='bytes'>0</allocation> + <available unit='bytes'>0</available> + <source> + <host name='localhost'/> + <dir path='/var/lib/libvirt/images'/> + <format type='nfs'/> + </source> + <target> + <path>/mnt</path> + <permissions> + <mode>0700</mode> + <owner>0</owner> + <group>0</group> + </permissions> + </target> +</pool> diff --git a/tests/storagepoolxml2xmltest.c b/tests/storagepoolxml2xmltest.c index b6f4cb4226..96dd04faec 100644 --- a/tests/storagepoolxml2xmltest.c +++ b/tests/storagepoolxml2xmltest.c @@ -73,6 +73,7 @@ mymain(void) DO_TEST("pool-iscsi"); DO_TEST("pool-iscsi-auth"); DO_TEST("pool-netfs"); + DO_TEST("pool-netfs-slash"); DO_TEST("pool-netfs-auto"); DO_TEST("pool-netfs-protocol-ver"); DO_TEST("pool-netfs-gluster"); -- 2.20.1

2 1

[libvirt] [PATCH 0/3] error: Introdude and use VIR_ERR_DEPRECATED error code
by Peter Krempa 25 Jun '19

25 Jun '19

Peter Krempa (3): error: Add VIR_ERR_DEPRECATED error code qemu: Use VIR_ERR_DEPRECATED in QemuAttach and DomainXMLFromNative stubs news: Mention VIR_ERR_DEPRECATED in improvements docs/news.xml | 10 ++++++++++ include/libvirt/virterror.h | 1 + src/qemu/qemu_driver.c | 4 ++-- src/util/virerror.c | 4 ++++ 4 files changed, 17 insertions(+), 2 deletions(-) -- 2.21.0

2 5

[libvirt] [PATCH 0/3] qemu: Fix handling of 'dir' volumes with disk type='volume'
by Peter Krempa 25 Jun '19

25 Jun '19

Peter Krempa (3): util: storage: Fix virStorageSourceGetActualType if volume was not translated qemu: domain: Allow 'VIR_STORAGE_TYPE_VOLUME' disks with 'fat' format qemu: Supply correct default type for 'dir' based VIR_STORAGE_TYPE_VOLUME src/qemu/qemu_domain.c | 13 ++++++++++--- src/util/virstoragefile.c | 12 +++++++++++- tests/qemuxml2argvdata/disk-source-pool.args | 8 ++++++++ tests/qemuxml2argvdata/disk-source-pool.xml | 16 +++++++++++++++- tests/qemuxml2xmloutdata/disk-source-pool.xml | 14 ++++++++++++++ 5 files changed, 58 insertions(+), 5 deletions(-) -- 2.21.0

1 4

[libvirt] Plans for next release
by Daniel Veillard 25 Jun '19

25 Jun '19

Ideally if we want to push the next release on Monday, we should enter freeze tomorrow, then push Rc2 on Friday and then if everything goes well we can push 5.5.0 final next Monday. Hopefully that plan works for everyone, Daniel -- Daniel Veillard | Red Hat Developers Tools http://developer.redhat.com/ veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] qemu: distinguish pr disk before qemuHotplugRemoveManagedPR
by Jie Wang 25 Jun '19

25 Jun '19

when a disk without PR perform attach or detach operation, need not call qemuHotplugRemoveManagedPR, otherwise, it will print err log about PR, let us fix it. Signed-off-by: Jie Wang <wangjie88(a)huawei.com> --- src/qemu/qemu_hotplug.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index efda539..7ef92d0 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -928,7 +928,9 @@ qemuDomainAttachDiskGeneric(virQEMUDriverPtr driver, if (qemuDomainObjExitMonitor(driver, vm) < 0) ret = -2; - if (qemuHotplugRemoveManagedPR(driver, vm, QEMU_ASYNC_JOB_NONE) < 0) + + if (virStorageSourceChainHasManagedPR(disk->src) && + qemuHotplugRemoveManagedPR(driver, vm, QEMU_ASYNC_JOB_NONE) < 0) ret = -2; virDomainAuditDisk(vm, NULL, disk->src, "attach", false); @@ -4481,7 +4483,8 @@ qemuDomainRemoveDiskDevice(virQEMUDriverPtr driver, dev.data.disk = disk; ignore_value(qemuRemoveSharedDevice(driver, &dev, vm->def->name)); - if (qemuHotplugRemoveManagedPR(driver, vm, QEMU_ASYNC_JOB_NONE) < 0) + if (virStorageSourceChainHasManagedPR(disk->src) && + qemuHotplugRemoveManagedPR(driver, vm, QEMU_ASYNC_JOB_NONE) < 0) goto cleanup; ret = 0; -- 1.8.3.1

3 3

[libvirt] [PATCH] storage: omit comma for ceph mon hosts to librados
by Yi Li 25 Jun '19

25 Jun '19

Silly mistakes omit the comma for multiple ipaddr Fixes: cdd362e0e7a34d4f8f102c75f2ca513d23dd1db0 Signed-off-by: Yi Li <yili(a)winhong.com> --- src/storage/storage_backend_rbd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c index bd6e3d5..315bef2 100644 --- a/src/storage/storage_backend_rbd.c +++ b/src/storage/storage_backend_rbd.c @@ -272,10 +272,10 @@ virStorageBackendRBDOpenRADOSConn(virStorageBackendRBDStatePtr ptr, const char *incFormat; if (virSocketAddrNumericFamily(source->hosts[i].name) == AF_INET6) { /* IPv6 address must be escaped in brackets on the cmd line */ - incFormat = "[%s]:%d"; + incFormat = "[%s]:%d,"; } else { /* listenAddress is a hostname or IPv4 */ - incFormat = "%s:%d"; + incFormat = "%s:%d,"; } virBufferAsprintf(&mon_host, incFormat, source->hosts[i].name, -- 2.7.5

3 2

[libvirt] [PATCH] deprecate -mem-path fallback to anonymous RAM
by Igor Mammedov 25 Jun '19

25 Jun '19

Fallback might affect guest or worse whole host performance or functionality if backing file were used to share guest RAM with another process. Patch deprecates fallback so that we could remove it in future and ensure that QEMU will provide expected behavior and fail if it can't use user provided backing file. Signed-off-by: Igor Mammedov <imammedo(a)redhat.com> --- PS: Patch is written on top of [PATCH v4 0/3] numa: deprecate '-numa node, mem' and default memory distribution to avoid conflicts in qemu-deprecated.texi numa.c | 4 ++-- qemu-deprecated.texi | 8 ++++++++ 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/numa.c b/numa.c index 91a29138a2..53d67b8ad9 100644 --- a/numa.c +++ b/numa.c @@ -494,8 +494,8 @@ static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner, if (mem_prealloc) { exit(1); } - error_report("falling back to regular RAM allocation."); - + warn_report("falling back to regular RAM allocation. " + "Fallback to RAM allocation is deprecated."); /* Legacy behavior: if allocation failed, fall back to * regular RAM allocation. */ diff --git a/qemu-deprecated.texi b/qemu-deprecated.texi index 2fe9b72121..2193705644 100644 --- a/qemu-deprecated.texi +++ b/qemu-deprecated.texi @@ -112,6 +112,14 @@ QEMU using implicit generic or board specific splitting rule. Use @option{memdev} with @var{memory-backend-ram} backend or @option{mem} (if it's supported by used machine type) to define mapping explictly instead. +@subsection -mem-path fallback to RAM (since 4.1) +Currently if system memory allocation from file pointed by @option{mem-path} +fails, QEMU fallbacks to allocating from anonymous RAM. Which might result +in unpredictable behavior since provided backing file wasn't used. In future +QEMU will not fallback and fail to start up, so user could fix his/her QEMU/host +configuration or explicitly use -m without -mem-path if system memory allocated +from anonymous RAM suits usecase. + @section QEMU Machine Protocol (QMP) commands @subsection block-dirty-bitmap-add "autoload" parameter (since 2.12.0) -- 2.18.1

4 8

[libvirt] [PATCH v2 0/6] cgroups v2 fixes and improvements
by Pavel Hrdina 25 Jun '19

25 Jun '19

Pavel Hrdina (6): util: vircgroup: pass parent cgroup into virCgroupDetectControllersCB util: vircgroup: improve controller detection util: vircgroupv2: use any controller to create thread directory util: vircgroupv2: enable CPU controller only if it's available util: vircgroupv2: separate return values of virCgroupV2EnableController util: vircgroupv2: don't error out if enabling controller fails src/util/vircgroup.c | 32 ++++++++++---------- src/util/vircgroupbackend.h | 3 +- src/util/vircgroupv1.c | 3 +- src/util/vircgroupv2.c | 59 +++++++++++++++++++++++++++++-------- 4 files changed, 66 insertions(+), 31 deletions(-) -- 2.21.0

1 7

[libvirt] [PATCH 0/3] conf: Fix parsing of <disk> <seclabel> (blockdev-add saga)
by Peter Krempa 25 Jun '19

25 Jun '19

The parser for the seclabel would mess up context leaving disk source private data unparsed. Peter Krempa (3): tests: qemustatusxml2xml: Split out the output file for 'modern' test tests: qemustatusxml2xml: Add a seclabel for a disk with privateData conf: domain: Restore XPath context after virSecurityDeviceLabelDefParseXML src/conf/domain_conf.c | 1 + tests/qemustatusxml2xmldata/modern-in.xml | 3 + tests/qemustatusxml2xmldata/modern-out.xml | 474 ++++++++++++++++++++- 3 files changed, 477 insertions(+), 1 deletion(-) mode change 120000 => 100644 tests/qemustatusxml2xmldata/modern-out.xml -- 2.21.0

2 8