This outputs a TOML template of the settings needed for setting the
upload credentials. It can be passed to 'upload start' and to 'compose
start', as well as used to set the profile for 'providers push'
Since we have both compose uuids and upload uuids they need to be
clearly named. This updates the upload naming to use 'upload_uuid' in
the inputs, and 'upload_id' in the output (_id instead of _uuid for
consistency with build_id naming in the status responses).
This also adds 'upload_id' to the /upload/log response.
This allows deleting a provider's profile. Pass the provider and profile
name like:
/api/v1/upload/providers/delete/azure/test-settings
A standard json response will be returned.
This extends the /compose/ route to support uploading with either an
existing profile, or with one-time use settings passed in the POST.
To select a profile include the provider and profile, as returned by
`/uploads/providers`:
"upload": {
"image_name": "My Image",
"provider": "azure",
"profile": "production-azure-settings"
}
uploads should only be included in the V1 API routes, add `api`
selection to the relevant helper functions and calls to them from v0.py
Add new V1 routes with api=1 to include the uploads information in the
results.
Also add tests to ensure that V0 requests do not include uploads.
Currently, Azure, vSphere, and OpenStack are supported. See
https://github.com/weldr/lorax/pull/826 for more details about this new
feature.
I've called the upload library "lifted" as a reference to Seuss'
The Lorax -- in the book, the Lorax lifts himself up by the seat of his
pants through a hole in the smog clouds, and they start calling him the
"Lifted Lorax."
This adds new features to the /compose route under API v1
A recipe that is valid TOML can still be an invalid recipe (eg. missing
the 'name' field) so this should also catch RecipeError.
Also added tests for this, as well as making sure commit_recipe_file()
raises the correct errors.
Resolves: rhbz#1755068
Add dmsquash-live-ntfs to the default dracut modules in livecd builds.
The omission of this is probably why:
https://bugzilla.redhat.com/show_bug.cgi?id=1449410
didn't work out as intended (I suspect it was closed unfixed).
Without this, initramfs winds up with the ntfs-3g tools lying
around, but no hooks cause them to actually get used.
AFAICT, this is a legit bugfix, not a "proposed enhancement"
so to speak. To test that this actually fixed the problem (well,
/a/ problem, at least), I used some scripts:
https://github.com/gmt/test-respin-fedora-ntfsable-iso
Signed-off-by: Greg Turner <gmt@be-evil.net>
Signed-off-by: Brian C. Lane <bcl@redhat.com>
This loads the system dnf vars from /etc/dnf/vars at startup if
system repos are enabled, and it substitutes the values in the sources
when loaded, and when a new source is added.
Also includes tests.
This changes the source 'name' field to match the DNF usage of it as a
descriptive string. 'id' is now used as the short name to refer to the
source. The v0 API remains unchanged.
Tests for v1 behavior have been added.
Now that the v1 API is in use the status message will return api: 1
This creates a tar suitable for use with the anaconda kickstart liveimg
command. It adds the kernel, grub2, and grub2-tools packages to the tar
template.
Normally the runtime image is an ext4 filesystem image that is
compressed with squashfs. dracut now supports setting up an overlayfs
when it detects a bare filesystem tree inside the squashfs.
This commit adds support for a --squashfs-only option which places the
root tree directly in the squashfs.img
In python 3 f.seek() on text doesn't work like it does in py2/C because
text is now unicode. So change read_tail to use byte mode and take
unicode into account. Also add tests for it.
The VALID_API_STRING function allows for characters that should not be
allowed in blueprint names. VALID_BLUEPRINT_NAME allows us to
specifically check if a blueprint contains a valid name.
dnf seems to have changed the default for skip_if_unavailable. Some
mock repositories are still around in later tests, which then fail
because metadata cannot be synced.
Also expose skip_if_unavailable in dnf_repo_to_file_repo(), so that
tests checking for equality of repo files continue to pass.
This makes sure that required fields are included, and that sections are
not empty. It does not check for all optional fields.
If there are errors it will gather up all of them and then raise a
RecipeError with a string of all the errors.
The new toml library, introduced with abe7df34f, outputs different
whitespace from the old one. Fix the test expectation and strip()
results from toml.dumps(), because it contains superfluous newlines at
the end.
This also includes extensive tests for each of the currently supported
customizations. It should be generic enough to continue working as long
as the list of dicts includes a 'name' or 'user' field in the dict.
Otherwise support for a new dict key will need to be added to the
customizations_diff function.
Instead of setting up the routes inside a function we can now use a
BlueprintSkip class, which allows us to register them at different
routes (eg. /api/v0/ and /api/v1/) and override any routes that will be
replaced by the new API version.
When adding a new API we want to use the old code for any routes that
aren't being overridden.
This modifies the Flask Blueprint class so that a skip_rules list can be
passed to server.register_blueprint()
To maintain consistency with the other options this changes firewall to
combine the existing settings from the image template with the settings
from the blueprint.
Also updated the docs, added a new test for it, and sorted the output
for consistency.
Add support for enabling and disabling systemd services in the
blueprint. It works like this:
[customizations.services]
enabled = ["sshd", "cockpit.socket", "httpd"]
disabled = ["postfix", "telnetd"]
They are *added* to any existing settings in the kickstart templates.
You can now open ports in the firewall, using port numbers or service
names:
[customizations.firewall]
ports = ["22:tcp", "80:tcp", "imap:tcp", "53:tcp", "53:udp"]
Or enable/disable services registered with firewalld:
[customizations.firewall.services]
enabled = ["ftp", "ntp", "dhcp"]
disabled = ["telnet"]
If the template contains firewall --disabled it cannot be overridden,
under the assumption that it is required for the image to boot in the
selected environment.
You can now set the keyboard layout and language. Eg.
[customizations.locale]
languages = ["en_CA.utf8", "en_HK.utf8"]
keyboard = "de (dvorak)"
Existing entries in the kickstart templates are replaced with the new
ones. If there are no entries then it will default to 'keyboard us' and
'lang en_US.UTF-8'
Includes tests, and leaves the existing keyboard and lang entries in the
templates with a note that they can be replaced by the blueprint.
This fixes the customizations list problem earlier than in
add_customizations.
In the recipe it should be [customizations] not [[customizations]]
which creates a list. If it was used that way grab the first element and
replace the list with it.
For example:
[customizations.timezone]
timezone = "US/Samoa"
ntpservers = ["0.pool.ntp.org"]
Also includes tests.
This removes the timezone kickstart command from all of the templates
except for google.ks which needs to set it's own ntp servers and timezone.
If timezone isn't included in the blueprint, and it is not already in a
template, it will be set to 'timezone UTC' by default.
If timezone is set in a template it is left as-is, under the assumption
that the image type requires it to boot correctly.
This is based on the VHD compose type, with the following differences:
* Use the vhdx format instead of vhd
* No WALinuxAgent
* Install hyperv-daemons
The hyperv-daemons are activated through udev rules, so there is no need
to add them to the services line.
This option will create an optionally compressed tarball containing a
disk image. This format is used by Google's Compute Engine.
This also adds a new option, tar_disk_name, to set the name of the disk
image that will be wrapped in the final tarball. opts.image_name
continues to be the final output file name.
If provided, round the disk image size up to a multiple of the value.
This allows for image formats with specific size-alignment requirements
(e.g., disk size must be in GiB).
rpmfluff was including / in the rpm, which conflicts with
filesystem.rpm
The rpm globs are pretty limited, and we don't actually know the file
paths until later, so we have to use a glob or a directory.
So when the destination is / it now uses /* to select all the files and
sub-directories in the archive. The limitation of this is that it cannot
support dotfiles directly under /, they will cause a rpmbuild error.
For destinations other than / it uses the name of the directory, so
dotfiles are fine in that situation.
This allows iso builds to include the extra kernel boot parameters by
passing them to the arch-specific live/*tmpl template.
Also adds tests to make sure it is written to config.toml in the build
metadata.
The shlex splitting can fail, resulting in error messages like:
ERROR livemedia-creator: No closing quotation
without any context in the log files. This logs the line that failed to
be split and expanded.
Sometimes it is necessary to modify the kernel command-line of the
image, this adds support for a [customizations.kernel] section to the
blueprint:
[customizations.kernel]
append = "nosmt=force"
This will be appended to the kickstart's bootloader --append argument.
Includes tests for modifying the bootloader line, the kickstart
template, and examining the final-kickstart.ks created for a compose.
This hooks up creation of the rpm to the build, adds it to the
kickstart, and passes the url to Anaconda. The dnf repo with the rpms is
created under the results directory so it will be included when
downloading the build's results.
This adds support, documentation, and testing for a [[repos.git]]
blueprint section that can be used to install files from a git
repository. It will create an rpm that will be added to the build,
and included in the metadata that can be downloaded. This allows you to
accurately keep track of the source of configuration files and extra
metadata that is added to the build.
The source repo and reference will be listed in the rpm's summary making
it easy to discover on the installed system.
Reading a blueprint wasn't checking to see if it had been deleted so it
was returning the most recent commit before it had been deleted. This
allowed things like starting a compose with a blueprint that technically
doesn't exist.
One exception to this is the /changes/ route, it must be available so
that you can use the commit hash to undo a delete.
This also adds tests for the various operations.
Resolves: rhbz#1682113
In order to support iso creation on multiple arches with the templates
we need to be able to select different packages based on arch.
lorax-composer uses the arch-specific Lorax templates in order to
generate the output iso so this patch:
1. Creates a new template and type to parse it, live-install.tmpl
which contains only installpkg commands and #if clauses for arch
2. Removes bootloader related packages from the live-iso.ks
3. Remove dracut-config-rescue exclusion because it can cause problems
with some blueprints.
4. Switch logo requirement to system-logos which is satisfied by
generic-logos or fedora-logos. This prevents conflicts when a blueprint
installs fedora-release-workstation.
So in the future, if x86.tmpl, etc. need a new package to support
creating the iso it should be added to the correct section in
./share/live/live-install.tmpl
reqpart can be used to make kickstarts more platform agnostic, creating
needed partitions without lmc having to keep track of the arch-specific
needs. eg. ppc64 needs prepboot and /boot
This increases the size of the disk based on whether reqpart or
reqpart --add-boot is in the kickstart.
Note that this is only valid for partitioned disk output types, not
for filesystem images or live iso output.
It is not actually needed. projects_info deduplicates the package list,
placing other builds into the builds list instead of making a new
package entry. So it returns a sorted and deduped list of packages, as
expected.
In livemedia-creator's usage of this it can never pass in None, but if
someone were to import the library and use it, it would crash with
NoneType. So add the extra checks to make sure cancel_func isn't None,
just in case.
When using LMC to virt-install a system to an image, cancel_func is not
provided in run_creator, causing a TypeError (NoneType object is not
callable).
Signed-off-by: Yuval Turgeman <yturgema@redhat.com>
The reason for the 3G minimum was because anaconda had a bug with how it
calculated minimum disk size when using kickstart. The gix for this has
been in Anaconda since 29.19-1, so we can now remove our limit and
create somewhat smaller disk images.
Anaconda can leave child processes and mounts around when it crashes or
is canceled before finishing. It also sometimes unmounts unrelated file
systems (https://github.com/rhinstaller/anaconda/issues/1791).
Run it in a mount and pid namespace to clean up after it.
We need to be root to read the certificates that give access to the
package repos. Right now, the alternative seems to be changing
permissions on the certs themselves, which seems less good. We're
running anaconda as root anyway.
If a repository has `sslcacert`, `sslclientcert`, or `ssclientkey` set,
pass them to anaconda through the kickstart file. This is mostly the
case when using RHEL repositories that are accessed through a
subscription.
In some cases when the host has, for whatever reason, multiple copies of
the same repo listed the build may fail with an error about running out
of space.
So this commit removes duplicate entries after the host's repos have been
loaded. It also adjusts some of the test repos to use different
temporary repo names for the tests.
If systemd's tmpfiles.d timer is executed while lorax is running it will
remove any files and directories older than 30 days. This is what has
been causing the occasional error where /proc/ would seem to vanish
during the install.
Upstream has proposed this solution, https://github.com/systemd/systemd/pull/11482
but until that is released we need a work-around to protect the lorax
files.
This commit does several things:
* Move the default tmpdir from /var/tmp/ to /var/tmp/lorax/
* Add a lorax.conf tmpfiles.d file that prevents systemd-tmpfiles from
removing anything under /var/tmp/lorax/
* Add an exit handler to lorax so that temporary directories are removed on
exit or on a python traceback.
* Use flock to lock access to the tempdir while lorax is running.
* Remove any unlocked tempdirs named /var/tmp/lorax/lorax.* at startup
Note that the exit handler will not remove the tempdir if lorax is
killed with a signal -- those are being caught by dnf and prevent the
exit handler from running.
systemd-tmpfiles cannot clean up the tempdirs at boot time because they
contain files labeled as shadow_t, so we have to remove those when lorax
runs. It uses the flock to prevent removing any directories created by
parallel instances of lorax and only removes ones that are unlocked.
Worst case they will be around until the first run of lorax after a
reboot.
If you want to keep the working directory around for debugging purposes
use --workdir /var/tmp/lorax/my-workdir and it won't be removed by
lorax.
