IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
It's generally not as valuable for non-x86 where we don't have the
history of supporting non-virtio OSes, but as time goes on it will
likely become more relevant for non-x86 arches, so let's make this
change now to get ahead of it.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
If the user selects virtiofs when editting or adding a new VM, and
we don't detect that they have shared memory enabled, show
a warning label in the UI pointing them to the Memory screen.
It would be nicer if we did this for them, but to get that totally
correct would require both duplicating libvirt's shared memory
detection logic, and some surgery to the addhw wizard. This is good
enough for now
Signed-off-by: Cole Robinson <crobinso@redhat.com>
- Move most xml suboption testing to many-devices test
- Clarify every specific bit we are testing in the singleton tests
- Consolidate/drop/reduce a lot of tests
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Split out tpmdetails.py, following the pattern of fsdetails.py. This
adds more UI editing fields for an already attached TPM.
Move the model and version under an 'Advanced options' expander,
since we should be getting this correct by default.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The code previously was just encoding the same defaults as libvirt,
which doesn't really add anything.
Instead, let's prefer type='emulator' model='tpm-crb', which
gives the most modern virtualization friendly config. When we don't
know if that will work, we mostly leave things up to libvirt to fill
in.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Add extra PCIe root ports to enable q35 device hotplug to work out
of the box. A typical modern linux guest has 7-8 PCI devices added
by default, so this gives plenty of wiggle room.
The smart thing to do would be to count the attached PCI devices
and add 4-5 extra, but that takes more work and isn't trivial.
The number can be overridden on the cli with:
--controller q35_pcie_root_ports=X
Use =0 to go back to the old behavior.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Adjust the UI to leave the box checked for both host-model and
host-passthrough, but host-passthrough is now what it means when
the user selects it. host-model can still be selected via the
CPU model drop down list
Signed-off-by: Cole Robinson <crobinso@redhat.com>
When libvirt and qemu are new enough, use host-passthrough for the
CPU default. Nowadays this is recommended over host-model for most
end user usage where migration isn't a critical feature.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
These have been used in the past, but no current code touches them,
so drop the parsing infrastructure
Signed-off-by: Cole Robinson <crobinso@redhat.com>
host-copy was the old default, but it's fundamentally flawed. Since
we switched to host-model default a few years back, it's not advertised
in the docs or selectable via virt-manager any more.
Have it print a warning and invoke host-model-only
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This was previously discussed here:
https://listman.redhat.com/archives/virt-tools-list/2020-September/msg00017.html
For the x86 + hvm case, failure to specify an --osinfo/--os-variant
OS, and failure to detect an OS from install media, will now throw
a big error:
```
--os-variant/--osinfo OS name is required, but no value was
set or detected.
This is now a fatal error. Specifying an OS name is required
for modern, performant, and secure virtual machine defaults.
If you expected virt-install to detect an OS name from the
install media, you can set a fallback OS name with:
--osinfo detect=on,name=OSNAME
You can see a full list of possible OS name values with:
virt-install --osinfo list
If your Linux distro is not listed, try one of generic values
such as: linux2020, linux2018, linux2016
If you just need to get the old behavior back, you can use:
--osinfo detect=on,require=off
Or export VIRTINSTALL_OSINFO_DISABLE_REQUIRE=1
```
The thread goes into more detail, but basically, for x86 VMs at least,
it's unlikely you will _ever_ want the default 'generic' behavior,
which gives gives no virtio, no PCIe, no usb3, IDE disks, slow
network devices, etc.
Many people use virt-install in scripts and CI, and this may now
cause breakage. The environment variable is there to help them
get things back to normal as quick as possible, but it will still
noisy up their logs with the warning to hopefully get them to make
a useful change to their virt-install invocations.
This is limited to x86, since that's where most of our defaults
historically differ, and where we can depend on libosinfo to give
the most accurate device info. This may be relevant to change for
other KVM architectures in the future.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Historically over time it's been more likely that an issue is reported
for osinfo-db missing the usb-tablet annotation. So for example we
always enable it for the default 'generic' case. We also want to
enable it for osinfo's 'unknown' case too.
Rather than add another check for that, let's just drop the osinfo
checking entirely. Some very old OS don't support usb-tablet, but
specifying it for those cases doesn't cause issues AFAIK, and users
can override it with `--input none` if they want.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
We are about to change the some defaults around os handling. Let's
start recommending the nicer named --osinfo more, since new error
messages are going to promote it a bit as well
Signed-off-by: Cole Robinson <crobinso@redhat.com>
I'm still seeing blog posts that recommend using
--os-type linux --os-variant XXX
Which has been a no op for a long time but is mostly harmless.
Current git would make this an error condition, but that's too
disruptive IMO. Just print a warning
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The canonical tool for this is `osinfo-query os`, which we still
reference in the man pages and in the list output.
However, we are about to make missing --os-variant fatal for common
usage, and I don't want to force users to install an extra tool just
to figure out what an acceptable --os-variant value is.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
- Move tooltip to the tree row instead of the finish button
- Some style cleanups
- Add a hack so we can hit it in the test suite
Signed-off-by: Cole Robinson <crobinso@redhat.com>
After checking with qemu devs, this option is not really recommended
for common usage and doesn't get used much in practice. So I don't
think it is suitable for the UI
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This recommendation came from an internal discussion. The cases are
* For block storage. This means guest requests are passed through
to the host device, which seems a more reasonable default than
ignoring them
* For sparse disk images we will create. discard=unmap helps preserve
the sparseness of the disk image. If a user requests non-sparse, they
are likely more concerned with performance than saving disk space,
so we leave the default as is. We limit this to disk images we will
create, since that's the easiest case to check, and it's less clear
if we should change the behavior here for an arbitrary existing
disk image.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This is from Gerd's suggestions here:
https://www.kraxel.org/blog/2019/09/display-devices-in-qemu/
When the guest supports it, we should use virtio. qxl is on the way
out, and the benefits are marginal and add a security and maintenance
burden.
While here, check domcaps that qxl or virtio are actually available.
Modern qemu has device modules, so device support may not be installed.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* Make it clear in code and UI that this is x86 only. Other arches
either require UEFI (aarch64) or don't support it
* Drop the internal 'bios' values since we don't handle them and may
not want them anyways, since when win11 support lands we will need
to explicitly throw an error if the user tries to force bios
* Add UI tests
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Let users choose libvirt's os.firmware=efi setting in the UI, putting
it about the firmware path list, since it's the preferred default
these days.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
<os firmware='efi'> is the libvirt official way to do what we
historically implement with `--boot uefi`, and UEFI setup in
virt-manager.
Let's prefer libvirt's official method if the support is advertised
in domcapabilities.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
I removed Portgroup UI in 4c3c53f773 release 3.0.0, but there's been
a steady stream of requests to bring it back. It seems it's commonly
used with some certain openvswitch config.
Maint burden isn't too bad. Let's bring it back
Fixes: https://github.com/virt-manager/virt-manager/issues/169
Signed-off-by: Cole Robinson <crobinso@redhat.com>
We were not correctly accounting for the internal representation of
some fields, and just trying to a string comparison. We need to be
a bit smarter than that
Fixes: https://github.com/virt-manager/virt-manager/issues/356
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Both these windows versions are now longer supported, and UEFI isn't
the default, so I don't think this hack is much needed anymore
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The bare metal world is moving to a situation where UEFI is going to be
the only supported firmware and there will be a strong expectation for
TPM and SecureBoot support.
With this in mind, if we're enabling UEFI on a VM, it makes sense to
also provide a TPM alongside it.
Since this requires swtpm to be installed we can't do this
unconditionally. The forthcoming libvirt release expands the domain
capabilities to report whether TPMs are supported, so we check that.
The user can disable the default TPM by requesting --tpm none
https://github.com/virt-manager/virt-manager/issues/310
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
When setting CPU defaults we want to force create the topology even if
the user has not specified anything. In particular this allows for
overriding the QEMU defaults, to expose vCPUs as cores instead of
sockets which is a much saner default for Windows.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
In real world silicon though it is rare to have high socket/die counts,
but common to have huge core counts.
Some OS will even refuse to use sockets over a certain count.
Thus we prefer to expose cores to the guest rather than sockets as the
default for missing fields.
This matches a recent change made in QEMU for new machine types
commit 4a0af2930a4e4f64ce551152fdb4b9e7be106408
Author: Yanan Wang <wangyanan55@huawei.com>
Date: Wed Sep 29 10:58:09 2021 +0800
machine: Prefer cores over sockets in smp parsing since 6.2
Closes: https://github.com/virt-manager/virt-manager/issues/155
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The product of sockets * dies * cores * threads must be equal to the
vCPU count. While libvirt and QEMU will report this error scenario,
it makes sense to catch it in virt-install, so we can test our local
logic for setting defaults for topology.
This exposes some inconsistent configurations in the test suite.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Any missing values in the topology need to be calculated based on the
other values which are set.
We can take account of fact that 'total_vcpus' treats any unset values
as being 1 to simplify the way we set topology defaults.
This ensures that topology defaulting takes account of dies.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
It is always permitted to set dies==1 regardless of architecture or
machine type. The only constraint is around setting values > 1, for
archs/machines that don't support the dies concept.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Although using --cpu topology.XXX is the preferred way to set topology,
it is still possible via the --vcpus parameter. For consistency, this
should support the full set of parameters, so dies needs to be added.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
We should only be returning a driver_type value for volumes that
report support_format(), meaning they support file type formats like
qcow2. Any other reported format should be ignored
Dropping the check for 'unknown' value changes one test case a bit,
but it hardcodes raw which is what libvirt gives us anyways, so it's
okay
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Usually, when storage volume is attached as a disk and disk xml is filled with
default values, the "<driver type=...>" value is copied from volume's
"<format type=...>". This makes sense for volumes of storage pool of type
"dir", where format types include "raw, qcow2...".
However, the same approach cannot be used for the storage pool of type "disk".
In that case, format types include "none, linux, fat16, fat32...". Such formats
cannot be used for disk's "<driver type=...>".
Therefore, when generating disk XML for volume of storage pool type "disk",
driver type should always be "raw".
The code comment suggests removing the aliases after a year. It has
now been three years, so it is time for them to go.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Add the ability to set the ioapic driver using the --features argument:
$ virt-install --features ioapic.driver=qemu ...
This results in the following xml:
<features>
...
<ioapic driver="qemu"/>
</features>
This is required in order to install a guest with >255 cpus. Such a
configuration requires an iommu with extended interrupt mode enabled,
which in turn requires IOMMU interrupt remapping to be enabled, which in
turn requires a split I/O APIC.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
The code was only checking the manual approach to enabling UEFI, not the
modern automatic approach.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
libvirt recently changed the nodedev names for mediated devices due to
the fact that mdevctl supports defining multiple mediated devices with
the same UUID as long as only one is active at a time. This means that
the nodedev name changed from the format 'mdev_$UUID' to the format
'mdev_$UUID_$PARENT'.
Unfortunately, virt-install was parsing the nodedev name to extract the
UUID of a mediated device. This fails with the new name format.
Fortunately, in libvirt 7.3.0, a <uuid> field was added to the xml
schema for mdev devices, so we can simply use this instead, and fall
back to the name parsing if it doesn't exist.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Libvirt now validates that all <hostdev> elements refer to distinct host
devices. The test suite violates that constraint by trying to build a
new guest with the same USB devices added to the guest twice, to
validate the various host device syntax options.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Libvirt now validates that all <hostdev> elements refer to distinct host
devices. The test suite violates that constraint by trying to hot-add a
device that alreadye exists in the config.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
For example, if both hotplugging and defining a new NIC, where we
generate the mac address, we need to use the initial generated device
XML for both operations, and not generate different MAC addresses
for each stage.
Resolves: https://github.com/virt-manager/virt-manager/issues/305
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Some test scenarios need to make sure different mac addresses would
_not_ be used in normal operations, but the test suite always generates
the same value. Add some hacks to let the test suite override the
default behavior and use incrementing addresses
Signed-off-by: Cole Robinson <crobinso@redhat.com>
If you call get_xml() on a device that's part of a Guest class,
the last element has correct indent but not the first element.
Steal the indent from the last element and prepend it to the returned
XML
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The libvirt test driver doesn't support hotplug. Add an env variable
to ignore failure, so we can get better test coverage here
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Shuffling bits of code around, adding comments and grouping CLI options
to make the code easier to read and understand at a glance.
Brings the ordering of XML options in line with libvirt's own output as
implemented in `src/conf/cpu_conf.c` and `src/conf/numa_conf.c`.
This includes support for the following options:
* numa.interconnects.latency[0-9]*.initiator
* numa.interconnects.latency[0-9]*.target
* numa.interconnects.latency[0-9]*.cache
* numa.interconnects.latency[0-9]*.type
* numa.interconnects.latency[0-9]*.value
* numa.interconnects.latency[0-9]*.unit
* The same suboptions for `numa.interconnects.bandwith[0-9]*`
Note that the cache= attribute is only explicitly defined for <latency>
nodes in the documentation. However, since <latency> and <bandwidth>
nodes are otherwise identical, the docs also don't explicitly forbid it
for <bandwidth> nodes, and libvirt happily accepts XML that does specify
it for for <bandwidth> nodes, this implements the cache= attribute for
<bandwidth> elements as well.
This includes support for the following options:
* numa.cell[0-9]*.cache[0-9]*.level
* numa.cell[0-9]*.cache[0-9]*.associativity
* numa.cell[0-9]*.cache[0-9]*.policy
* numa.cell[0-9]*.cache[0-9]*.size.value
* numa.cell[0-9]*.cache[0-9]*.size.unit
* numa.cell[0-9]*.cache[0-9]*.line.value
* numa.cell[0-9]*.cache[0-9]*.line.unit
This includes support for the following options:
* `emulatorsched.scheduler`
* `emulatorsched.priority`
* `iothreadsched.iothreads`
* `iothreadsched.scheduler`
* `iothreadsched.priority`
This includes support for the following options:
* `shares`
* `period`
* `quota`
* `global_period`
* `global_quota`
* `emulator_period`
* `emulator_quota`
* `iothread_period`
* `iothread_quota`
libvirt 7.2.0 introduced support for a list of firmware features
that should or should not be present. Libvirt takes these into
account when auto-selecting a firmware. Currently supported features
are `enrolled-keys` and `secure-boot`.
This adds support for evdev inputs which were introduced in 7.4.0,
as well as passthrough inputs and some other misc options to complete
the --input command.
New suboptions:
* source.evdev
* source.dev
* source.repeat
* source.grab
* source.grabToggle
* model
This includes support for the following suboptions:
* name (<shmem name=X>)
* role (<shmem role=X>)
* model.type (<shmem><model type=X/>)
* size (<shmem><size>X)
* size.unit (<shmem><size unit=X/>)
* server.path (<shmem><server path=X/>)
* msi.vectors (<shmem><msi vectors=X/>)
* msi.ioeventfd (<shmem><msi ioeventfd=X/>)
There are two domain XML knobs specific to NVDIMMs that
virt-install doesn't allow to set: <pmem/> and <alignsize/>.
Implement them.
Closes: https://github.com/virt-manager/virt-manager/issues/267
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
As already done for a few architectures, let's add the default
graphics when using aarch64.
This has been tested on a NVIDIA Jetson Xavier NX SBC.
Signed-off-by: Fabiano Fidêncio <fabiano@fidencio.org>
This has been reported for the libvirt qemu driver since v1.3.5,
released June 2016. But we need to keep some fallback logic for
the test driver, and to keep the testsuite happy
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Enable virt-manager GUI to support add, edit, remove, hot-plug and
hot-unplug of mediated devices (like DASDs, APQNs and PCIs) in virtual
server.
It is not possible to edit MDEV when a virtual server is in
running state, as this is not supported by libvirt.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Shalini Chellathurai Saroja <shalini@linux.ibm.com>
Implement kvm.poll-control.state to `virt-install --feature`. It requires
libvirt >= v6.10.0.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Han Han <hhan@redhat.com>
Implement "<transient shareBacking=yes/>" to virtinst to allow a transient disk
to be shared across VMs. It is introduced to libvirt since:
75871da0ec qemu: Allow <transient> disks with images shared accross
VMs
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Han Han <hhan@redhat.com>
Since libvirt v6.9, the element <transient/> is to configure a disk
which discards its changes while VM was active. Support this element
by cmdline option `--disk ...,transient=on`.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Han Han <hhan@redhat.com>
Support rotation_rate attrib which is introduced since libvirt v7.3.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Han Han <hhan@redhat.com>
xorisso is the still maintained isoinfo alternative, and may be
the only iso reading tool in RHEL9, so we need to support it.
Make it the default for our spec file and test suite too
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Add support to install a virtual server with passed-through mediated
device. Mediated device can be created using vGPU attached to
vfio_pci driver or DASD attached to vfio_ccw driver or APQNs attached
to vfio_ap driver.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Shalini Chellathurai Saroja <shalini@linux.ibm.com>
This adds the following suboptions to configure <cputune>:
- vcpusched.vcpus
- vcpusched.scheduler
- vcpusched.priority
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
Bhyve only supports nmdm console type, so use it instead of TYPE_PTY.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Bhyve requires explicit loader configuration. So query
domain capabilities, try to find the "official"
firmware and configure all the necessary loader options.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Choose SATA as a default bus for bhyve as it doesn't support IDE.
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Various checks in the test suite try to match for error/status messages,
so using a localized libvirt (or even using translations of an installed
version of virt-manager in the system) makes those check fail.
Hence, force (using $LANG, and unsetting $LANGUAGE) an English UTF-8
locale.
Thanks Cole Robinson for spotting the right place where to inject the
locale settings.
Fixes: #199
Signed-off-by: Pino Toscano <ptoscano@redhat.com>
Make it work more like gfxdetails. The problem with the current
approach is that it requires effectively rebuilding the whole device
to match the original device when we want to edit a single field,
which is error prone.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This makes it more clear that 'path' is really a special designation
with a bunch of complicated logic behind it. It's also easier to
grep for
Signed-off-by: Cole Robinson <crobinso@redhat.com>
We do this by faking an install phase whenever cloudinit media is
specified, which isn't really the right abstraction and will leak
into virt-install behavior (like doing 2 boots), but it's the simplest
fix for now
Fixes: #178
Signed-off-by: Cole Robinson <crobinso@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1883008
In the example above this is because the VM XML has an invalid machine
type, so domcaps fetching entirely fails, and a get_enum() call
then fails. But this could happen if using virt-manager against an
older libvirt that doesn't advertise the enum
Signed-off-by: Cole Robinson <crobinso@redhat.com>
delete and createvm tests launch a dialog which obstructs the
manager UI. The location can be kinda random, and it might obstruct
selecting the connection in the manager window. Go back to using
the drag() window pattern to make this more deterministic
Signed-off-by: Cole Robinson <crobinso@redhat.com>
And absord device building from addhardware. This moves all the
knowledge to gfxdetails, which saves sprinkling it around in other
places
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Close accelerator changes ctrl+w -> ctrl+shift+w
Quit accelerator changes ctrl+q -> ctrl+shift+q
After aafb874c8, if the mouse pointer isn't inside the console
window, it has keyboard focus but ctrl+w will be sent to the vmwindow
and not the VM. ctrl+w is a common shortcut for deleting a word so
this is pretty disruptive if you are typing inside the VM
Use gnome-terminal-esque accelerators starting with ctrl+shift to
reduce the chance of collision.
https://bugzilla.redhat.com/show_bug.cgi?id=1880295
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Historically we have not advertised host-passthrough because it was
not recommended for general usage. That stance is softening,
tools like gnome-boxes already set it as the default, and users
continue to ask about it.
We may change the default in virt-manager but it will take more
discussion. This is a tiny move in the direction of hiding it less
than we already do.
Drop the label for host-model and call it by its libvirt XML name,
since otherwise it's hard to tell which combo choice is for each
value
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Inspired by some discussion from here:
https://bugzilla.redhat.com/show_bug.cgi?id=1759454
Most libvirt storage volume creation doesn't actually do anything
with allocation, besides interpreting cap == alloc and cap != alloc.
The exceptions are zfs volumes, and raw file volumes. But it's unclear
what the usecase is for the latter at all.
This drops the allocation spinner and adds checkbox in its place
'Allocate entire volume now'. When enabled, it sets cap == alloc.
We only show this for file volumes. For qcow2 it defaults to unselected
(sparse), for all others it defaults to selected. If it's not showing,
it defaults to selected.
Bundled with this change is showing this field for qcow2, where
we previously only allowed nonsparse here. Libvirt and qemu-img
support non-sparse qcow2 these days.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1759454
See 15a6a7e210
The idea behind virt-manager's sparse vs nonsparse default, is that if
the user selected 'raw' for as the default image format, assume they
want to maximize performance, so fully allocate the disk.
qcow2 didn't support anything except sparse, so the sparse=True vs
sparse=False made no difference. So we always set sparse=False
Then qcow2 grows non-sparse support, and virt-manager is suddenly
defaulting to it, which is not the intention.
Default to sparse when requested format isn't raw
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This is another preference that was added before anyone ever asked
for it. I'm fine with suggesting users remove the device manually
if they don't want it
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Remove the preference option to disable this. This was added with
the initial usbredir support because I was afraid people would
complain. They did complain, but only about the auto redir behavior
of the spice client. We still have a toggle to disable that behavior
If people don't want usbredir devices, I'm comfortable telling
them to remove them manually, or use virt-install
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* Convert to pytest style functions
* Move lots of shared code to our App class
* Reduce dogtail sleep amounts to speed up the whole testsuite
* Improve robustness in a lot of areas
Signed-off-by: Cole Robinson <crobinso@redhat.com>
For the few bits we are hitting specific code paths, break them
out or fold them into other test cases
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Make use of the new helper for showing a standard error message for two
conflicting cli options. This also catches one untranslatable message.
Signed-off-by: Pino Toscano <ptoscano@redhat.com>
Instead of using the title of the dialog and prepending the connection
label, create a new title as a single string. This way it is possible to
translate this title as single sentence.
Signed-off-by: Pino Toscano <ptoscano@redhat.com>
Latest libvirt uses socket activation, so libvirtd.service in
offline state does not indicate a problem necessarily.
Also on Fedora nowadays we have a weak RPM dependency on
libvirt-daemon which we didn't in the past.
Both things combine to make this code less useful and less accurate,
so let's remove most of it.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
virtManager/createvm.py:982:8: W0128: Redeclared variable 'ignore' in assignment (redeclared-assigned-name)
tests/uitests/lib/app.py:12:0: C0411: third party import "from gi.repository import Gio" should be placed before "import dogtail.rawinput" (wrong-import-order)
Signed-off-by: Chen Hanxiao <chenhx.fnst@cn.fujitsu.com>
If specified, this errors if no OS name was detected or manually set.
So --os-variant detect=on,require=on will error if no OS is detected.
name= can be used as a fallback, so test and document this case
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This adds the following --os-variant suboptions
* name=, short-id=
* id=
* detect=on|off
Functionally this does not change behavior, just adds explicit
sub options for behavior we already support
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The --os-variant option naming is pretty crappy and mostly a historical
artifact. Ideally this would be named just `--os` but I'm afraid that
would cause confusion with libvirt's <os> XML
Add --osinfo as an alternate commandline naming. If we ever want to
transition documented use of --os-variant it will help to have the
alternative around for a few releases
Signed-off-by: Cole Robinson <crobinso@redhat.com>
The --xml option allows users to request raw XML edits to virt-install
or virt-xml generated XML. This gives users a bit of a workaround
incase we don't have proper support for some XML property. The --xml
option can gain more features in the future if it makes sense, like
setting XML namespaces for example.
Basic usage is like: virt-install --xml ./@foo=bar ...
Which will change the generated <domain> XML to have
<domain foo='bar' ...
virt-xml works similarly. It can only be combined with --edit currently.
This only works with xpaths rooted against the entire document.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Break utils.py apart into a whole uitests/lib/ directory with
* lib/_dogtailinit.py: all the dogtail library init we need
* lib/_node.py: extending our dogtail node class with more functions
* lib/app.py: VMMDogtailApp
* lib/util.py: util functions plus all the special helpers previously
in our custom TestCase
* lib/testcase.py: The TestCase that sets and tears down self.app
Adjust callers to match and make it easier to eventually convert to
native pytest usage
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This uncovered some areas in details.py we weren't handling that
the search view unselects the current selection
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Similar to behavior we have in virt-manager, if the user destroys the
VM during the VM install process, don't invoke the post install
reboot.
https://bugzilla.redhat.com/show_bug.cgi?id=1818089
Signed-off-by: Cole Robinson <crobinso@redhat.com>
As part of making virt-manager cooperate better with external viewers,
add an option to disable console autoconnect. When opening a VM window
for a running VM, you'll see a 'Connect to console' button in place
of the spice/vnc viewer. Click that and things proceed like normal.
This is useful to prevent virt-manager from disconnecting a virt-viewer
instance that's already attached to a VM
https://bugzilla.redhat.com/show_bug.cgi?id=1793876
Signed-off-by: Cole Robinson <crobinso@redhat.com>
When a VM window is launched for the first for a VM, currently we
set the top window size to 800x600 which is small and arbitrary and
is universally shrinks the viewer too much to fit any OS installer
I can find.
Instead do some hacks to resize the window to accomodate a viewer
widget of 1024x768 which seems to be what QXL graphics give us for
win10 and Fedora 32 installers. So for new VMs hitting the OS installer
we don't see scrollbars.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Right now this is tied to widget focus, which is too strong. This
changes it so that say clicking on the window title or toolbar then
allows the user to use Alt+F to trigger the File menu for example.
This roughly matches how virt-viewer works
https://bugzilla.redhat.com/show_bug.cgi?id=1824480
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* Move all the menu building to its own class, for clarity
* Rename the menu 'Consoles' since it contains graphical choice as well
* Strip out the VM console duplicate if it exists
* Simplify the code a bit
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* rename kernelupload.py to volumeupload.py and make the entrypoint
more generic
* move all upload invocation to the Installer class
* use it with cloudinit and unattended ISO generation if required
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This is a quirk of libvirt that the first <device> is usually a
logical duplicate of the first <serial> device. Adjust virt-xml to
understand this quirk and remove both devices at the same time,
like we already do in virt-manager
https://bugzilla.redhat.com/show_bug.cgi?id=1685541
Signed-off-by: Cole Robinson <crobinso@redhat.com>
coreos is going to start using disk serial for ignition disk, so
setting this in the UI for distro testing will become more common
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Share the UI for changing all these disk properties:
- shareable
- readonly
- removable
- cache
- discard
- detect zeroes
Move them all under the 'Advanced options' expander in details, and
add the checkbox options to the addhardware wizard.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Allows the user to tweak the XML at the destination, which is already
something that libvirt supports
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* Drop the network editing, users can use the details window
* Drop the combo box approach in favor of a regular treeview
* Drop a lot validation checks which are redundant with modern
virtinst. We probably lose some checks but I don't think it's
too important
* Use the cloner API
* Add uitest coverage
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This let's us move more of the preserve logic to virtclone.py
and prep more things to share with virt-manager
Signed-off-by: Cole Robinson <crobinso@redhat.com>
test:///default saves changes for the lifetime of the connecting
process, which can cause weird interdependent issues in the test
suite. Duplicating it with driver XML will avoid those issues
popping up.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
* Centralize lots of disk building
* Open code virt-clone specific behavior at the source
* Drop a lot or properties
* Move most testing to test_cli.py
* Generally a ton of cleanup
virt-manager clone wizard has not been converted yet so is totally
broken after this commit
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Cover more cases via the virt-clone CLI testing, than just the
test_cloner.py unit style testing.
Change most of the virt-clone --print-xml testing to also attempt the
clone operation as well via a hidden cli option, to ensure we aren't
testing XML of any bogus operations
Signed-off-by: Cole Robinson <crobinso@redhat.com>
This is just a test suite hack from before we had other methods
of specifying cloneable guests. It will be removed soon
We need to tweak some test XML to fix code coverage after
Signed-off-by: Cole Robinson <crobinso@redhat.com>
As much as we can. There's not any current way to actually test
the full guestfs implementation, so we exclude it all
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Refactor the internals to cleanly separate the pieces that fill in
the UI, and the pieces that react to UI state to dynamically show/hide
fields.
Improve spice GL warnings while he are here, and several other minor
fixes
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Make it explicit that all uses of this is actually the object
name. We already leaked this abstraction in several places so better
to make it explicit. This also communicates to users that this is a
field that is not immutable so it shouldn't be used as a unique key
Signed-off-by: Cole Robinson <crobinso@redhat.com>
qxl doesn't add anything for the VNC case, and qxl is more likely
to be compiled out of qemu since it is dependent on spice, so vga
is the safer default for getting a working config
https://bugzilla.redhat.com/show_bug.cgi?id=1833704
Signed-off-by: Cole Robinson <crobinso@redhat.com>
- create a versioned directory for the current VERSION file, fixing the
typo in the distro name
- move the test to an own function
Reviewed-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Pino Toscano <ptoscano@redhat.com>
This code path was never hit because it came after caps.guest_lookup
which errors in this case. We need to check things earlier
Signed-off-by: Cole Robinson <crobinso@redhat.com>
In all modern local cases we will be using openFD to get a direct
file descriptor from the VM, so this code should never be triggered
nowadays
Signed-off-by: Cole Robinson <crobinso@redhat.com>
