- Devel - libvirt List Archives

[libvirt] on shutdown
by Ruben Kerkhof 26 Oct '16

26 Oct '16

Hi all, virsh(1) has this to say about virsh shutdown: "The exact behavior of a domain when it shuts down is set by the on_shutdown parameter in the domain's XML definition." on_shutdown isn't documented in https://libvirt.org/formatdomain.html Is the virsh(1) manpage wrong? Kind regards, Ruben Kerkhof

2 4

[libvirt] [PATCH] Resuming running domain raise error message now
by Sławek Kapłoński 26 Oct '16

26 Oct '16

This patch introduce raising XML error when user tries to resume domain (Qemu or LXC) which is already running. Closes bug: https://bugzilla.redhat.com/show_bug.cgi?id=1009008 Sławek Kapłoński (1): Resuming running domain raise error message src/lxc/lxc_driver.c | 8 +++++++- src/qemu/qemu_driver.c | 4 ++++ 2 files changed, 11 insertions(+), 1 deletion(-) -- 2.10.1

2 5

[libvirt] [PATCH 0/5] wireshark: Fix $(prefix) handling
by Andrea Bolognani 26 Oct '16

26 Oct '16

Well, almost :) There are still some cases that are not handled correctly, but at least this will unbreak 'make rpm' while I work on the rest. Tested by running 'make rpm' successfully on Fedora 23, Fedora 24 and Fedora rawhide. Andrea Bolognani (5): wireshark: Introduce $ws_modversion wireshark: Hoist $ws_prefix declaration wireshark: Strip prefix correctly wireshark: Inject $(prefix) at the right time wireshark: Rename plugindir to ws_plugindir m4/virt-wireshark.m4 | 29 ++++++++++++++++++----------- tools/Makefile.am | 2 +- 2 files changed, 19 insertions(+), 12 deletions(-) -- 2.7.4

3 8

[libvirt] [PATCH] virt-wireshark.m4: Defer $(prefix) substitution
by Michal Privoznik 26 Oct '16

26 Oct '16

The problem with evaluating $(prefix) in configure phase is that autoconf does a lot of magic with this variable. Firstly, if the --prefix argument is not set for the configure script, the variable will have value "NONE". Wait, wat? Secondly, even autoconf scripts then have special cases for when variable is "NONE". It's even worse - whenever they need to construct new path based on prefix, they save its current value, substitute it with proper value (like /usr/local/), do what they need and then restore the original value. Le sigh. This mad behaviour is, however, suppressed if there's --prefix on the configure command line. Now, the problem is that we tried to construct path for wireshark plugin in the m4 file. Based on the knowledge described above, we shouldn't do it and just construct the full path in Makefile later where prefix has proper value. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- m4/virt-wireshark.m4 | 2 +- tools/Makefile.am | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/m4/virt-wireshark.m4 b/m4/virt-wireshark.m4 index f383e2b..e1e4a59 100644 --- a/m4/virt-wireshark.m4 +++ b/m4/virt-wireshark.m4 @@ -38,7 +38,7 @@ AC_DEFUN([LIBVIRT_CHECK_WIRESHARK],[ if test "x$ws_prefix" = "x" ; then ws_prefix="/usr"; fi - plugindir="${prefix}${plugindir#$ws_prefix}" + plugindir="${plugindir#$ws_prefix}" fi elif test "x$with_ws_plugindir" = "xno" || test "x$with_ws_plugindir" = "xyes"; then AC_MSG_ERROR([ws-plugindir must be used only with valid path]) diff --git a/tools/Makefile.am b/tools/Makefile.am index e7e42c3..08e1680 100644 --- a/tools/Makefile.am +++ b/tools/Makefile.am @@ -398,7 +398,7 @@ EXTRA_DIST += \ if WITH_WIRESHARK_DISSECTOR -ws_plugindir = $(plugindir) +ws_plugindir = $(prefix)$(plugindir) ws_plugin_LTLIBRARIES = wireshark/src/libvirt.la wireshark_src_libvirt_la_CPPFLAGS = \ -I wireshark/src $(WIRESHARK_DISSECTOR_CFLAGS) -- 2.8.4

4 6

[libvirt] [PATCH v11 0/5] Add native TLS encrypted chardev TCP support
by John Ferlan 26 Oct '16

26 Oct '16

v10: http://www.redhat.com/archives/libvir-list/2016-October/msg00866.html Changes since v10... This is essentially the adding the secret to/for the TLS chardev TCP. Lots of messy stuff. Patches 1-3 are new... In particular, I kept looking at the RNG removal code and kept thinking, something just doesn't look right when compared to the RemoveVirtioDisk code. If I'm incorrect with order of removal it's an easy enough change. Kept trying to think what depends on what and for the RNG it would seem a need a to unplug the backend before removing the front end object would be more logical. Patches 4-5 are the old patch 3-4. They (hopefully) encompass the previous review plus other things since discovered. John Ferlan (5): qemu: Move TLS object remove from DetachChr to RemoveChr qemu: Swap order of RNG hot unplug removal qemu: Need to remove TLS object in RemoveRNGDevice qemu: Add a secret object to/for a char source dev qemu: Add secret object hotplug for TCP chardev TLS src/qemu/qemu_command.c | 30 +++- src/qemu/qemu_command.h | 1 + src/qemu/qemu_domain.c | 173 +++++++++++++++++- src/qemu/qemu_domain.h | 17 +- src/qemu/qemu_driver.c | 6 +- src/qemu/qemu_hotplug.c | 195 +++++++++++++++++---- src/qemu/qemu_hotplug.h | 9 +- src/qemu/qemu_process.c | 9 +- tests/qemuhotplugtest.c | 2 +- ...xml2argv-serial-tcp-tlsx509-secret-chardev.args | 38 ++++ ...uxml2argv-serial-tcp-tlsx509-secret-chardev.xml | 50 ++++++ tests/qemuxml2argvtest.c | 17 ++ 12 files changed, 495 insertions(+), 52 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.xml -- 2.7.4

2 16

[libvirt] [PATCH v5 00/15] Use more PCIe less legacy PCI
by Laine Stump 26 Oct '16

26 Oct '16

These are just the rebased versions of the patches from v4 that I haven't yet pushed, to make review easier. Patch 1 and 2 (and several others) are already ACKed, but included for completeness. Laine Stump (15): qemu: new functions qemuDomainMachineHasPCI[e]Root() qemu: new functions to calculate/set device pciConnectFlags qemu: set/use info->pciConnectFlags when validating/assigning PCI addresses qemu: set/use proper pciConnectFlags during hotplug qemu: set pciConnectFlags to 0 instead of PCI|HOTPLUGGABLE if device isn't PCI qemu: assign virtio devices to PCIe slot when appropriate qemu: assign e1000e network devices to PCIe slots when appropriate qemu: assign nec-xhci (USB3) controller to a PCIe address when appropriate qemu: only force an available legacy-PCI slot on domains with pci-root qemu: auto-add pcie-root-port/dmi-to-pci-bridge controllers as needed [RFC] qemu: if pci-bridge is in PCIe config w/o dmi-to-pci-bridge, add it qemu: don't force-add a dmi-to-pci-bridge just on principle qemu: add a USB3 controller to Q35 domains by default qemu: try to put ich9 sound device at 00:1B.0 qemu: initially reserve one open pcie-root-port for hotplug src/conf/device_conf.h | 5 + src/conf/domain_addr.c | 112 ++- src/conf/domain_addr.h | 3 +- src/qemu/qemu_domain.c | 52 +- src/qemu/qemu_domain.h | 2 + src/qemu/qemu_domain_address.c | 805 +++++++++++++++++---- src/qemu/qemu_domain_address.h | 4 + src/qemu/qemu_hotplug.c | 20 +- tests/qemuxml2argvdata/qemuxml2argv-autoindex.args | 10 +- .../qemuxml2argv-bios-nvram-secure.args | 1 + .../qemuxml2argv-machine-smm-opt.args | 1 + tests/qemuxml2argvdata/qemuxml2argv-pcie-root.args | 3 +- .../qemuxml2argv-q35-default-devices-only.args | 23 + .../qemuxml2argv-q35-default-devices-only.xml | 18 + .../qemuxml2argv-q35-pcie-autoadd.args | 57 ++ .../qemuxml2argv-q35-pcie-autoadd.xml | 51 ++ tests/qemuxml2argvdata/qemuxml2argv-q35-pcie.args | 58 ++ tests/qemuxml2argvdata/qemuxml2argv-q35-pcie.xml | 67 ++ .../qemuxml2argv-q35-pm-disable-fallback.args | 1 + .../qemuxml2argv-q35-pm-disable.args | 1 + .../qemuxml2argv-q35-virt-manager-basic.args | 56 ++ .../qemuxml2argv-q35-virt-manager-basic.xml | 76 ++ .../qemuxml2argv-q35-virtio-pci.args | 58 ++ .../qemuxml2argv-q35-virtio-pci.xml | 1 + tests/qemuxml2argvtest.c | 164 ++++- .../qemuxml2xmlout-autoindex.xml | 10 +- .../qemuxml2xmlout-pcie-root.xml | 4 - .../qemuxml2xmlout-q35-default-devices-only.xml | 45 ++ .../qemuxml2xmlout-q35-pcie-autoadd.xml | 148 ++++ .../qemuxml2xmloutdata/qemuxml2xmlout-q35-pcie.xml | 152 ++++ .../qemuxml2xmlout-q35-virt-manager-basic.xml | 121 ++++ .../qemuxml2xmlout-q35-virtio-pci.xml | 152 ++++ tests/qemuxml2xmltest.c | 136 +++- 33 files changed, 2189 insertions(+), 228 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-default-devices-only.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-default-devices-only.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-pcie-autoadd.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-pcie-autoadd.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-pcie.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-pcie.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-virt-manager-basic.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-virt-manager-basic.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-q35-virtio-pci.args create mode 120000 tests/qemuxml2argvdata/qemuxml2argv-q35-virtio-pci.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-q35-default-devices-only.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-q35-pcie-autoadd.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-q35-pcie.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-q35-virt-manager-basic.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-q35-virtio-pci.xml -- 2.7.4

2 18

[libvirt] [PATCH] qemu: Fix depedency order in qemuRemoveDiskDevice
by John Ferlan 26 Oct '16

26 Oct '16

Need to remove the drive first, then the secobj and/or encobj if they exist. This is because the drive has a dependency on secobj (or the secret for the networked storage server) and/or the encobj (or the secret for the LUKS encrypted volume). Deleting either object first leaves an drive without it's respective objects. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/qemu/qemu_hotplug.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 7a21dc6..30565d5 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3176,6 +3176,9 @@ qemuDomainRemoveDiskDevice(virQEMUDriverPtr driver, qemuDomainObjEnterMonitor(driver, vm); + qemuMonitorDriveDel(priv->mon, drivestr); + VIR_FREE(drivestr); + /* If it fails, then so be it - it was a best shot */ if (objAlias) ignore_value(qemuMonitorDelObject(priv->mon, objAlias)); @@ -3186,8 +3189,6 @@ qemuDomainRemoveDiskDevice(virQEMUDriverPtr driver, ignore_value(qemuMonitorDelObject(priv->mon, encAlias)); VIR_FREE(encAlias); - qemuMonitorDriveDel(priv->mon, drivestr); - VIR_FREE(drivestr); if (qemuDomainObjExitMonitor(driver, vm) < 0) return -1; -- 2.7.4

2 1

[libvirt] [PATCH v2 00/12] Add length (duration) params for iotune throttling
by John Ferlan 25 Oct '16

25 Oct '16

v1: http://www.redhat.com/archives/libvir-list/2016-September/msg01090.html Differences since v1: Patches 1-5 are new... Patch 1 is essentially the former patch 8.5 added after initial review based on a review comment. Erik Skultety and I went back and forth on this one a few times and this is the result Patch 2 is a result of being frustrated by a generic error message when I know we could return whatever qemu gives us Patch 3 is simple code motion - the conf_disk is closer to usage now Patch 4 & 5 create and use the same algorithm to set the default values that weren't provided. Patches 6-7 have already been ACK'd... I've been waiting to push because they will interfere with libvirt-perl builds and since the 2.3.0 hasn't been published yet, I'm still holding onto them. Patch 8 is the former patch 8, but reworked mostly in qemu_driver.c, but also a change in qemu_monitor_json.c to use "P:" instead of "U:" for then length values. For _length a 0 is an invalid value, so we'll use it to mean "nothing changed" Patch 9 is the former patch 11 and is mostly intact, with major difference being the removal of the CHECK_IOTUNE_MAX_LENGTH macro (it was causing domain disappearance) Patch 10 is the former patch 12 and is unchanged. It also has been ACKd Patch 11-12 are new - it's the virsh code. John Ferlan (12): qemu: Create a macro to handle setting bytes/iops iotune values qemu: Return real error message for block_set_io_throttle qemu: Move setting of conf_disk in qemuDomainSetBlockIoTune qemu: Introduce qemuDomainSetBlockIoTuneSetDefaults qemu: Use qemuDomainsetBlockIoTuneSetDefaults for config include: Add new definitions for duration for bps/iops throttling caps: Add new capability for the bps/iops throttling length qemu: Add length for bps/iops throttling parameters to driver conf: Add support for blkiotune "_length" options qemu: Add the length options to the iotune command line virsh: Create a macro to add IOTUNE values virsh: Add _length parameters to virsh output docs/formatdomain.html.in | 40 ++- docs/schemas/domaincommon.rng | 38 +++ include/libvirt/libvirt-domain.h | 102 ++++++ src/conf/domain_conf.c | 24 +- src/conf/domain_conf.h | 6 + src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 21 ++ src/qemu/qemu_driver.c | 341 +++++++++++---------- src/qemu/qemu_monitor.c | 7 +- src/qemu/qemu_monitor.h | 3 +- src/qemu/qemu_monitor_json.c | 39 ++- src/qemu/qemu_monitor_json.h | 3 +- .../caps_2.6.0-gicv2.aarch64.xml | 1 + .../caps_2.6.0-gicv3.aarch64.xml | 1 + tests/qemucapabilitiesdata/caps_2.6.0.ppc64le.xml | 1 + tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.7.0.x86_64.xml | 1 + tests/qemumonitorjsontest.c | 17 +- .../qemuxml2argv-blkdeviotune-max-length.args | 34 ++ .../qemuxml2argv-blkdeviotune-max-length.xml | 65 ++++ tests/qemuxml2argvtest.c | 4 + .../qemuxml2xmlout-blkdeviotune-max-length.xml | 1 + tests/qemuxml2xmltest.c | 1 + tools/virsh-domain.c | 196 +++++------- tools/virsh.pod | 21 ++ 26 files changed, 673 insertions(+), 298 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-blkdeviotune-max-length.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-blkdeviotune-max-length.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-blkdeviotune-max-length.xml -- 2.7.4

2 29

Re: [libvirt] question about rdma migration
by Michael R. Hines 25 Oct '16

25 Oct '16

Hi Roy, On 02/09/2016 03:57 AM, Roy Shterman wrote: > Hi, > > I tried to understand the rdma-migration in qemu code and i have two > questions about it: > > 1. I'm working with qemu-kvm using libvirt and i'm getting > > MEMLOCK max locked-in-memory address space 65536 65536 bytes > > in qemu process so I don't understand how can you use rdma-pin-all > with such low MEMLOCK. > > I found a solution in libvirt to lock all vm memory in advance and to > enlarge MEMLOCK. > It uses memoryBacking locking and memory tuning hard_limit of vm > memory but I couldn't find a usage of this in rdma-migration code. > You're absolutey right, the RDMA migration code itself doesn't set this lock limit explicitly because there are system-wide restrictions in both appArmour, /etc/security, as well as SELINUX that restrict applications from arbitrarily setting their maximum memory lock limits. The other problem is CGROUPS: If someone sets a cgroup control for maximum memory and forgets about that mlock() limits, then there will be a conflict. So, libvirt must have a policy to deal with all of these possibilities, not just handle a special case for RDMA migration. The only way "simple" way (without patching the problems above) to apply a higher lock limit to QEMU is to set the ulimit for libvirt (or for QEMU if starting QEMU manually) in your environment or the command line with $ ulimit # before attempting the migration, then the RDMA subsystem will be able to lock the memory successfully. The other option is to use /etc/security/limits.conf and set the option for a specific libvirt process user and make sure your libvirt/qemu are not running as root. QEMU itself also has a "mlock" option built into the command line, but it also suffers from the same problem --- you have to find a way (currently) to increase the limit before using the option. > 2. Do you have any comparison of IOPS and bandwidth between TCP > migration and rdma migration? > Yes, lots of comparisons. http://wiki.qemu.org/Features/RDMALiveMigration http://www.canturkisci.com/ETC/papers/IBMJRD2011/preprint.pdf > Regards, > Roy > >

5 15

[libvirt] [PATCH] qemu_driver: unlink new domain cfg file when rollbak
by Chen Hanxiao 25 Oct '16

25 Oct '16

From: Chen Hanxiao <chenhanxiao(a)gmail.com> If we failed to unlink old dom cfg file, we goto rollback. But inside rollback, we fogot to unlink the new dom cfg file. This patch fixes this issue. Signed-off-by: Chen Hanxiao <chenhanxiao(a)gmail.com> --- src/qemu/qemu_driver.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index e6f845d..3f4a2fb 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -19869,6 +19869,7 @@ qemuDomainRenameCallback(virDomainObjPtr vm, char *new_dom_name = NULL; char *old_dom_name = NULL; char *old_dom_cfg_file = NULL; + char *new_dom_cfg_file = NULL; virCheckFlags(0, ret); @@ -19882,6 +19883,11 @@ qemuDomainRenameCallback(virDomainObjPtr vm, goto cleanup; } + if (!(new_dom_cfg_file = virDomainConfigFile(cfg->configDir, + new_dom_name))) { + goto cleanup; + } + event_old = virDomainEventLifecycleNewFromObj(vm, VIR_DOMAIN_EVENT_UNDEFINED, VIR_DOMAIN_EVENT_UNDEFINED_RENAMED); @@ -19909,6 +19915,7 @@ qemuDomainRenameCallback(virDomainObjPtr vm, cleanup: VIR_FREE(old_dom_cfg_file); + VIR_FREE(new_dom_cfg_file); VIR_FREE(old_dom_name); VIR_FREE(new_dom_name); qemuDomainEventQueue(driver, event_old); @@ -19922,6 +19929,10 @@ qemuDomainRenameCallback(virDomainObjPtr vm, vm->def->name = old_dom_name; old_dom_name = NULL; } + + if (virFileExists(new_dom_cfg_file)) + unlink(new_dom_cfg_file); + goto cleanup; } -- 1.8.3.1

1 2