May 2016 - Devel - libvirt List Archives

[libvirt] [PATCH v2 0/3] Improve init script choice
by Andrea Bolognani 02 May '16

02 May '16

Changes from [v1]: * look for systemctl(1) to detect systemd instead of relying on whether or not the /etc/systemd directory is present, as suggested by Martin * use 'systemd+redhat' instead of plain 'systemd' on RHEL * move everything to a macro tucked away in a separate file Tested on Debian sid, Fedora rawhide, FreeBSD 10.3 and CentOS 6. [v1] https://www.redhat.com/archives/libvir-list/2016-April/msg01930.html Andrea Bolognani (3): configure: Improve --with-init-script=check configure: Add systemd detection to --with-init-script=check configure: Introduce LIBVIRT_{CHECK,RESULT}_INIT_SCRIPT configure.ac | 48 ++--------------------------- m4/virt-init-script.m4 | 84 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 86 insertions(+), 46 deletions(-) create mode 100644 m4/virt-init-script.m4 -- 2.5.5

2 7

[libvirt] [PATCH v3 0/6] Add panic device support for S390
by Boris Fiuczynski 02 May '16

02 May '16

S390 architecture inherently provides a crash detection capability which has not been reflected in the domain xml. This series adds an s390 model to the panic device which does not allow an address to be specified and is by default created on S390 guests unless already provided. v3: * Merged S390 and S390X in the add default device function * Changed and added tests * Split out ISA panic device message into separate patch * Adjusted documentation v2: * Grouped default panic device creation and required test code changes into one patch. * Added an s390 model instead of using default model for toleration. Boris Fiuczynski (6): qemu: add panic device support for S390 qemu: add default panic device to S390 guests tests: add tests for panic device model s390 docs: align spelling of S390 qemu: merge S390 and S390X default device creation qemu: fix error message for default panic device docs/formatdomain.html.in | 26 ++++++++++++------- docs/schemas/domaincommon.rng | 1 + src/conf/domain_conf.c | 3 ++- src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 21 +++++++++++++++- src/qemu/qemu_domain.c | 11 +++++--- .../qemuargv2xml-machine-aeskeywrap-off-argv.xml | 1 + .../qemuargv2xml-machine-aeskeywrap-on-argv.xml | 1 + .../qemuargv2xml-machine-deakeywrap-off-argv.xml | 1 + .../qemuargv2xml-machine-deakeywrap-on-argv.xml | 1 + .../qemuargv2xml-machine-keywrap-none-argv.xml | 1 + .../qemuxml2argv-s390-panic-address.xml | 26 +++++++++++++++++++ .../qemuxml2argv-s390-panic-missing.args | 25 +++++++++++++++++++ .../qemuxml2argv-s390-panic-missing.xml | 23 +++++++++++++++++ .../qemuxml2argv-s390-panic-no-address.args | 25 +++++++++++++++++++ .../qemuxml2argv-s390-panic-no-address.xml | 22 ++++++++++++++++ tests/qemuxml2argvdata/qemuxml2argv-s390-panic.xml | 22 ++++++++++++++++ tests/qemuxml2argvtest.c | 11 +++++++- .../qemuxml2xmlout-iothreads-disk-virtio-ccw.xml | 1 + .../qemuxml2xmlout-s390-defaultconsole.xml | 1 + .../qemuxml2xmlout-s390-panic-missing.xml | 29 ++++++++++++++++++++++ .../qemuxml2xmlout-s390-panic-no-address.xml | 28 +++++++++++++++++++++ .../qemuxml2xmlout-s390-panic.xml | 28 +++++++++++++++++++++ tests/qemuxml2xmltest.c | 6 +++++ 24 files changed, 300 insertions(+), 15 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic-address.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic-missing.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic-missing.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic-no-address.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic-no-address.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-s390-panic.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-s390-panic-missing.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-s390-panic-no-address.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-s390-panic.xml -- 2.3.0

2 7

[libvirt] [PATCH 0/9] stream error reporting fixes
by Cole Robinson 02 May '16

02 May '16

This series attempts to address various stream finish/abort error reporting issues: * Unclear error messages * virStreamAbort always returns an error * Spurious errors logged on the daemon side * Various issues in stream error cleanup (found through code inspection) Cole Robinson (9): fdstream: separate out virCommandPtr cleanup fdstream: Report error with virProcessTranslateStatus fdstream: Raise explicit error when iohelper gets SIGPIPE daemon: stream: Close stream on send failure daemon: stream: don't update events if stream->closed daemon: stream: set stream->closed on removal rpc: protocol: Clarify VIR_NET_ERROR usage with streams daemon: stream: Don't force error when client aborts fdstream: don't raise error on SIGPIPE if abort requested daemon/stream.c | 42 +++++++++++++++++------- src/fdstream.c | 84 ++++++++++++++++++++++++++++++------------------ src/rpc/virnetprotocol.x | 7 ++-- 3 files changed, 88 insertions(+), 45 deletions(-) -- 2.7.3

2 13

[libvirt] [PATCH v2 00/12] Add IV Secret Object support
by John Ferlan 02 May '16

02 May '16

v1: http://www.redhat.com/archives/libvir-list/2016-April/msg00596.html Differences since v1: - Add qemuBuildiSCSICommandLine (and BuildDiskiSCSI && BuildHostdeviSCSI) These will do the magic necessary in order to support IV secret objects for the impending iSCSI -drive argument. This API doesn't require any qemu patches in order to work AFAICT. I also determined that the "id=" *isn't* required for an '-iscsi ...' argument, which made using the complete 'path' string for 'initiator-name' possible. The other option was to break it up and pass the "iqn.*" string as the initiator-name and a "modified" remaining string as the "id=" parameter. The modified would be to ensure only alphanumeric, '-', '.', and '_' characters are in the 'id=' string. - Fix up some logic found while actually working through the tests. Some of it related to what was found for the 'iscsi' options. A couple of other minor nits. - Add tests and mocks for virRandomBytes and gnutls_rnd (note: the former could be used to "randomly" (hah!) generate a UUID of all '0xff'). A mock of 'gnutls_encrypt' is not necessary since, it can only be called if the function gnutls_encrypt exists *and* we have a secret object capability. Not having a mock function allows us to validate that gnutls_encrypt actually generates a value we expect based on some less than stellar and totally non random key's! - Remove the hotplug IV code (I've saved it off for future expansion). Although not needing to do hotplug probably means patches 6-9 are not required, but still I think better than the existing so I kept them even though they have nothing to do with IV secrets (they'd need to go in after patches 1-5 anyways). - Ran the changes through the coverity checker... John Ferlan (12): qemu: Introduce qemuDomainSecretInfo qemu: Introduce qemuDomainSecretPrepare and Destroy qemu: Introduce qemuDomainHostdevPrivatePtr qemu: Introduce qemuDomainSecretHostdevPrepare and Destroy qemu: Use qemuDomainSecretInfoPtr in qemuBuildNetworkDriveURI qemu: hotplug: Assume support for -device for attach virtio disk qemu: hotplug: Adjust error path for attach scsi disk qemu: hotplug: Adjust error path for attach virtio disk qemu: hotplug: Adjust error path for attach hostdev scsi disk qemu: hotplug: Fix possible memory leak of props qemu: Introduce qemuDomainSecretIV qemu: Utilize qemu secret objects for SCSI/RBD auth/secret configure.ac | 1 + src/conf/domain_conf.c | 33 +- src/conf/domain_conf.h | 5 +- src/lxc/lxc_native.c | 4 +- src/qemu/qemu_alias.c | 23 + src/qemu/qemu_alias.h | 2 + src/qemu/qemu_command.c | 445 ++++++++++++++---- src/qemu/qemu_command.h | 13 +- src/qemu/qemu_domain.c | 516 ++++++++++++++++++++- src/qemu/qemu_domain.h | 81 +++- src/qemu/qemu_driver.c | 13 +- src/qemu/qemu_hotplug.c | 247 +++++----- src/qemu/qemu_hotplug.h | 4 +- src/qemu/qemu_parse_command.c | 4 +- src/qemu/qemu_process.c | 13 +- src/vbox/vbox_common.c | 4 +- src/xenconfig/xen_common.c | 4 +- src/xenconfig/xen_sxpr.c | 4 +- ...uxml2argv-disk-drive-network-iscsi-auth-IV.args | 39 ++ ...muxml2argv-disk-drive-network-iscsi-auth-IV.xml | 43 ++ ...emuxml2argv-disk-drive-network-rbd-auth-IV.args | 31 ++ ...qemuxml2argv-disk-drive-network-rbd-auth-IV.xml | 42 ++ ...emuxml2argv-hostdev-scsi-lsi-iscsi-auth-IV.args | 41 ++ ...qemuxml2argv-hostdev-scsi-lsi-iscsi-auth-IV.xml | 48 ++ ...xml2argv-hostdev-scsi-virtio-iscsi-auth-IV.args | 43 ++ ...uxml2argv-hostdev-scsi-virtio-iscsi-auth-IV.xml | 48 ++ tests/qemuxml2argvmock.c | 31 +- tests/qemuxml2argvtest.c | 19 + tests/virhostdevtest.c | 3 +- 29 files changed, 1557 insertions(+), 247 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-network-iscsi-auth-IV.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-network-iscsi-auth-IV.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-network-rbd-auth-IV.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-network-rbd-auth-IV.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-hostdev-scsi-lsi-iscsi-auth-IV.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-hostdev-scsi-lsi-iscsi-auth-IV.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-hostdev-scsi-virtio-iscsi-auth-IV.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-hostdev-scsi-virtio-iscsi-auth-IV.xml -- 2.5.5

4 36

[libvirt] [PATCH] Adapt augeas profile to handle negative int values.
by Cédric Bosdonnat 02 May '16

02 May '16

Introducing keepalive_interval = -1 breaks to augeas lens. Fix the lens by allowing signed ints in the regular expression. --- daemon/libvirtd.aug | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/daemon/libvirtd.aug b/daemon/libvirtd.aug index a70aa1d..4d40ee2 100644 --- a/daemon/libvirtd.aug +++ b/daemon/libvirtd.aug @@ -13,7 +13,7 @@ module Libvirtd = let str_val = del /\"/ "\"" . store /[^\"]*/ . del /\"/ "\"" let bool_val = store /0|1/ - let int_val = store /[0-9]+/ + let int_val = store /-?[0-9]+/ let str_array_element = [ seq "el" . str_val ] . del /[ \t\n]*/ "" let str_array_val = counter "el" . array_start . ( str_array_element . ( array_sep . str_array_element ) * ) ? . array_end -- 2.6.6

4 3

[libvirt] [PATCH v3 0/4] Fix parsing our own XMLs
by Martin Kletzander 02 May '16

02 May '16

v3: - Actually use virPCIDeviceAddress. It was actually ACKed with that changed but I feel like it's way too big of a change for just pushing it. Also, I'm in no rush with this, so I'l gladly wait after therelease with this. v2: - Just a rebase - I did *not* use virPCIDeviceAddress wording instead as discussed in the v1 thread. That's because we have lot of functions working with virDevicePCIAddress named exactly after that and renaming those would be ugly IMHO. - https://www.redhat.com/archives/libvir-list/2016-April/msg00714.html v1: - https://www.redhat.com/archives/libvir-list/2016-April/msg00081.html Martin Kletzander (4): Change virDevicePCIAddress to virPCIDeviceAddress Move capability formatting together schemas: Update nodedev schema to match reality conf: Parse more of our nodedev XML docs/schemas/nodedev.rng | 29 +++--- src/bhyve/bhyve_device.c | 4 +- src/conf/device_conf.c | 16 +-- src/conf/device_conf.h | 25 ++--- src/conf/domain_addr.c | 24 ++--- src/conf/domain_addr.h | 22 ++-- src/conf/domain_conf.c | 26 ++--- src/conf/domain_conf.h | 10 +- src/conf/network_conf.c | 8 +- src/conf/network_conf.h | 2 +- src/conf/node_device_conf.c | 111 +++++++++++++++++++-- src/conf/storage_conf.c | 14 +-- src/conf/storage_conf.h | 2 +- src/libvirt_private.syms | 8 +- src/network/bridge_driver.c | 4 +- src/qemu/qemu_agent.c | 2 +- src/qemu/qemu_domain_address.c | 12 +-- src/qemu/qemu_hotplug.c | 10 +- src/qemu/qemu_monitor.c | 12 +-- src/qemu/qemu_monitor.h | 14 +-- src/qemu/qemu_monitor_json.c | 12 +-- src/qemu/qemu_monitor_json.h | 12 +-- src/qemu/qemu_monitor_text.c | 14 +-- src/qemu/qemu_monitor_text.h | 12 +-- src/storage/storage_backend_scsi.c | 2 +- src/util/virpci.c | 2 +- src/util/virpci.h | 3 +- src/xen/xend_internal.c | 2 +- .../pci_0000_00_1c_0_header_type.xml | 2 +- tests/nodedevschemadata/pci_0000_02_10_7_sriov.xml | 23 +++++ .../pci_0000_02_10_7_sriov_pf_vfs_all.xml | 29 ++++++ ...i_0000_02_10_7_sriov_pf_vfs_all_header_type.xml | 30 ++++++ .../pci_0000_02_10_7_sriov_vfs.xml | 26 +++++ .../pci_0000_02_10_7_sriov_zero_vfs_max_count.xml | 21 ++++ tests/nodedevxml2xmltest.c | 7 ++ 35 files changed, 391 insertions(+), 161 deletions(-) create mode 100644 tests/nodedevschemadata/pci_0000_02_10_7_sriov.xml create mode 100644 tests/nodedevschemadata/pci_0000_02_10_7_sriov_pf_vfs_all.xml create mode 100644 tests/nodedevschemadata/pci_0000_02_10_7_sriov_pf_vfs_all_header_type.xml create mode 100644 tests/nodedevschemadata/pci_0000_02_10_7_sriov_vfs.xml create mode 100644 tests/nodedevschemadata/pci_0000_02_10_7_sriov_zero_vfs_max_count.xml -- 2.8.1

2 5

[libvirt] [PATCH] spec: add vz daemon related packages to libvirt.spec.in
by Maxim Nestratov 02 May '16

02 May '16

--- libvirt.spec.in | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 50 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 8036fa3..cddf447 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -54,6 +54,7 @@ %define with_uml 0%{!?_without_uml:%{server_drivers}} %define with_libxl 0%{!?_without_libxl:%{server_drivers}} %define with_vbox 0%{!?_without_vbox:%{server_drivers}} +%define with_vz 0%{!?_without_vz:%{server_drivers}} %define with_qemu_tcg %{with_qemu} @@ -423,6 +424,9 @@ Requires: libvirt-daemon-driver-vbox = %{version}-%{release} %if %{with_nwfilter} Requires: libvirt-daemon-driver-nwfilter = %{version}-%{release} %endif + %if %{with_vz} +Requires: libvirt-daemon-driver-vz = %{version}-%{release} + %endif %if %{with_interface} Requires: libvirt-daemon-driver-interface = %{version}-%{release} @@ -988,6 +992,19 @@ VirtualBox %endif + %if %{with_vz} +%package daemon-driver-vz +Summary: Virtuozzo driver plugin for the libvirtd daemon +Group: Development/Libraries +Requires: libvirt-daemon = %{version}-%{release} + +%description daemon-driver-vz +The vz driver plugin for the libvirtd daemon, providing +an implementation of the hypervisor driver APIs using +Virtuozzo + %endif + + %if %{with_libxl} %package daemon-driver-libxl Summary: Libxl driver plugin for the libvirtd daemon @@ -1141,6 +1158,27 @@ Requires: libvirt-daemon-driver-storage = %{version}-%{release} Server side daemon and driver required to manage the virtualization capabilities of VirtualBox %endif + + %if %{with_vz} +%package daemon-vz +Summary: Server side daemon & driver required to run Virtuozzo guests +Group: Development/Libraries + +Requires: libvirt-daemon = %{version}-%{release} + %if %{with_driver_modules} +Requires: libvirt-daemon-driver-vz = %{version}-%{release} +Requires: libvirt-daemon-driver-interface = %{version}-%{release} +Requires: libvirt-daemon-driver-network = %{version}-%{release} +Requires: libvirt-daemon-driver-nodedev = %{version}-%{release} +Requires: libvirt-daemon-driver-nwfilter = %{version}-%{release} +Requires: libvirt-daemon-driver-secret = %{version}-%{release} +Requires: libvirt-daemon-driver-storage = %{version}-%{release} + %endif +%description daemon-vz +Server side daemon and driver required to manage the virtualization +capabilities of Virtuozzo + %endif + %endif # %{with_libvirtd} %package client @@ -2228,6 +2266,13 @@ exit 0 %defattr(-, root, root) %{_libdir}/%{name}/connection-driver/libvirt_driver_vbox.so %endif + + %if %{with_vz} +%files daemon-driver-vz +%defattr(-, root, root) +%{_libdir}/%{name}/connection-driver/libvirt_driver_vz.so + %endif + %endif # %{with_driver_modules} %if %{with_qemu_tcg} @@ -2259,6 +2304,11 @@ exit 0 %files daemon-vbox %defattr(-, root, root) %endif + + %if %{with_vz} +%files daemon-vz +%defattr(-, root, root) + %endif %endif # %{with_libvirtd} %if %{with_sanlock} -- 2.4.3

2 1

[libvirt] fix some problems with configure in libvirt-sandbox-service
by Daniel Brooks 01 May '16

01 May '16

I had some trouble building it and ended up with these patches; they basically just fix up the configure checks so that they find things which would later cause the build to fail.

1 2