Commit Graph

1571 Commits

Author SHA1 Message Date
Valentin Rothberg
f269be3a31 add {generate,play} kube
Add the `podman generate kube` and `podman play kube` command.  The code
has largely been copied from Podman v1 but restructured to not leak the
K8s core API into the (remote) client.

Both commands are added in the same commit to allow for enabling the
tests at the same time.

Move some exports from `cmd/podman/common` to the appropriate places in
the backend to avoid circular dependencies.

Move definitions of label annotations to `libpod/define` and set the
security-opt labels in the frontend to make kube tests pass.

Implement rest endpoints, bindings and the tunnel interface.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-05-06 17:08:22 +02:00
OpenShift Merge Robot
0eb905ff2c Merge pull request #6096 from mheon/fix_small_issues
Add small fixes for 'podman run' from diffing inspect
2020-05-06 11:27:05 +02:00
OpenShift Merge Robot
7885b5cd52 Merge pull request #6063 from QiWang19/manifest-annotate
manifest annotate
2020-05-06 03:41:09 +02:00
Matthew Heon
f7c3cfde77 Add small fixes for 'podman run' from diffing inspect
To try and identify differences between Podman v1.9 and master,
I ran a series of `podman run` commands with various flags
through each, then inspecting the resulting containers and diffed
the inspect JSON between each. This identified a number of issues
which are fixed in this PR.

In order of discovery:
- Podman v2 gave short names for images, where Podman v1 gave the
  fully-qualified name. Simple enough fix (get image tags and use
  the first one if they're available)
- The --restart flag was not being parsed correctly when a number
  of retries was specified. Parsing has been corrected.
- The -m flag was not setting the swap limit (simple fix to set
  swap in that case if it's not explicitly set by the user)
- The --cpus flag was completely nonfunctional (wired in its
  logic)

Tests have been added for all of these to catch future
regressions.

Signed-off-by: Matthew Heon <matthew.heon@pm.me>
2020-05-05 20:21:01 -04:00
Qi Wang
1090d4d5d9 manifest annotate
Signed-off-by: Qi Wang <qiwan@redhat.com>
2020-05-05 15:01:32 -04:00
OpenShift Merge Robot
864aec8bb0 Merge pull request #6081 from baude/v2system
v2 system subcommand
2020-05-05 20:49:17 +02:00
baude
25312bb5b4 v2 system subcommand
add system df, info, load, renumber, and migrate

Refactor for specialized libpod engines

add the ability to prune images, volumes, containers, and pods

Signed-off-by: baude <bbaude@redhat.com>
2020-05-05 12:24:33 -05:00
OpenShift Merge Robot
9db97dbab8 Merge pull request #6080 from baude/v2stats
v2 podman stats
2020-05-05 17:29:56 +02:00
OpenShift Merge Robot
e6235ef8f1 Merge pull request #6076 from vrothberg/rmi-v2.2
image removal: refactor part 2
2020-05-05 17:25:04 +02:00
baude
b5a235df90 v2 podman stats
Signed-off-by: baude <bbaude@redhat.com>
2020-05-05 08:46:51 -05:00
Matthew Heon
7ac3d906b5 Rework port parsing to support --expose and -P
As part of this, make a major change to the type we use to
represent port mappings in SpecGen (from using existing OCICNI
structs to using our own custom one). This struct has the
advantage of supporting ranges, massively reducing traffic over
the wire for Podman commands using them (for example, the
`podman run -p 5000-6000` command will now send only one struct
instead of 1000). This struct also allows us to easily validate
which ports are in use, and which are not, which is necessary for
--expose.

Once we have parsed the ports from the new struct, we can produce
an accurate map including all currently requested ports, and use
that to determine what ports need to be exposed (some requested
exposed ports may already be included in a mapping from --publish
and will be ignored) and what open ports on the host we can map
them to.

Signed-off-by: Matthew Heon <matthew.heon@pm.me>
2020-05-04 20:57:27 -04:00
Valentin Rothberg
7f97896c59 image removal: refactor part 2
Continue the refactoring of image removal.  I didn't manage to break all
the following changes into smaller and easier to digest commits due to
time constraints:

 * Return an error slice instead of a single error. Use multierror only
   in the client/frontend.  Reflect that in the types.

 * Use the batch image removal in the client while preserving the more
   rest-idiomatic single-image removal endpoint.

 * Add a new handler for the single-image removal endpoint to make it
   share the same code as the batch endpoint.

 * Expose bindings for the single and batch endpoints, so we can
   properly test them.

 * Add several convenience functions for error handling to
   pkg/errorhandling.

 * Set the correct error type in libpod to set the exit code to 2 when
   one or more containers are using an image.

 * Massage the bindings tests a bit and tackle compilation errors.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-05-04 16:01:45 +02:00
OpenShift Merge Robot
51d0be4204 Merge pull request #6051 from rhatdan/containers.conf
Fixes for test/e2e/containers_conf_test.go
2020-05-04 11:08:43 +02:00
OpenShift Merge Robot
0fd8242f91 Merge pull request #6058 from rhatdan/coverity
Fix errors found in coverity scan
2020-05-01 21:32:22 +02:00
OpenShift Merge Robot
bffd3f5134 Merge pull request #6060 from sujil02/systemprune-v2
And system prune feature for v2.
2020-05-01 21:08:50 +02:00
Daniel J Walsh
7d37f4bbfe Fix errors found in coverity scan
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-05-01 15:00:33 -04:00
Daniel J Walsh
97fcbfcbec cgroupsns was not following containers.conf
Implement ParseCgroupsNamespace to handle defaults.

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-05-01 15:00:33 -04:00
Daniel J Walsh
4a2765c498 Properly handle default capabilities listed in containers.conf
If user/admin specifies a different list of default capabilties
we need to honor these.

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-05-01 15:00:26 -04:00
Daniel J Walsh
730fbc7628 Properly handle containers.conf devices
We need to add the default devices listed in containers.conf

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-05-01 15:00:26 -04:00
OpenShift Merge Robot
95b9b72c0c Merge pull request #6062 from jwhonce/wip/docs
[CI:DOC] Bring README.md up to date
2020-05-01 20:59:05 +02:00
Jhon Honce
b22a39b6b1 [CI:DOCS] Bring README.md up to date
* Add notes on helper functions
* Update example

Signed-off-by: Jhon Honce <jhonce@redhat.com>
2020-05-01 11:22:39 -07:00
Sujil02
b94862171b And system prune feature for v2.
Adds podman system prune for v2.
Refactoring for code reuse from pods containers images and volume prune.
Adds and enables testcases to support the added feature.

Signed-off-by: Sujil02 <sushah@redhat.com>
2020-05-01 13:57:16 -04:00
Daniel J Walsh
8173e83054 Fix errors found in coverity scan
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-05-01 13:26:50 -04:00
OpenShift Merge Robot
226e0da6fe Merge pull request #6057 from baude/v2networking
v2networking enable commands
2020-05-01 18:27:16 +02:00
OpenShift Merge Robot
49107a5a2e Merge pull request #6004 from rhatdan/ulimits
Set up ulimits for rootless containers.
2020-05-01 15:58:24 +02:00
OpenShift Merge Robot
1230499e45 Merge pull request #6016 from giuseppe/fix-create
v2, podman: fix create and entrypoint tests
2020-05-01 15:32:00 +02:00
baude
e88a418528 v2networking enable commands
Enable the networking commands for v2.

Signed-off-by: baude <bbaude@redhat.com>
2020-04-30 14:53:54 -05:00
Giuseppe Scrivano
9d373408b5 podman, start: propagate back the raw input
this is necessary as we expect "podman start $ID_NAME" to print the
same arguments the user passed in instead of the full ID.

Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-30 12:23:19 +02:00
Giuseppe Scrivano
c11cff4542 cmd, podman: do not override entrypoint if unset
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-30 11:33:34 +02:00
Giuseppe Scrivano
65d7f22720 cmd, podman: handle --pod new:POD
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-30 11:33:33 +02:00
Qi Wang
75eeb40f41 testv2: enable attach test
testv2: enable attach test

Signed-off-by: Qi Wang <qiwan@redhat.com>
2020-04-29 23:47:15 -04:00
Jhon Honce
a9cc13448e V2 Restore images list tests
* Fix history --quiet formatting
* Fix image inspect --format=json
* Fix image list --sort

Signed-off-by: Jhon Honce <jhonce@redhat.com>
2020-04-29 08:24:56 -07:00
Valentin Rothberg
2d10471b8d pull/search options: tls verify -> skip
Change the logic in the options from tls-verify to skipping
verification.  It require a constant brain yoga to translate
from doing verification (CLI logic) to skipping it (c/image logic).
As the code is using c/image, let's make it consistent.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-04-29 16:28:12 +02:00
OpenShift Merge Robot
a76a99352a Merge pull request #6037 from vrothberg/enable-push-tests
Enable push tests
2020-04-29 16:20:44 +02:00
OpenShift Merge Robot
6246165f16 Merge pull request #6036 from giuseppe/fix-rootlessport-panic
rootlessport: use two different channels
2020-04-29 16:07:44 +02:00
Valentin Rothberg
5c108f75eb login system test: enable "push ok"
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-04-29 15:23:30 +02:00
OpenShift Merge Robot
a22dacfa1b Merge pull request #6035 from giuseppe/move-rootless-open-before-fork
rootless: move ns open before fork
2020-04-29 15:07:07 +02:00
Valentin Rothberg
d6d1e3860c push: fix --tls-verify
Fix --tls-verify parsing and make the associated options reflect the
correct logic.  Other commands are affected as well but will be fixed
later.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-04-29 14:33:00 +02:00
Giuseppe Scrivano
6d545bb2f7 rootlessport: use two different channels
The same channel is written to by two different goroutines.

Use a different channel for each of them so to avoid writing to a
closed channel.

Closes: https://github.com/containers/libpod/issues/6018

Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-29 12:35:23 +02:00
Giuseppe Scrivano
b8db112436 specgen: honor slirp4netns
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-29 12:13:05 +02:00
Giuseppe Scrivano
89d4940a37 rootless: move ns open before fork
commit 788fdc685b introduced a race
where the target process dies before the child process opens the
namespace files.  Move the open before the fork so if it fails the
parent process can attempt to join a different container instead of
failing.

Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-29 11:44:24 +02:00
OpenShift Merge Robot
3e912f711b Merge pull request #6022 from vrothberg/enable-inspect-tests
enable inspect tests
2020-04-29 11:21:47 +02:00
OpenShift Merge Robot
62a4befe04 Merge pull request #5998 from vrothberg/generate-systemd
generate systemd
2020-04-29 11:19:15 +02:00
OpenShift Merge Robot
4e21d09f98 Merge pull request #6029 from rhatdan/env
User specified environment happen after other environments are set
2020-04-29 08:52:14 +02:00
Valentin Rothberg
8700c2fd03 enable inspect tests
A surprisingly big change.  A core problem was that `podman inspect`
allows for passing containers AND images with the default `--type=all`.
This only worked partially as the data was processed in isolation which
caused various issues (e.g., two separate outputs instead of one) but it
also caused issues regarding error handling.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-04-29 07:25:31 +02:00
Valentin Rothberg
b2414b580e generate systemd
Implement `podman generate systemd` for Podman v2 and enable associated
tests.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
2020-04-29 07:22:18 +02:00
OpenShift Merge Robot
bf4efc1953 Merge pull request #6026 from baude/v2forcesystemtests
system tests must pass
2020-04-28 23:29:20 +02:00
OpenShift Merge Robot
30b27819d0 Merge pull request #5991 from sujil02/pod-rm-test
Fix typos in messages pod rm
2020-04-28 22:40:09 +02:00
Daniel J Walsh
e84beeaf94 User specified environment happen after other environments are set
When using varlink we want to make sure that user specified environment variables
take precedence over http-proxy environment.

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-04-28 16:29:01 -04:00
Brent Baude
517bc28360 system tests must pass
Signed-off-by: Brent Baude <bbaude@redhat.com>
2020-04-28 15:17:00 -05:00