CLI reference for the base command was generated to cli.md
Changed it to docker.md to handle broken links.
Signed-off-by: David Karlsson <35727626+dvdksn@users.noreply.github.com>
Alias pages redirect to the canonical names, but these pages still
linked to the aliases, causing broken links when building the docs site.
Signed-off-by: David Karlsson <35727626+dvdksn@users.noreply.github.com>
- Add section about cdi-spec-dirs daemon configuration
- Add subsection about cdi in --device section for docker run
- Update `docker info` output example
Signed-off-by: David Karlsson <35727626+dvdksn@users.noreply.github.com>
The Docker daemon provides a versioned API for backward compatibility with old
clients. Docker clients can perform API-version negotiation to select the most
recent API version supported by the daemon (downgrading to and older version of
the API when necessary). API version negotiation was introduced in Docker v1.12.0
(API 1.24), and clients before that used a fixed API version.
Docker Engine versions through v25.0 provide support for all [API versions](https://docs.docker.com/engine/api/#api-version-matrix)
included in stable releases for a given platform. For Docker daemons on Linux,
the earliest supported API version is 1.12 (corresponding with Docker Engine
v1.0.0), whereas for Docker daemons on Windows, the earliest supported API
version is 1.24 (corresponding with Docker Engine v1.12.0).
Support for legacy API versions (providing old API versions on current versions
of the Docker Engine) is primarily intended to provide compatibility with recent,
but still supported versions of the client, which is a common scenario (the Docker
daemon may be updated to the latest release, but not all clients may be up-to-date
or vice versa). Support for API versions before that (API versions provided by
EOL versions of the Docker Daemon) is provided on a "best effort" basis.
Use of old API versions is very rare, and support for legacy API versions
involves significant complexity (Docker 1.0.0 having been released 10 years ago).
Because of this, we'll start deprecating support for legacy API versions.
Docker Engine v25.0 by default disables API version older than 1.24 (aligning
the minimum supported API version between Linux and Windows daemons). When
connecting with a client that uses an API version version older than 1.24,
the daemon returns an error. The following example configures the docker
CLI to use API version 1.23, which produces an error:
DOCKER_API_VERSION=1.23 docker version
Error response from daemon: client version 1.23 is too old. Minimum supported API version is 1.24, please upgrade your client to a newer version
An environment variable (`DOCKER_MIN_API_VERSION`) is introduced that allows
re-enabling older API versions in the daemon. This environment variable must
be set in the daemon's environment (for example, through a [systemd override
file](https://docs.docker.com/config/daemon/systemd/)), and the specified
API version must be supported by the daemon (`1.12` or higher on Linux, or
`1.24` or higher on Windows).
Support for API versions lower than `1.24` will be permanently removed in Docker
Engine v26, and the minimum supported API version will be incrementally raised
in releases following that.
We do not recommend depending on the `DOCKER_MIN_API_VERSION` environment
variable other than for exceptional cases where it's not possible to update
old clients, and those clients must be supported.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
In order to solve the "double notification" issue (see:
ef5e5fa03f)
without running the plugin process under a new pgid (see:
https://github.com/moby/moby/issues/47073) we instead check if we're
attached to a TTY, and if so skip signalling the plugin process since it
will already be signalled.
Signed-off-by: Laura Brehm <laurabrehm@hey.com>
This reverts commit ef5e5fa03f.
Running new plugins under a new pgid isn't a viable solution due to
it causing issues with plugin processes attempting to read from the
TTY (see: https://github.com/moby/moby/issues/47073).
Signed-off-by: Laura Brehm <laurabrehm@hey.com>
Build and push an image containing a static CLI binary for master branch
and every release branch and tag.
This is a slightly adjusted copy of the bin-image workflow from
docker/buildx (by @crazy-max).
Co-authored-by: CrazyMax <crazy-max@users.noreply.github.com>
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Looks like this change caused docker-ce-packaging to fail;
dest=$PWD/build/mac; cd /home/ubuntu/workspace/release-packaging_ce-nightly/packaging/src/github.com/docker/cli/build && for platform in *; do \
arch=$(echo $platform | cut -d_ -f2); \
mkdir -p $dest/$arch/docker; \
cp $platform/docker-darwin-* $dest/$arch/docker/docker && \
tar -C $dest/$arch -c -z -f $dest/$arch/docker-25.0.0-rc.2.tgz docker; \
done
cp: cannot stat 'darwin_amd64/docker-darwin-*': No such file or directory
cp: cannot stat 'darwin_arm64/docker-darwin-*': No such file or directory
This reverts commit 6ad07f2a4b.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Changes were made in 1554ac3b5f to provide
a mechanism for the CLI to notify running plugin processes that they
should exit, in order to improve the general CLI/plugin UX. The current
implementation boils down to:
1. The CLI creates a socket
2. The CLI executes the plugin
3. The plugin connects to the socket
4. (When) the CLI receives a termination signal, it uses the socket to
notify the plugin that it should exit
5. The plugin's gets notified via the socket, and cancels it's `cmd.Context`,
which then gets handled appropriately
This change works in most cases and fixes the issue it sets out to solve
(see: https://github.com/docker/compose/pull/11292) however, in the case
where the user has a TTY attached and the plugin is not already handling
received signals, steps 4+ changes:
4. (When) the CLI receives a termination signal, before it can use the
socket to notify the plugin that it should exit, the plugin process
also receives a signal due to sharing the pgid with the CLI
Since we now have a proper "job control" mechanism, we can simplify the
scenarios by executing the plugins with their own process group id,
thereby removing the "double notification" issue and making it so that
plugins can handle the same whether attached to a TTY or not.
In order to make this change "plugin-binary" backwards-compatible, in
the case that a plugin does not connect to the socket, the CLI passes
the signal to the plugin process.
Co-authored-by: Bjorn Neergaard <bjorn.neergaard@docker.com>
Signed-off-by: Laura Brehm <laurabrehm@hey.com>
Signed-off-by: Bjorn Neergaard <bjorn.neergaard@docker.com>
As macOS does not support the abstract socket namespace, use a temporary
socket in $TMPDIR to connect with the plugin. Ensure this socket is
cleaned up even in the case of crash/ungraceful termination by removing
it after the first connection is accepted.
Co-authored-by: Laura Brehm <laurabrehm@hey.com>
Signed-off-by: Bjorn Neergaard <bjorn.neergaard@docker.com>
Sebastiaan van Stijn
Sebastiaan van Stijn
Tonis Tiigi
David Karlsson
Laura Brehm
Paweł Gronowski
Bjorn Neergaard