We now have only a few entrypoints that are called externally,
so make the rest private. This will make it more obvious that
we are not breaking any external users.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
This will allow us to write unit tests without setting up the complete Podman runtime
(and without the Linux dependency).
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
NOTE: This does not edit the use-sigstore-attachments value
in registries.d, similarly to how (podman image trust set) didn't
set the lookaside paths for simple signing.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
- Also reject public keys with types that don't use them
- Reject unknown trust types
- And add unit tests
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
This will allow us to write unit tests without setting up the complete Podman runtime
(and without the Linux dependency).
Also, actually add a basic smoke test of the core functionality.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Only process the incoming args[] (which is a single-element array
for some reason) once, and use a semantic variable name for the value
we care about.
Should not change behavior, the only caller already supposedly ensures
that len(args) == 1.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Split the existing code into policy.go and registries.go,
depending on which files it concerns.
Only moves unchanged code, should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
We can always recover it from git, but it seems to serve
no purpose anyway.
Should not change behavior.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Also, do a general cleanup of all the timeout code. Changes
include:
- Convert from int to *uint where possible. Timeouts cannot be
negative, hence the uint change; and a timeout of 0 is valid,
so we need a new way to detect that the user set a timeout
(hence, pointer).
- Change name in the database to avoid conflicts between new data
type and old one. This will cause timeouts set with 4.2.0 to be
lost, but considering nobody is using the feature at present
(and the lack of validation means we could have invalid,
negative timeouts in the DB) this feels safe.
- Ensure volume plugin timeouts can only be used with volumes
created using a plugin. Timeouts on the local driver are
nonsensical.
- Remove the existing test, as it did not use a volume plugin.
Write a new test that does.
The actual plumbing of the containers.conf timeout in is one line
in volume_api.go; the remainder are the above-described cleanups.
Backported to v4.2.0-rhel per RHBZ 2125241
Signed-off-by: Matthew Heon <mheon@redhat.com>
Trying to print the image id on a failed inspect will result in a nil
pointer panic because the image will be nil. Replace image.id with the
image name which is defined as a string without the use of inspect.
Fixes: bz#2131836
[NO NEW TESTS NEEDED]
Signed-off-by: Brent Baude <bbaude@redhat.com>
For systems that have extreme robustness requirements (edge devices,
particularly those in difficult to access environments), it is important
that applications continue running in all circumstances. When the
application fails, Podman must restart it automatically to provide this
robustness. Otherwise, these devices may require customer IT to
physically gain access to restart, which can be prohibitively difficult.
Add a new `--on-failure` flag that supports four actions:
- **none**: Take no action.
- **kill**: Kill the container.
- **restart**: Restart the container. Do not combine the `restart`
action with the `--restart` flag. When running inside of
a systemd unit, consider using the `kill` or `stop`
action instead to make use of systemd's restart policy.
- **stop**: Stop the container.
To remain backwards compatible, **none** is the default action.
Backport of commit aad29e759c
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2097708
Signed-off-by: Valentin Rothberg <vrothberg@redhat.com>
when verifying that the memory was set correctly for a podman machine
instance, we check if the number is between a range because based on
architecture, operating system, and memory itself this number can differ
significantly.
Signed-off-by: Brent Baude <bbaude@redhat.com>
use the sandbox id instead of the name for the
io.kubernetes.cri-o.SandboxID annotation used by gVisor.
Closes: https://github.com/containers/podman/issues/15223
[NO NEW TESTS NEEDED] it is specific to gVisor
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
Tying filtering logic for podman stop and start to same place in getContainersAndInputByContext() to reduce code redundancy
Signed-off-by: Karthik Elango <kelango@redhat.com>
podman run/create can accept `-h <hostname>` as argument. When parsing
flags -h throws an help requested error from pflag. To prevent this
error we have to define the help flag.
Fixes#15124
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
When a container was created with `--sdnotify value` we would remove
this arg instead of using it like with `--sdnotfiy=value`.
Also when the arg is set to ignore we should force conmon in order to
make the resulting Type=notify units work.
Fixes#15052
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
This bug is reproduced when we execute the following command:
1. podman manifest add <manifest list> <images exist on local storage>
2. podman manifest push --rm <manifest list> dir:<directory>
If pushing succeeds, it is expected to remove only a manifest list.
However, manifest list remains on local storage and images are removed.
This commit fixes `podman manifest push --rm` to remove only a manifest list.
And, supports `manifest push --rm option` in remote environment,
like host environment.
Fixes: https://github.com/containers/podman/issues/15033
Signed-off-by: Toshiki Sonoda <sonoda.toshiki@fujitsu.com>
When the break out or the WSL environment fails to start, the config
and connections should not be written. Placing them at the end of the
provisioning step will mitigate the issue.
[NO NEW TESTS NEEDED]
Signed-off-by: Gerard Braad <me@gbraad.nl>
- Allow creating sigstore signatures via --sign-by-sigstore-private-key .
Like existing --sign-by, it does not work remote (in this case
because we would have to copy the private key to the server).
- Allow passing a passphrase (which is mandatory for sigstore private keys)
via --sign-passphrase-file; if it is not provided, prompt interactively.
- Also, use that passphrase for --sign-by as well, allowing non-interactive
GPG use. (But --sign-passphrase-file can only be used with _one of_
--sign-by and --sign-by-sigstore-private-key.)
Note that unlike the existing code, (podman build) does not yet
implement sigstore (I'm not sure why it needs to, it seems not to
push images?) because Buildah does not expose the feature yet.
Also, (podman image sign) was not extended to support sigstore.
The test for this follows existing (podman image sign) tests
and doesn't work rootless; that could be improved by exposing
a registries.d override option.
The test for push is getting large; I didn't want to
start yet another registry container, but that would be an
alternative. In the future, Ginkgo's Ordered/BeforeAll
would allow starting a registry once and using it for two
tests.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
Before making / mutable/immutable, podman-machine checks if the mount is
being done in /home or /mnt. However the current check is always going
to be true:
```
!strings.HasPrefix(mount.Target, "/home") || !strings.HasPrefix(mount.Target, "/mnt")
```
is false when mount.Target starts with "/home" and mount.Target starts
with "/mnt", which cannot happen at the same time.
The correct check is:
```
!strings.HasPrefix(mount.Target, "/home") && !strings.HasPrefix(mount.Target, "/mnt")
```
which can also be written as:
```
!(strings.HasPrefix(mount.Target, "/home") || strings.HasPrefix(mount.Target, "/mnt"))
```
The impact is not too bad, it results in extra 'chattr -i' calls which
should be unneeded.
[NO NEW TESTS NEEDED]
Signed-off-by: Christophe Fergeau <cfergeau@redhat.com>
Option defaults in API must be the same as in CLI.
```
% podman image push --help
% podman image pull --help
% podman manifest push --help
% podman image search --help
```
All of these CLI commands them have --tls-verify=true by default:
```
--tls-verify require HTTPS and verify certificates when accessing the registry (default true)
```
As for `podman image build`, it doesn't have any means to control
`tlsVerify` parameter but it must be true by default.
Signed-off-by: Vladimir Kochnev <hashtable@yandex.ru>
Some refer to issues that are closed. Remove them.
Some are runc bugs that will never be fixed. Say so, and remove
the FIXME.
One (bps/iops) should probably be fixed. File an issue for it, and
update comment to include the issue# so my find-obsolete-skips script
can track it.
And one (rootless mount with a "kernel bug?" comment) is still
not fixed. Leave the skip, but add a comment documenting the symptom.
Signed-off-by: Ed Santiago <santiago@redhat.com>
Fix some network option parsing logic to use constants.
Always use the isolate option since this is what docker does.
Remove the icc option, this is different from isolate and it is not
implemented.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Docker uses "bridge" as default network name so some tools expect this
to work with network list or inspect. To fix this we change "bridge" to
the podman default ("podman") name.
Fixes#14983
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Progress bar in JSONMessage is missing compared to docker output both in
pull and push. Additionaly, pull was not using JSONMessage while push
was using the type.
[NO NEW TESTS NEEDED]
Signed-off-by: Jakub Guzik <jguzik@redhat.com>
`podman-remote push` has shown absolutely no progress at all. Fix that
by doing essentially the same as the remote-pull code does.
The get-free-out-of-jail-card for backwards compatibility is to let the
`quiet` parameter default to true. Since the --quioet flag wasn't
working before either, older Podman clients do not set it.
Also add regression tests to make sure we won't regress again.
Fixes: #11554Fixes: #14971
Signed-off-by: Valentin Rothberg <vrothberg@redhat.com>
The memory both local and in the CI test is converted to 3822. I don't
know why this changed but I want to have this working again. For the
future we should look at a more robust solution.
Fixes#15012
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The same problem again as 4374038cc6.
Also fix the incorrect --format autocompletion struct.
It should be avoided to import cmd/podman/... packages from outside of
cmd/podman. This can lead in weird hard to debug import paths but also
can have negative consequences when imported in unit tests. In this case
it will set XDG_CONFIG_HOME and thus the machine tests this dir over the
tmp HOME env variable which is set at a later point. This caused machine
files to be leaked into the actual users home dir.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Rename all files to _test.go and rename the package to e2e_test. This
makes the linter less strict about things like dot imports.
Add some unused nolint directives to silence some warnings, these can be
used to find untested options so someone could add tests for them.
Fixes#14996
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
PR https://github.com/containers/common/pull/1071 moved `pkg/hooks` to
`c/common` hence remove that from podman and use `pkg/hooks` from
`c/common`
[NO NEW TESTS NEEDED]
[NO TESTS NEEDED]
Signed-off-by: Aditya R <arajan@redhat.com>
Update the init container type default to once instead
of always to match k8s behavior.
Add a new annotation that can be used to change the init
ctr type in the kube yaml.
Signed-off-by: Urvashi Mohnani <umohnani@redhat.com>
There is a possibility that podman does not output expected error message.
(e.g. When pause rootless cgroups v1 container on host)
This problem is solved by appending `report.Errs` to `reports` before `continue`.
[NO NEW TESTS NEEDED]
Signed-off-by: Toshiki Sonoda <sonoda.toshiki@fujitsu.com>
- Also save the file using this convention.
- Change the general pull mechanism to print the local file
as opposed to the remote to enable this - no change in
observed behavior on mac
Signed-off-by: Jason T. Greene <jason.greene@redhat.com>
When initing machines, we download a machine image, and uncompress and
copy the image for the actual vm image. When a user constantly pulls new
machines, there may be a buildup of old, unused machine images. This
commit cleans ups the unused cached images.
Changes:
- If the machine is pulled from a URL or from the FCOS releases, we pull
them into XDG_DATA_HOME/containers/podman/machine/vmType/cache
- Cache cleanups only happen if there is a cache miss, and we need to
pull a new image
- For Fedora and FCOS, we actually use the cache, so we go through the
cache dir and remove any images older than 2 weeks (FCOS's release cycle), on a cache miss.
- For generic files pulled from a URL, we don't actually cache, so we
delete the pulled file immediately after creating a machine image
- For generic files from a local path, the original file will never be
cleaned up
Note that because we cache in a different dir, this will not clean up
old images pulled before this commit.
[NO NEW TESTS NEEDED]
Signed-off-by: Ashley Cui <acui@redhat.com>
