Commit Graph

2399 Commits

Author SHA1 Message Date
OpenShift Merge Robot 7984842d7e
Merge pull request #8556 from mheon/fix_8539
Use Libpod tmpdir for pause path
2020-12-02 22:03:19 +01:00
OpenShift Merge Robot e74072e742
Merge pull request #8112 from QiWang19/load-optional-name
Drop name argument from Load API
2020-12-02 21:35:31 +01:00
Matthew Heon ab88632835 Use Libpod tmpdir for pause path
Previously, we always computed pause path from the Rootless
runtime directory. Problem: this does not match the behavior of
Libpod when the directory changes. Libpod will continue to use
the previous directory, cached in the database; Pause pidfiles
will swap to the new path. This is problematic when the directory
needs to exist to write the pidfile, and Libpod is what creates
the directory.

There are two potential solutions - allow the pause pidfile to
move and just make the directory when we want to write it, or use
the cached Libpod paths for a guaranteed location. This patch
does the second, because it seems safer - we will never miss a
previously-existing pidfile because the location is now
consistent.

Fixes #8539

Signed-off-by: Matthew Heon <mheon@redhat.com>
2020-12-02 14:18:37 -05:00
OpenShift Merge Robot e82ec90007
Merge pull request #8552 from baude/buildtarget
target is not tag
2020-12-02 16:58:28 +01:00
Qi Wang f525d8b843 Do not pass name argument to Load API
Not pass the name argument to Load API. Specify in the document the usage of the optional argument is tagging an additional image.
Close #7337

Signed-off-by: Qi Wang <qiwan@redhat.com>
2020-12-02 09:25:01 -05:00
baude 3a5cd57bb4 target is not tag
remove mistaken use of target being used for tag

Signed-off-by: baude <bbaude@redhat.com>
2020-12-02 08:21:03 -06:00
Paul Holzinger c31a5c0d9c Add support for network ids
The network ID is not stored. It is just the sha256 hash from
the network name. There is a risk of a potential hash collision.
However it's very unlikely and even if we hit this it will
complain that more than network with this ID exists.

The main benefit is that the compat api can have proper
network ID support. Also this adds the support for
`podman network ls --format "{{.ID}}"` and `--filter id=<ID>`.

It also ensures that we can do network rm <ID> and network
inspect <ID>.

Since we use a hash this commit is backwards compatible even for
already existing networks.

Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
2020-12-02 10:27:00 +01:00
OpenShift Merge Robot 9c5fe954cc
Merge pull request #8543 from mheon/no_syslog_true
Do not use "true" after "syslog" in exit commands
2020-12-02 01:03:40 +01:00
Anders F Björklund b1b35707aa Add podman network create option for bridge mtu
Signed-off-by: Anders F Björklund <anders.f.bjorklund@gmail.com>
2020-12-01 22:33:03 +01:00
OpenShift Merge Robot c71ad9a557
Merge pull request #8541 from rhatdan/man
Fix potential race condition in testing
2020-12-01 22:09:38 +01:00
OpenShift Merge Robot b7ff6f0912
Merge pull request #8505 from Luap99/network-labels
podman network label support
2020-12-01 21:43:27 +01:00
Matthew Heon bf6b228092 Do not use "true" after "syslog" in exit commands
Instead of being interpreted as an argument to the boolean flag,
the 'true' is being intepreted as the Podman command to be run -
so we're trying to run `podman true`, which does not exist. This
causes the cleanup command to fail when `--log-level=debug` is
set, so containers are not cleaned up or removed.

This problem is easily reproduced with any command combining the
`--rm`, `-d`, and `--log-level=debug` flags - the command will
execute and exit, but the container will not be removed.

Separate, but worth looking into later: the errors we get on
trying `podman true` with any flags are terrible - if you just
type `podman true` you get a quite sane "Unrecognized command"
error, but if you try `podman true --rm` you get an "unknown flag
--rm" error - which makes very little sense given the command
itself doesn't exist.

Signed-off-by: Matthew Heon <mheon@redhat.com>
2020-12-01 15:12:50 -05:00
Daniel J Walsh c734c13904
Fix potential race condition in testing
The It("podman wait to pause|unpause condition"... test is
flaking every so often when a messages is sent in the second
function to a channel.  It is my believe that in between the time
the first function sends a message to the channel and before it closes
the channel the second errChan=make() has happened.  This would mean that
the fist function closes the second errChan, and then when the second
function sends a message to the second errChan, it fails and blows up with
the error you are seeing.

By creating a different variable for the second channel, we eliminate the race.

Fixes: https://github.com/containers/podman/issues/6518

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-12-01 14:11:04 -05:00
OpenShift Merge Robot ce45b71dcf
Merge pull request #8126 from matejvasek/impl-apiv2-archive
Implement containers/{id or name}/archive api
2020-12-01 19:56:49 +01:00
OpenShift Merge Robot e3f0b7db75
Merge pull request #8400 from rhatdan/varlink
Remove varlink support from podman
2020-12-01 15:30:06 +01:00
OpenShift Merge Robot 24383906f8
Merge pull request #8427 from rhatdan/ps
Handle ps container created field as a time.Time
2020-12-01 14:40:18 +01:00
OpenShift Merge Robot e2c406f199
Merge pull request #8465 from rhatdan/pull
Document docker transport is the only supported remote transport
2020-11-30 19:52:04 +01:00
OpenShift Merge Robot 1613921ecc
Merge pull request #8514 from Luap99/revert-8410-fix-multiple-networks
Revert "Allow multiple --network flags for podman run/create"
2020-11-30 19:43:19 +01:00
OpenShift Merge Robot f24812ac3b
Merge pull request #8230 from mheon/port_net_host_conflict
Ensure that --net=host/pod/container conflicts with -p
2020-11-30 19:41:26 +01:00
Luap99 f3402c73d4 Revert "Allow multiple --network flags for podman run/create"
As described in issue #8507 this commit contains a breaking
change which is not wanted in v2.2.

We can discuss later if we want this in 3.0 or not.

Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
2020-11-30 18:10:26 +01:00
jortkoopmans 84e8b2afa7
Merge branch 'master' into patch-1 2020-11-30 15:58:23 +01:00
Daniel J Walsh 9ba52e8ef0
Document docker transport is the only supported remote transport
The goal is to improve errors when users use the wrong transport
in certain cases we stutter, in other cases we don't give enough
information.

Remove stutters when failing to pull remote images, because of
lack of support.

Fix errors returned by reference.Parse to wrap in image that was being
checked.

Fixes: https://github.com/containers/podman/issues/7116

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-11-29 06:28:39 -05:00
Paul Holzinger 8494bcb866 podman network label support
Add label support for podman network create. Use the `args`
field in the cni config file to store the podman labels.
Use `podman_labels` as key name and store the labels as
map[string]string.

For reference: https://github.com/containernetworking/cni/blob/master/CONVENTIONS.md#args-in-network-config
https://github.com/containernetworking/cni/blob/spec-v0.4.0/SPEC.md#network-configuration

Example snippet:

```
...
"args": {
	"podman_labels": {
		"key1":"value1",
		"key2":"value2"
	}
}
...
```

Make podman network list support several filters. Supported filters are name,
plugin, driver and label. Filters with different keys work exclusive. Several label
filters work exclusive and the other filter keys are working inclusive.

Also adjust the compat api to support labels in network create and list.

Breaking changes:

- podman network ls -f shortform is used for --filter instead --format
This matches docker and other podman commands (container ps, volume ps)

- libpod network list endpoint filter parameter is removed. Instead the
filters paramter should be used as json encoded map[string][]string.

Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
2020-11-28 18:35:43 +01:00
Alban Bedel 66944baad6 Add support for persistent volume claims in kube files
In k8s a persistent volume claim (PVC) allow pods to define a volume
by referencing the name of a PVC. The PVC basically contains criterias
that k8s then use to select which storage source it will use for the
volume.

Podman only provide one abtracted storage, the named volumes, and
create them if they don't exists yet. So this patch simply use a
volume with the name of the PVC.

Signed-off-by: Alban Bedel <albeu@free.fr>
2020-11-27 11:38:33 +01:00
Alban Bedel b84304da5e Prepare support in kube play for other volume types than hostPath
Replace the simple map of names to paths with a map of names to a struct
to allow passing more parameters. Also move the code to parse the volumes
to its own file to avoid making the playKubePod() function overly complex.
Finally rework the kube volumes test to also be ready to support more
volume types.

Signed-off-by: Alban Bedel <albeu@free.fr>
2020-11-27 11:38:33 +01:00
Daniel J Walsh f62a356515
Remove varlink support from Podman
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-11-26 16:50:42 -05:00
OpenShift Merge Robot ad2439264d
Merge pull request #8487 from riyad/fix-ping-typo
REST API v2 - ping - fix typo in header
2020-11-26 16:41:53 +01:00
OpenShift Merge Robot 201b0acd02
Merge pull request #8486 from riyad/improve-ping-api-compat
REST API v2 - ping - remove newline from response to improve Docker compatibility
2020-11-26 16:40:11 +01:00
Paul Holzinger f7f85fef7c Do not validate the volume source path in specgen
The volume src path should not be validated in specgen since
the remote client also uses that part and the path must only
exists on the server. This now fails later and only on the
server and not the client.

I don't think I can add a test for this because the CI runs
server and client always on the same vm.

Fixes #8473

Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
2020-11-26 12:55:33 +01:00
Riyad Preukschas 0ae1221a46 REST API v2 - ping - fix typo in header
Signed-off-by: Riyad Preukschas <riyad@informatik.uni-bremen.de>
2020-11-25 20:49:08 +01:00
Riyad Preukschas af6106f3ff REST API v2 - ping - remove newline from response to improve Docker compatibility
Signed-off-by: Riyad Preukschas <riyad@informatik.uni-bremen.de>
2020-11-25 20:43:36 +01:00
jortkoopmans 5cfbe0b78e squash
Signed-off-by: jortkoopmans <jort@jabo-solutions.eu>
2020-11-25 18:27:31 +01:00
Jhon Honce 44da01f45c Refactor compat container create endpoint
* Make endpoint compatibile with docker-py network expectations
* Update specgen helper when called from compat endpoint
* Update godoc on types
* Add test for network/container create using docker-py method
* Add syslog logging when DEBUG=1 for tests

Fixes #8361

Signed-off-by: Jhon Honce <jhonce@redhat.com>
2020-11-23 15:20:39 -06:00
OpenShift Merge Robot 5d55285188
Merge pull request #8409 from mlegenovic/master
Docker compat API fixes
2020-11-23 21:17:05 +01:00
Milivoje Legenovic 6f7b7060e8 [WIP] Docker compat API fixes
These are the first fixes that are needed for development environments like
Eclipse or IntelliJ that have Docker plug-ins and use the Docker API to speak
with container engine (#7857)

Signed-off-by: Milivoje Legenovic <m.legenovic@gmail.com>
2020-11-23 17:45:24 +01:00
OpenShift Merge Robot dd343418ce
Merge pull request #8263 from rhatdan/restart
Allow containers to --restart on-failure with --rm
2020-11-23 13:44:37 +01:00
OpenShift Merge Robot 5292d5a7b8
Merge pull request #8429 from psakar/fix-json-capabilities
APIv2 - strip CAP_ prefix from capabilities in inspect container json
2020-11-21 10:32:28 +01:00
OpenShift Merge Robot b30ca110ec
Merge pull request #8410 from Luap99/fix-multiple-networks
Allow multiple --network flags for podman run/create
2020-11-21 02:11:22 +01:00
OpenShift Merge Robot 41c88a6608
Merge pull request #8423 from psakar/fix-8418
REST API v2 - list of images - mandatory Created attribute
2020-11-20 22:05:50 +01:00
Petr Sakař 435f61f497 APIv2 - strip CAP_ prefix from capabilities in json
strip prefix "CAP_" from capabilities in json generated by container inspect operation

Signed-off-by: Petr Sakař <petr.sakar@chare.eu>
2020-11-20 21:46:51 +01:00
Daniel J Walsh dc8996ec84
Allow containers to --restart on-failure with --rm
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-11-20 13:55:19 -05:00
Petr Sakař e24e6b6372 REST API v2 - list of images - mandatory Created attribute
fixes https://github.com/containers/podman/issues/8418

created attribute is missing for images with created datetime set to 0 Unix time (January 1, 1970 UTC) because 0 is considered as default value, and thus attribute was ommited because of 'omitempty' flag

Signed-off-by: Petr Sakař <petr.sakar@chare.eu>
2020-11-20 19:21:03 +01:00
OpenShift Merge Robot 864fe21ed0
Merge pull request #8406 from jwhonce/issues/8390
Make podman service log events
2020-11-20 18:10:56 +01:00
Paul Holzinger f441190d10 Allow multiple --network flags for podman run/create
We allow a container to be connected to several cni networks
but only if they are listed comma sperated. This is not intuitive
for users especially since the flag parsing allows multiple string
flags but only would take the last value. see: spf13/pflag#72

Also get rid of the extra parsing logic for pods. The invalid options
are already handled by `pkg/specgen`.

A test is added to prevent a future regression.

Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
2020-11-20 17:02:48 +01:00
Daniel J Walsh 2d861ac14a
Handle ps container created field as a time.Time
In the current code we were translating the created time
from a time.Time to a unix epoch, this was leading to a loss
of precession, and some unexpected results where the sorting
order of containers was misordered because of the precession loss.

If we pass around created as time.Time, we do not loose the precission.

Fixes: https://github.com/containers/podman/issues/8414

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-11-20 10:50:30 -05:00
Jhon Honce d8795a36b2 Make podman service log events
* Log endpoint calls at level Info
* Ensure API server started at level Info

Fixes #8390

Signed-off-by: Jhon Honce <jhonce@redhat.com>
2020-11-19 14:42:56 -07:00
Daniel J Walsh 9770947818
Document containers.conf settings for remote connections
Currently we don't document which end of the podman-remote client server
operations uses the containers.conf.  This PR begins documenting this
and then testing to make sure the defaults follow the rules.

Fixes: https://github.com/containers/podman/issues/7657

Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
2020-11-19 14:48:10 -05:00
Matej Vasek 4d0346c028 not forcing unmount
Signed-off-by: Matej Vasek <mvasek@redhat.com>
2020-11-19 20:31:47 +01:00
Matej Vasek be7e9f63f2 add comment
Signed-off-by: Matej Vasek <mvasek@redhat.com>
2020-11-19 20:31:47 +01:00
Matej Vasek 6ad2f1d248 fix: unmount container without force
Signed-off-by: Matej Vasek <mvasek@redhat.com>
2020-11-19 20:31:47 +01:00