This patch adds tests for common errors when working with patches.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
In ac4373088 the error handling was made more strict. By that, failures
during patch apply instantly bailed out instead of returning and
providing the more useful error message. This patch restores the error
handling by marking the commands as not-fail and encapsulating the error
into a more specific exception with a useful error message.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
As the tasks with fail=True property return via an exception for a
long time, the handling of the return code is not required anymore.
This simplifies the code by removing dead code paths.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
When async tasks fail, all exceptions need to be collected to not get
subsequent exceptions about invalid future states. This is achieved by
gathering the task results, instead of just waiting for them. By
gathering the results, also user-requested cancellation (e.g. via
ctrl-c) works without throwing tons of additional exceptions.
Since ac437308 we more likely run into that case, which unvealed the bug.
By properly handling the exception, a TaskResultError is returned
instead of the underlying CommandExecError. This change is reflected in
the corresponding unit test.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This adds a test for common command exec errors when checking out
repositories. The test checks if the correct error is thrown
(CommandExecError) instead of failing silently or with an
unspecific exception.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
When executing async commands, bail out in case the fail flag is set
and a command returns a non zero exit code. This is required to
ensure that further downstream commands are not executed as well.
Previously, the execution continued till the first error of the command
execution itself happened (e.g. command not found).
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
The common case that an include file cannot be found is now wrapped into
a LoadConfigException. By that, the user gets a more meaningful error
message on the console (instead of a stacktrace).
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch configures Coverage.py to measure only the code coverage of
kas itself. In addition, we add a note in the devguide how to create and
inspect the coverage data.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch handles common Kconfig errors (file not found, invalid
config) and reports them as user errors. By that, the root cause of the
error is easier to spot by the user.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch replaces all direct invocations of sys.exit outside of the
main invocation to KasUserError based exceptions. By that, only one
method for returning is used and return codes can be handled
consistently. In addition, this makes it possible to handle specific
errors differently.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch adds the KasUserError exception class to distinguish between
internal kas exceptions and user or configuration errors. Exceptions
previously raised on user errors are ported over by deriving
KasUserError. In case of user errors, only the exception message is
shown, but no stacktrace. This makes it easier for users to locate the
issue as the reason is now stated in the last line of the output.
Kas internal exceptions are not subject to this change to help the
developers to find the root cause more easily.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
The 'kas' function does not return any value, hence we must not use the
return value.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
As part of the tests of the dump plugins a variable was shadowed. By
that, one branch of the test was not executed.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch forwards the log-level parameter from kas-container to kas
inside the container. For backward compatibility, the -d parameter is
deprecated but kept and also forwarded as-is to kas. By that, a more recent
kas-container script can still be used to run an older kas container.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
KAS by default will output a lot of information (INFO) messages for
all operations, which makes it difficult to spot warnings thru all
that 'noise'.
Add a command line argument so that the default log level can be
modified.
For backward compatibility, the --debug parameter is still supported
but marked as deprecated in the help message.
Signed-off-by: Hugo Villeneuve <hvilleneuve@dimonoff.com>
[Jan: style fixes]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
To make it easier to display (and modify) the default log level,
especially with the introduction of the new (future commit) argument
--log-level.
Signed-off-by: Hugo Villeneuve <hvilleneuve@dimonoff.com>
[Jan: style fix]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch improves the documentation regarding how lockfiles work, where
kas searches for them and how to create/update them. A dedicated section
about the locking mechanism is added to the userguide. The documentation
of the kas internal logic is improved by making the wordings more precise
(e.g. lockspec vs. lockfile).
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
[Jan: fixed overlong lines]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch makes the creation and update of a lockfile more convenient.
When running the dump plugin in inplace mode, a lockfile is created next
to the first file on the kas cmdline. By that, the repo directory also
needs to be mounted rw. Otherwise the kas inside the container cannot
create the file.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
The auto-generated lockfiles should be terminated with a newline to be
POSIX compliant. Previously, this only worked for output via stdout but
not for inplace operations. This is fixed by appending the newline to
the active output target.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This commit adds a test that check the creation, effectiveness and
update of a lockfile. Testing this functionality via the dump plugin is
sufficient, as the plugin directly uses the checkout workflow.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
[Jan: fix over-long lines and removed assert brackets]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch adds the --lock option to the dump plugin. When enabled, the
output only contains the resolved refspecs of each repo (as valid kas
format). By that, floating branches can be used in the projects kas files
and these can be pinned to fixed revisions, when required.
When using --lock in combination with --inplace, a lockfile named
<filename>.lock.<ext> is created next to the <filename>.<ext>. In case
multiple files are added to the kas CLI, the lockfile is only created
for the first file (by considering the merged information from all files).
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
[Jan: fold in Python 3.6 support]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
When checking out repositories, check if a file <filename>.lock.<ext>
exists next to the file specified first on the kas CLI. In case this
file exists and the --update option is not specified, automatically
append this file to the kas CLI before performing any other kas
operations.
When --update is specified, the lockfile is ignored.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch adds the top-level `overrides` entry, which is used to
override (or pin) the refspec of repositories. The main difference to a
direct override is that this logic only applies to repos that are
already defined. By that, a superset of all repos can be added to this
entry (similar to a global lockfile), but only the currently active ones
are affected. A new top-level keyword is required because everything
below the "repos" keyword is potentially defined by "default" values.
For the locking mechanism, a clear separation between overrides (only
override if existing) and definitions is required to be able to define a
global lockfile with all possible repos, while just defining some repos.
Proposed-by: Ross Burton <ross@burtonini.com>
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
[Jan: also bump __file_version__]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
As all currently supported distros provide at least python 3.6, we
drop the support for 3.5. At time of this commit, the python versions
of the supported distros are as following:
- Debian: 3.7 (buster)
- Ubuntu: 3.7 (18.04)
- Fedora: 3.11 (Fedora 36)
- RHEL: 2.x (RHEL 7), 3.6 (RHEL 8)
- OpenSUSE: 3.6 (Leap / 15.4)
While updating the lower bound version, we also unify the upper bound in
the setup.py script with the versions tested in the CI.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch adds the required meta-schema identifiers to allow automatic
validation against a fixed version of json-schema.
Reported-by: Ross Burton <ross@burtonini.com>
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This allows the configuration of the git option 'credential.usehttppath' if
the used credential helper requires this.
Signed-off-by: Christoph Freundl <Christoph.Freundl@ifm.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Currently kas-container fails silently if readlink is unable to resolve
a path.
Add -v to each readlink command to get errors reported.
Before:
$ KAS_BUILD_DIR=/scratch/rwtypo/leia kas-container shell leia.yml
$ echo $?
1
After:
$ KAS_BUILD_DIR=/scratch/rwtypo/leia kas-container shell leia.yml
readlink: /scratch/rwtypo/leia: No such file or directory
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Without this package kas will not use colorful logging.
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
kas-container is carefully written to be POSIX shell compliant.
Let's do the same with container-entrypoint to be consistent.
While we're here, remove the only bashism used in container-entrypoint.
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch puts the code examples of the dump plugin into code-blocks in
the documentation. By that, the documentation is easier to read.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
If we run kas-container with --isar flag and did not set build_system,
we end up calling both enable_isar_mode and enable_oe_mode. This can
trigger:
Error: keep-id is only supported in rootless mode
Signed-off-by: Stefan Müller-Klieser <s.mueller-klieser@phytec.de>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Since 492b2c56, we create user and group upfront, now using 1000:1000 as
IDs. This can cause unexpected glitches when using the container without
kas-container in environments where older version already created files
with the previously used IDs. In order to stay compatible, switch the
default IDs back to 30000:30000.
This will not affect any user of kas-container.
Reported-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
As we no longer create the builder user at runtime, placing data into
/etc/skel at runtime is semantically not correct anymore. Instead, we
bind mount host paths below /var/kas/userdata. By that, we now place
the data into a directory which is fully handled by us.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This patch fixes a regression introduced in 492b2c5. As the builder user
is no longer created in the entrypoint, the data from the home skeleton
is also not copied anymore. This breaks the ssh config (including
known_hosts) when using kas-container with --ssh-dir, as the ssh dir is
mounted into the skeleton, but not copied to the builders home.
With this patch, we now explicitly copy the .ssh folder into the builder
users home, in case it is mounted from the host.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
With the added auto-caching logic of the referenced repos in 1c2c859,
kas-container has to mount this directory in read-write mode.
Otherwise, initial clones will fail.
Signed-off-by: Felix Moessbauer <felix.moessbauer@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Neither git nor hg currently provide a production-ready replacement for
weak SHA-1 commit IDs. Furthermore, kas mixes commit IDs and symbolic
commit names in refspec. This permits attackers who gained control over
a repository that kas fetches from to present manipulated content
without kas noticing this.
Aditya Sirish A Yelgundhalli recently reported one potential attack
scenario, using branches that shadow commit IDs. While trying to
mitigate this particular case, it became clear that there is no simple
solutions with the given tools and interfaces.
For now, warn prominently that only trusted sources should be used.
There are extensions planned to address the issue at its root, likely by
introducing content checksums.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
The user environment may lack /usr/sbin while certain podman
configuration will need, e.g., iptables for the setup. This can cause
Error: plugin type="bridge" failed (add): cni plugin bridge failed: failed to locate iptables: exec: "iptables": executable file not found in $PATH
Resolved that by appending /usr/sbin to the PATH in privileged podman
mode.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Dockerfile and container-entrypoint were missing that header.
Furthermore, the leading comment in Dockerfile got out-of-date, and we
should rather add section marker for the kas and kas-isar targets.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This is for the sake of the gitlab-ci runner which does not properly
aligns the ownership of the repo it checks out with the UID:GID of
our builder user. Reason not yet understood and hard to debug (logging
of the runner is incomplete).
Work around this issue by disabling safe.directory checks in case the
container is called without kas-container as wrapper (means, when it is
called without setting "--user=root"). This preserves git's checks for
the common interactive case, the more critical one.
Reported-by: Ross Burton <ross@burtonini.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Already create the builder user/group during container image build and
only align the IDs in the entrypoint if started with a non-zero USER_ID.
The primary gain is code simplification because this removes some
dynamics from the entrypoint.
As this refactoring avoids that gitlab-ci runners start the container as
root, it was also supposed to resolve the mismatch between the owner of
the checked-out repo and builder user. Unfortunately, this does not work
yet, and the reason is still unclear.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Leave the isar-only commands commented-out in the container-entrypoint
and simply remove the comment when building kas-isar. This is simpler,
more readable and also more robust against changes of the entrypoint
file.
While at it, avoid a separate layer for modifying container-entrypoint.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
The pattern
mkdir test_commands
rm -rf test_commands
makes no sense. If there were test_commands before, mkdir would have
failed. Simply build the path and copy the source content in.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
This file shall describe the vulnerability disclosure process for kas
and the security context in which kas should be seen. Reporting
vulnerabilities via github has already been activated.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
For a long time already (c9326cc1ed), the documentation suggests a
different ordering of entries generated by *_conf_header than we
actually create. Finally fix this.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Was lost in the refactoring of a6b18abc8a.
Signed-off-by: Hannah Kiekens <hannah.kiekens@mind.be>
[Jan: refer to the causing commit]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Since a6b18abc8a, we only have a single Dockerfile and a --target
instead.
We actually also need to touch the unmodified proxy line, wrapping it
around, so that doc8 remains happy.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
