Allowing for multiple manifest per artifact just makes the code and cli
design harder to work with it. It is not clear how mounting, extracting
or edit on a multi manifest artifact should have worked.
A single manifest should make the code much easier to work with.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
like images and containers, it could be handy to have a --noheading
option that removes the headings on the output.
Signed-off-by: Brent Baude <bbaude@redhat.com>
added a --no-trunc flag to artifact ls, which follows what images has
done. by default now, the ls output will have the shortened 12
character digest. the --no-trunc will output the full digest.
Signed-off-by: Brent Baude <bbaude@redhat.com>
podman artifact add now supports two new command line switches.
--type string that describes the type of artifact
--annotation string slice in the form of key=val
These new options allow users to "tag" information in on their artifacts
for any number of purposes down the line
RUN-2446
Signed-off-by: Brent Baude <bbaude@redhat.com>
BuildOrigin is a field that can be set at build time by packagers. This helps us trace how and where the binary was built and installed from, allowing us to see if the issue is due to a specfic installation or a general podman bug. This field shows up in podman version and in podman info when populated. Note that podman info has a new field, Client, that only appears when running podman info using the remote client.
Automatically set the BuildOrigin field when building the macOS pkginstaller to pkginstaller.
Usage: make podman-remote BUILD_ORIGIN="mypackaging"
Signed-off-by: Ashley Cui <acui@redhat.com>
They are new and failing on remote, needs to be looked at (#25138)
For now skip them so we can have a proper buildah vendored for rc2.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The `podman system prune` command is able to remove build containers that were created during the build, but were not removed because the build terminated unexpectedly.
By default, build containers are not removed to prevent interference with builds in progress. Use the **--build** flag when running the command to remove build containers as well.
Fixes: https://issues.redhat.com/browse/RHEL-62009
Signed-off-by: Jan Rodák <hony.com@seznam.cz>
The test `podman selinux: check unsupported relabel` has been failing
recently on Fedora rawhide.
This is due to a regression in the `ls` command itself. Workaround for
now is to switch to `getfattr -n security.selinux ...`.
Ref: https://github.com/containers/podman/issues/25132#issuecomment-2615744915Fixes: #25132
Signed-off-by: Lokesh Mandvekar <lsm5@fedoraproject.org>
This is a generalization of PodmanExitCleanly, scalable
to an arbitrary number of possible options.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
It seems this utility is not all that generally useful,
so eliminate it from the global namespace and use
PodmanWithOptions directly.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Instaed, inline the implementation into callers, calling
PodmanWithOptions directly, demonstrating how to use
PodmanWithOptions.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
... and, to an extent, centralize the PodmanSessionIntegration
creation in that function.
This reduces duplication, and we will further eliminate
some of the callers.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Pass exactly the same PodmanExecOptions to makeOptions
and to PodmanExecBaseWithOptions. This will allow
simplifying the code further.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Eliminate this helper / indirection, and pass around
PodmanExecOptions explicitly.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
This will make it easier to structure the API, at the cost
of making it a bit more opaque about which parts of PodmanExecOptions
are implemented where.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
... replacing the many parameters with a struct with named fields.
This makes the meaning of parameters more explicit, and more importantly
it makes it easier to just edit _one_ of the parameters without requiring
specialized wrappers for every single case.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
This should be set only by podman as it is used for the podman generate
systemd --new command. For the api it was set to the system service
command which is simply pointless. It must be empty in these cases.
Fixes#25026
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
the podman artifact verb is used to manage OCI artifacts. the following
verbs were added to `podman artifact`:
* add
* inspect
* ls
* pull
* push
* rm
Notable items with this PR:
* all artifact commands and their output are subject to change. i.e.
consider all of this tech preview
* there is no way to add a file to an artifact that already exists in
the store. you would need to delete and recreate the artifact.
* all references to artifacts names should be fully qualified names in
the form of repo/name:tag (i.e. quay.io/artifact/foobar:latest)
* i understand that we will likely want to be able to attribute things
like arch, etc to artifact files. this function is not available yet.
Many thanks to Paul Holzinger for autocompletion PRs and review PRs that
fixed issues early on.
Also fix up some Args function to specify the correct number of args.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Signed-off-by: Brent Baude <bbaude@redhat.com>
Fixes: https://github.com/containers/podman/issues/25002
Also add the ability to inspect containers for
UseImageHosts and UseImageHostname.
Finally fixed some bugs in handling of --no-hosts for Pods,
which I descovered.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
This significantly simplifies the ceromony of running a Podman command
in integration tests, from
> session := p.Podman([]string{"stop", id})
> session.WaitWithDefaultTimeout()
> Expect(session).Should(ExitCleanly())
to
> p.PodmanExitCleanly("stop", id)
There are >4650 instances of ExitCleanly() in the tests,
and many could be migrated; this does not do that.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Introduce a new option "size" to configure the maximum size of the
user namespace configured by keep-id.
Closes: https://github.com/containers/podman/issues/24837
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
Passing the hostname allows netavark to include it in DHCP lease
requests which, in an environment where DDNS is used, can cause
DNS entries to be created automatically.
* The current Hostname() function in container.go was updated to
check the new `container_name_as_hostname` option in the
CONTAINERS table of containers.conf. If set and no hostname
was configured for the container, it causes the hostname to be
set to a version of the container's name with the characters not
valid for a hostname removed. If not set (the default), the original
behavior of setting the hostname to the short container ID is
preserved.
* Because the Hostname() function can return the host's hostname
if the container isn't running in a private UTS namespace, and we'd
NEVER want to send _that_ in a DHCP request for a container, a new
function NetworkHostname() was added which functions like Hostname()
except that it will return an empty string instead of the host's
hostname if the container is not running in a private UTS namespace.
* networking_common.getNetworkOptions() now uses NetworkHostname()
to set the ContainerHostname member of the NetworkOptions structure.
That member was added to the structure in a corresponding commit in
common/libnetwork/types/network.go.
* Added test to containers_conf_test.go
Signed-off-by: George Joseph <g.devel@wxy78.net>
When running this test on a system without unqualifiedsearch registries
it will fail with a different error causing the test to fail. to avoid
that case define our own registries.conf that defines quay.io as
registry. This should make the test pass in the debian env.
Signed-off-by: Paul Holzinger <git@holzinger.dev>
If the source dir is owned by another user then the test the chown will
fail assuming we run the tests rootless. This function is only used by
the quadlet tests and for the purpose all we need is to read the files
so the simple fix is remove the chown as this should make the tests pass
on the special debian gating env.
Fixes#24802
Signed-off-by: Paul Holzinger <git@holzinger.dev>
This test a pretty much useless, it checks that a connection attempt on
the default socket fails. But this is incorrect as the socket is outside
of the test control as such it might be ready to accept connections as
thus the test can fail locally or as reported here in the debian tests.
Given that a simple connection fails does not add any value I opted to
remove it.
Fixes#24803
Signed-off-by: Paul Holzinger <git@holzinger.dev>
Found in debian testing where by default there are no unqualified search
registries installed. As such the test failed as the FIXME said. Now
there is no need for the test to assume anything.
Instead set our own config via CONTAINERS_REGISTRIES_CONF then we can
do exact matches, except that env was not read in the shell completion
code so move some code around to make it read the var in the same way as
podman login/logout.
Signed-off-by: Paul Holzinger <git@holzinger.dev>
This commit simplifies the systemd parser logic, and it solves an
infinite loop when using a continuation line.
Closes: https://github.com/containers/podman/issues/24810
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
When the --index flag is used, `manifest annotate` shouldn't be
expecting a second non-flag argument.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
with defaults values when changes configuration with podman update.
The new LinuxResource structure does not represent the current unchanged configuration, which was not affected by the change.
Fixes: https://issues.redhat.com/browse/RUN-2375
Signed-off-by: Jan Rodák <hony.com@seznam.cz>
The command `podman info` returned only one imagestore in
`store.graphOptions.<driver>.imagestore` even if multiple
image stores were configured.
This change replaces the field `<driver>.imagestore` with
the field `<driver>.additionalImageStores`, that instead
of a string is an array of strings and that includes all
the configured additional image stores.
Fix https://github.com/containers/storage/issues/2094
Signed-off-by: Mario Loriedo <mario.loriedo@gmail.com>
This highlights a bug in common where the compression format is reset if the format is v2s2, even if its a valid compression format.
Signed-off-by: James Hewitt <james.hewitt@uk.ibm.com>
This solves several problems with copying into volumes on a
container that is not running.
The first, and most obvious, is that we were previously entirely
unable to copy into a volume that required mounting - like
image volumes, volume plugins, and volumes that specified mount
options.
The second is that this fixed several permissions and content
issues with a fresh volume and a container that has not been run
before. A copy-up will not have occurred, so permissions on the
volume root will not have been set and content will not have been
copied into the volume.
If the container is running, this is very low cost - we maintain
a mount counter for named volumes, so it's just an increment in
the DB if the volume actually needs mounting, and a no-op if it
doesn't.
Unfortunately, we also have to fix permissions, and that is
rather more complicated. This involves an ugly set of manual
edits to the volume state to ensure that the permissions fixes
actually worked, as the code was never meant to be used in this
way. It's really ugly, but necessary to reach full Docker
compatibility.
Fixes#24405
Signed-off-by: Matthew Heon <matthew.heon@pm.me>
* Add --hosts-file flag to container create, container run and pod create
* Add HostsFile field to pod inspect and container inspect results
* Test BaseHostsFile config in containers.conf
Signed-off-by: Gavin Lam <gavin.oss@tutamail.com>
New flags in a `podman update` can change the configuration of HealthCheck when the container is started, without having to restart or recreate the container.
This can help determine why a given container suddenly started failing HealthCheck without interfering with the services it provides. For example, reconfigure HealthCheck to keep logs longer than the usual last X results, store logs to other destinations, etc.
Fixes: https://issues.redhat.com/browse/RHEL-60561
Signed-off-by: Jan Rodák <hony.com@seznam.cz>
Test is failing on 1mt because of differences between 'stat'
command output and /proc/mounts. Solution: compare stat %t
(hex filesystem type), not %T (human-readable). This should
match no matter what kernel version or version of stat on
host/container.
Fixes: #24611
Signed-off-by: Ed Santiago <santiago@redhat.com>
This matches the behavior of other volume and mount types. Image
volumes and volumes/mounts from the `--volumes-from` flag should
be overridden by actual user-specified named volumes and mounts,
but this was not true for overlay mounts. Fortunately, our
duplicate-mount detection logic still works, so we got a good
error message at least.
The fix is simple - extend our supersede logic, which currently
only works with named volumes and mounts, to also work with
overlay mounts.
Fixes#24555
Signed-off-by: Matt Heon <mheon@redhat.com>
Paul Holzinger
Brent Baude
openshift-merge-bot[bot]
Giuseppe Scrivano
Ashley Cui
Jan Rodák
renovate[bot]
Lokesh Mandvekar
Miloslav Trmač
Daniel J Walsh
Florian Apolloner
Ahmed Moalla
Erik Sjölund
George Joseph
Odilon Sousa
Paul Holzinger
Misaki Kasumi
Nalin Dahyabhai
Mario Loriedo
James Hewitt
Matthew Heon
Gavin Lam
Ygal Blum
Ed Santiago
Matt Heon