These were originally added in 6f61cf053a,
but at the time, the error wasn't returned. Now that it is, we shouldn't
log _and_ return the error.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Adds support for setting memory swap settings on Swarm services
* Adds flags `memory-swap` and `memory-swappiness` to `docker service
create` and `docker service update` commands.
* Adds compose fields `memswap_limit` and `mem_swappiness` for `docker
stack` commands.
Signed-off-by: Drew Erny <derny@mirantis.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Prevent the linter from recursing to other modules (cmd/docker-trust),
which don't have their dependencies vendored.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Create a copy of the registry package to use, so that code used only
for trust can be removed from the cli/internal package.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
move the `trust` subcommands to a plugin, so that the subcommands can
be installed separate from the `docker trust` integration in push/pull
(for situations where trust verification happens on the daemon side).
make binary
go build -o /usr/libexec/docker/cli-plugins/docker-trust ./cmd/docker-trust
docker info
Client:
Version: 28.2.0-dev
Context: default
Debug Mode: false
Plugins:
buildx: Docker Buildx (Docker Inc.)
Version: v0.24.0
Path: /usr/libexec/docker/cli-plugins/docker-buildx
trust: Manage trust on Docker images (Docker Inc.)
Version: unknown-version
Path: /usr/libexec/docker/cli-plugins/docker-trust
docker trust --help
Usage: docker trust [OPTIONS] COMMAND
Extended build capabilities with BuildKit
Options:
-D, --debug Enable debug logging
Management Commands:
key Manage keys for signing Docker images
signer Manage entities who can sign Docker images
Commands:
inspect Return low-level information about keys and signatures
revoke Remove trust for an image
sign Sign an image
Run 'docker trust COMMAND --help' for more information on a command.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- term: remove duplicate flag and add comment on windows
- term: allow multi-line bracketed paste to not create single line
with verbatim LFs (fixes "x/term: multi line bracketed paste fails"
to issue line by line commands
full diff: https://github.com/golang/term/compare/v0.32.0...v0.36.0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This code was using swarmkit's genericresource package as intermediate;
add a local copy of that code that skips the protobufs as intermediate.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Various dependencies, including "golang.org/x/.." started to update
the minimum required version,so we should follow suit for the next
release.
Note that the `//go:build` directives not necesserily have to be
updated, but it's good to keep them in sync until we have a go.mod
to control this.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Add WS2025 to Windows matcher and code optimizations
- use windowsMatchComparer for OSVersion match order
Windows OS version should match based on the full OSVersion. When
sorting a manifest, the entries should be sorted using the `Less`
function.
full diff: https://github.com/containerd/platforms/compare/v1.0.0-rc.1...v1.0.0-rc.2
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This was introduced in dad1d367c8, which
did not add a `//go:build` constraint to enable the use of generics (`any`).
Which causes an error when used;
could not import github.com/docker/cli/cli/command/container (-: # github.com/docker/cli/cli/command/container
/Users/thajeztah/go/pkg/mod/github.com/docker/cli@v29.0.0-rc.2+incompatible/cli/command/container/stats.go:148:39: predeclared any requires go1.18 or later (-lang was set to go1.16; check go.mod)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The allManagementSubCommands function is used to present plugin-commands
in the docker --help output; these commands are included in the "management
commands" section, but for plugins we don't know if they have sub-commands.
However, plugin stubs may be hidden (for placeholders that are not yet loaded),
or not be runnable, which was previously ignored.
This patch treats plugin-stubs the same as other commands, with the exception
of checking if they have subcommands (which is not yet known for plugin-stubs).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Use the new tree view by default and only fallback if format or old
view-related options are used.
The expanded view is shown when `--tree` is passed.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The logic for applying the dangling filter when `--all` is not used was
inverted. The filter was being applied when the dangling filter was
present, but it should be applied when the dangling filter is NOT
present.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Sort images alphabetically by their repository tags rather than by
creation date.
When an image has multiple tags, they are sorted internally and the
first tag is used as the representative for sorting the image in the
list. Untagged images are placed at the end.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Use separate branches to handle windows/unix results to reduce intermediate
variables, and make it more transparent what's set for each platform.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
construct the decoder inside the go-routine, including closing the body,
and add handling for context-cancellation.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
We should consider splitting this out to a separate function, but
start with just an early return before we hit the timer-loop.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The JSON field was added in [moby@9fd2c0f], to address [moby#19177], which
reported an incompatibility with Classic (V1) Swarm, which produced a non-
standard response;
> Make docker load to output json when the response content type is json
> Swarm hijacks the response from docker load and returns JSON rather
> than plain text like the Engine does. This makes the API library to return
> information to figure that out.
A later change in [moby@96d7db6] added additional logic to make sure the
correct content-type was returned, depending on whether the `quiet` option
was set (which produced a non-JSON response). This caused inconsistency in
the API response, and [moby@2f27632] changed the endpoint to always produce
JSON (only skipping the "progress" output if `quiet` was set).
This means that the "load" endpoint ([`imageRouter.postImagesLoad`]) now
unconditionally returns JSON, making the `JSON` field fully redundant.
This patch removes the use of the JSON field, as it's redundant, and the way it handles
the content-type is incorrect because it would not handle correct, but different
formatted response-headers (`application/json; charset=utf-8`), which could
result in malformed output on the client.
[moby@9fd2c0f]: 9fd2c0feb0
[moby#19177]: https://github.com/moby/moby/issues/19177
[moby@96d7db6]: 96d7db665b
[moby@2f27632]: 2f27632cde
[`imageRouter.postImagesLoad`]: 7b9d2ef6e5/api/server/router/image/image_routes.go (L248-L255)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This field is set for the request as a whole, so can be obtained
from the server-info instead. Docker v29 will provide per-stats
information.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This change prevents changes to the otel version from affecting the otel
sdk version. This is done by copying the telemetry sdk implementation
locally and using our own choice for semconv from within that.
This prevents a schema conflict from happening since the otel version of
the sdk gets implicitly updated whenever the semconv changes while we
have to manually change ours. Now, we manually change both and they're
locked to each other.
Signed-off-by: Jonathan A. Sternberg <jonathan.sternberg@docker.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `--tree` implementation already does this.
Make the behavior consistent for the legacy image list implementation.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Some tests had to be skipped as there's some issues to address, and
some of the result-types cannot be mocked / stubbed.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This release addresses breakage caused by a security patch included in
Go 1.25.2 and 1.24.8, which enforced overly restrictive validation on
the parsing of X.509 certificates. We've removed those restrictions
while maintaining the security fix that the initial release addressed.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This utility was used internally and will be removed in the next release.
Use `DetectContextType` to detect the context-type, and use `GetContextFromLocalDir`,
`GetContextFromLocalDir`, `GetContextFromGitURL`, or `GetContextFromURL`
instead.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
follow-up to f81816ef88
We could decide to not initialize a default, and do this only when
setting a value, but this may require more changes, so doing this
as a starting-point.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This method was added as part of a refactor in [moby@1ba1138], at which
time it was used to delete original values for "--host" and "--volume"
after normalizing. This beccame redundant in [moby@6200002], which added
specialized options that used a validate function, which both validated
and normalized inputs.
It's no longer used, so let's mark it deprecated so that we can remove it.
[moby@1ba1138]: 1ba11384bf
[moby@6200002]: 6200002669
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was added in 7405ac5c2d as
a fallback for API < v1.41, which did not include the service status
in the response. Current API versions return this information, so there's
no need to fetch it manually.
It was not gated by API version for some tests (which didn't set API
version), but should not be needed for non-test situations.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Support for API versions < v1.44 was removed in the client in [moby@96b29f5]
and [moby@7652f38], so we can remove fallback-code from the CLI as well,
as it won't be able to use those versions.
[moby@96b29f5]: 96b29f5a1f
[moby@7652f38]: 7652f38c28
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- fix minor linting issues (unhandled errors)
- rename vars to prevent shadowing
- use sub-tests for tests that already prepared for it
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This minor release includes 10 security fixes following the security policy:
- net/mail: excessive CPU consumption in ParseAddress
The ParseAddress function constructed domain-literal address components through repeated string concatenation. When parsing large domain-literal components, this could cause excessive CPU consumption.
Thanks to Philippe Antoine (Catena cyber) for reporting this issue.
This is CVE-2025-61725 and Go issue https://go.dev/issue/75680.
- crypto/x509: quadratic complexity when checking name constraints
Due to the design of the name constraint checking algorithm, the processing time
of some inputs scales non-linearly with respect to the size of the certificate.
This affects programs which validate arbitrary certificate chains.
Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-58187 and Go issue https://go.dev/issue/75681.
- crypto/tls: ALPN negotiation errors can contain arbitrary text
The crypto/tls conn.Handshake method returns an error on the server-side when
ALPN negotation fails which can contain arbitrary attacker controlled
information provided by the client-side of the connection which is not escaped.
This affects programs which log these errors without any additional form of
sanitization, and may allow injection of attacker controlled information into
logs.
Thanks to National Cyber Security Centre Finland for reporting this issue.
This is CVE-2025-58189 and Go issue https://go.dev/issue/75652.
- encoding/pem: quadratic complexity when parsing some invalid inputs
Due to the design of the PEM parsing function, the processing time for some
inputs scales non-linearly with respect to the size of the input.
This affects programs which parse untrusted PEM inputs.
Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-61723 and Go issue https://go.dev/issue/75676.
- net/url: insufficient validation of bracketed IPv6 hostnames
The Parse function permitted values other than IPv6 addresses to be included in square brackets within the host component of a URL. RFC 3986 permits IPv6 addresses to be included within the host component, enclosed within square brackets. For example: "http://[::1]/". IPv4 addresses and hostnames must not appear within square brackets. Parse did not enforce this requirement.
Thanks to Enze Wang, Jingcheng Yang and Zehui Miao of Tsinghua University for reporting this issue.
This is CVE-2025-47912 and Go issue https://go.dev/issue/75678.
- encoding/asn1: pre-allocating memory when parsing DER payload can cause memory exhaustion
When parsing DER payloads, memories were being allocated prior to fully validating the payloads.
This permits an attacker to craft a big empty DER payload to cause memory exhaustion in functions such as asn1.Unmarshal, x509.ParseCertificateRequest, and ocsp.ParseResponse.
Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-58185 and Go issue https://go.dev/issue/75671.
- net/http: lack of limit when parsing cookies can cause memory exhaustion
Despite HTTP headers having a default limit of 1 MB, the number of cookies that can be parsed did not have a limit.
By sending a lot of very small cookies such as "a=;", an attacker can make an HTTP server allocate a large amount of structs, causing large memory consumption.
net/http now limits the number of cookies accepted to 3000, which can be adjusted using the httpcookiemaxnum GODEBUG option.
Thanks to jub0bs for reporting this issue.
This is CVE-2025-58186 and Go issue https://go.dev/issue/75672.
- crypto/x509: panic when validating certificates with DSA public keys
Validating certificate chains which contain DSA public keys can cause programs
to panic, due to a interface cast that assumes they implement the Equal method.
This affects programs which validate arbitrary certificate chains.
Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-58188 and Go issue https://go.dev/issue/75675.
- archive/tar: unbounded allocation when parsing GNU sparse map
tar.Reader did not set a maximum size on the number of sparse region data blocks in GNU tar pax 1.0 sparse files. A maliciously-crafted archive containing a large number of sparse regions could cause a Reader to read an unbounded amount of data from the archive into memory. When reading from a compressed source, a small compressed input could result in large allocations.
Thanks to Harshit Gupta (Mr HAX) - https://www.linkedin.com/in/iam-harshit-gupta/ for reporting this issue.
This is CVE-2025-58183 and Go issue https://go.dev/issue/75677.
- net/textproto: excessive CPU consumption in Reader.ReadResponse
The Reader.ReadResponse function constructed a response string through
repeated string concatenation of lines. When the number of lines in a response is large,
this could cause excessive CPU consumption.
Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-61724 and Go issue https://go.dev/issue/75716.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This warning is better handled by the daemon, where applicable, as
the client does not have all information available to determine
if using a localhost / loopback-address for the DNS is possible.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The ResolveDefaultContext function was exported in [cli@f820766] to allow
(unit) testing, but did not document that it was only exported for this
purpose. The only external use of this function is in buildx, which uses
it in a unit test that can be implemented without this function.
This patch deprecates the function so that we can remove it.
[cli@f820766]: f820766f6a
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's just a wrapper around WithAPIClient, and not needed for this
test, which validates that "Initialize" properly creates the context
store, even if a client was already set;
3b26cfce8b
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
On Docker CLI versions before v28.0.0, using an unknown flag would print
the usage output, showing all available top-level flags and commands;
docker --badopt
unknown flag: --badopt
See 'docker --help'.
Usage: docker [OPTIONS] COMMAND
A self-sufficient runtime for containers
Options:
--config string Location of client config files (default "/root/.docker")
...
This output did not include plugin-commands, making the usage output
incomplete. That issue was fixed in [cli@40a6cf7], which loaded all
available cli-plugins, so that a stub was created for printing the
plugin commands in the usage output. Similarly, [cli@79a75da] added
code to hide experimental commands and commands not supported by the
daemon.
However, since 28.0.0 (commit [cli@f28fc7f]), the usage output was
removed for this error, so loading plugins is no longer needed;
docker --badopt
unknown flag: --badopt
Usage: docker [OPTIONS] COMMAND [ARG...]
Run 'docker --help' for more information
This patch removes the code added in [cli@40a6cf7] and [cli@79a75da].
With this patch, the output is still the same;
docker --unknown-flag buildx ls --no-such
unknown flag: --unknown-flag
Usage: docker [OPTIONS] COMMAND [ARG...]
Run 'docker --help' for more information
This function only handles flags defined by the CLI itself; invalid
flags for plugins are handled by the plugin itself, so are not
impacted;
docker buildx ls --no-such
unknown flag: --no-such
Usage: docker buildx ls
Run 'docker buildx ls --help' for more information
[cli@f28fc7f]: f28fc7f82f
[cli@40a6cf7]: 40a6cf7c47
[cli@79a75da]: 79a75da0fd
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Support for daemon-side auto-remove was added in API v1.25; on older
versions of the daemon, the client was responsible for removing the
container after it exited (see [moby@6dd8e10])
On API versions < 1.30, it used the events API for this purpose, and
would wait for a "die", "detach" or "detroy" events to know the container
exited, and could be removed or (when attached, but without a TTY) to
get the container's exit-status. (see [cli@38591f2]).
API version 1.24 (docker 1.12) is 9 Years old (July 29, 2016), and API
1.30 (docker 17.06) is 8 Years old (Jun 20, 2017), and long EOL. While
technically, a CLI could negotiate API 1.30 or older, this would only
be in cases where either API version negotiation failed, or the version
was explicitly overridden through `DOCKER_API_VERSION` for testing.
Either of those cases would be rare, and not worth the technical complexity
to support. This patch removes support for AutoRemove on API < 1.30.
[moby@6dd8e10]: 6dd8e10d6e
[cli@38591f2]: 38591f20d0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `VirtualSize` field was deprecated in [moby@1261fe6], and omitted / removed
in API v1.44 in [moby@913b0f5], and the corresponding formatting placeholder
was deprecated in [cli@f02301a].
This patch removes the formatting function, which also removes it from the
`docker image ls --format=json` output.
[moby@1261fe6]: 1261fe69a3
[moby@913b0f5]: 913b0f51ca
[cli@f02301a]: f02301ab5d
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Relates to [cli@27b2797], which forked this type from the Moby API, and
[cli@aab947d], which fixed the deprecation comment.
This field is no longer used since Docker 1.11 (API version 1.23) through
[moby@aee260d] and [engine-api@9a9e468], and the fix of the deprecation
comment was included in the 28.4.0 release.
This patch removes the field.
[cli@27b2797]: 27b2797f7d
[cli@aab947d]: aab947de8f
[moby@aee260d]: aee260d4eb
[engine-api@9a9e468]: 9a9e468f50
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This utility was added in 9b83d5bbf9, but
was never used. Remove the utility, and rewrite the error returned to
implement the errdefs.NotFound interface, so that it can be detected
using the errdefs.IsNotFound() utility if needed.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Commit [cli@27b2797] forked the AuthConfig type from the API, and changed
existing code to do a direct cast / convert of the forked type to the API
type. This can cause issues if the API types diverges, such as the removal
of the Email field.
This patch explicitly maps each field to the corresponding API type, but
adds some TODOs, because various code-paths only included a subset of the
fields, which may be intentional for fields that were meant to be handled
on the daemon / registry-client only.
We should evaluate these conversions to make sure these fields should
be sent from the client or not (and possibly even removed from the API
type).
[cli@27b2797]: 27b2797f7d
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function is a wrapper around apiClient.ImagePull and not directly
related to docker content trust; it just happens to also be called
when using content trust (through the trustedPull utility).
Move it together with the `runPull` function to separate it from
trust-related code.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function is a wrapper around apiClient.ImagePull; the use of
trust.ImageRefAndAuth was out of convenience because it's also called
when using content trust (through the trustedPull utility).
Let's pull away the layers to separate it from trust code.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This interface is used in tests to provide a dummy notary client,
but none of the tests require any arguments, so let's remove them.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The IndexInfo was only used to detect if the target was an official
image, which we can deduct from the hostname. Adding some normalizing
just in case (but we should only get "docker.io" here).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Docker Hub's Notary service is being retired, and now produces
failures in most cases. Add a warning when attempting to use
it, pending full removal of trust;
https://www.docker.com/blog/retiring-docker-content-trust/
With this PR:
DOCKER_CONTENT_TRUST=1 docker pull -q hello-world
WARNING: Docker is retiring DCT for Docker Official Images (DOI).
For details, refer to https://docs.docker.com/go/dct-deprecation/
could not validate the path to a trusted root: unable to retrieve valid leaf certificates
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
All information needed can be deducted from the image reference, which
is used to create a indexInfo, repoInfo, and to resolve auth-config.
In some situations this may result in resolving the auth-config twice
after it already was resolved to an encoded auth-config.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The Apply method was added when CLI options for constructing the CLI were
rewritten into functional options in [cli@7f207f3]. There was no mention
in the pull request of this method specifically, and this may have been
related to work being done elsewhere on compose-on-kubernetes or the
compose-cli plugin that may have needed options to modify the CLI config
after it was already initialized.
The CLI itself no longer depends on this method since [cli@133279f], and
there are no known external users. It was deprecated in [cli@24bfedf],
which is included in the 28.5.0 release, so we can remove it for 29.0.
[cli@7f207f3]: 7f207f3f95
[cli@133279f]: 133279fb0d
[cli@24bfedf]: 24bfedf3f8
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The Apply command was deprecated in 24bfedf3f8,
and has no known external users, but we didn't remove it from the interface.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was used internally, but is no longer used.
This method was deprecated in 0270b2d6f7,
which was included in the 28.5.0 release, and has no known external users,
so removing it for 29.0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These options were used internally as defaults for the constructor and
only impact commands implemented in the CLI itself.
They were deprecated in 40cdfc0d81, which
was included in the 28.5.0 release, so removing it for 29.0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was used internally, but is no longer used. Users should check
the value of the `DOCKER_CONTENT_TRUST` environment variable instead.
This method was deprecated in 11d40488dd,
which was included in the 28.5.0 release, and has no known external users,
so removing it for 29.0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The Apply method was added when CLI options for constructing the CLI were
rewritten into functional options in [cli@7f207f3]. There was no mention
in the pull request of this method specifically, and this may have been
related to work being done elsewhere on compose-on-kubernetes or the
compose-cli plugin that may have needed options to modify the CLI config
after it was already initialized.
The CLI itself no longer depends on this method since [cli@133279f], and
the only known consumer (docker compose) no longer needs it since [cli@2711800]
and [cli@048e931].
This patch deprecates the method with the intent to remove it in a future
release.
[cli@7f207f3]: 7f207f3f95
[cli@133279f]: 133279fb0d
[cli@2711800]: 2711800430
[cli@048e931]: 048e931b42
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was used internally, but is no longer used. Users should check
the value of the `DOCKER_CONTENT_TRUST` environment variable instead.
There are no known external users of this method, so already removing it
from the Cli interface; this method will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This is a follow-up to 7609dde8d0 and
3f5b1bdd32, which removed support for
DCT for build and plugin commands.
As these flags are just stubs, hidden by default and no longer functional,
they don't have to reflect the current state of DCT.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These options were used internally as defaults for the constructor and
only impact commands implemented in the CLI itself.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was used internally, but is no longer used. There are
no known users of this method, so already removing it from the Cli
interface.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Add support to the `cli/command` package to accept a custom User
Agent to pass to the underlying client.
This is used as the `UpstreamClient` portion of the `User-Agent`
when the Moby daemon makes requests.
For example, pushing and pulling images with Compose might result
in the registry seeing a `User-Agent` value of:
```
docker/24.0.7 go/go1.20.10 git-commit/311b9ff kernel/6.5.13-linuxkit os/linux arch/arm64 UpstreamClient(docker-cli-plugin-compose/v2.24.0)
```
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The CLI currently hard-codes the schema-version for CLI plugins to
"0.1.0", which doesn't allow us to expand the schema for plugins.
As there's many plugins that we shipped already, we can't break
compatibility until we reach 2.0.0, but we can expand the schema
with non-breaking changes.
This patch makes the validation more permissive to allow new schema
versions <= 2.0.0. Note that existing CLIs will still invalidate
such versions, so we cannot update the version until such CLIs are
no longer expected to be used, but this patch lays the ground-work
to open that option.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Before this patch, a broken symlink would print a warning;
docker info > /dev/null
WARNING: Plugin "/Users/thajeztah/.docker/cli-plugins/docker-feedback" is not valid: failed to fetch metadata: fork/exec /Users/thajeztah/.docker/cli-plugins/docker-feedback: no such file or directory
After this patch, such symlinks are ignored:
docker info > /dev/null
With debug enabled, we don't ignore the faulty plugin, which will
make the warning shown on docker info;
mkdir -p ~/.docker/cli-plugins
ln -s nosuchplugin ~/.docker/cli-plugins/docker-brokenplugin
docker --debug info
Client:
Version: 29.0.0-dev
Context: default
Debug Mode: true
Plugins:
buildx: Docker Buildx (Docker Inc.)
Version: v0.25.0
Path: /usr/libexec/docker/cli-plugins/docker-buildx
WARNING: Plugin "/Users/thajeztah/.docker/cli-plugins/docker-brokenplugin" is not valid: failed to fetch metadata: fork/exec /Users/thajeztah/.docker/cli-plugins/docker-brokenplugin: no such file or directory
# ...
We should als consider passing a "seen" map to de-duplicate entries.
Entries can be either a direct symlink or in a symlinked path (for
which we can filepath.EvalSymlinks). We need to benchmark the overhead
of resolving the symlink vs possibly calling the plugin (to get their
metadata) further down the line.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Currently, the plugin.Run command constructs the DockerCli using
the default options, assuming plugins run with all the same options
as the CLI itself; to customize the CLI there's a "Apply" option,
but this means mutating the CLI after it's already constructed, which
is not ideal.
This patch adds a variadic ops argument to allow CLI plugins to pass
custom options to use for the CLI, so that there's no need to mutate
its config in most cases.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The Apply method was added when CLI options for constructing the CLI were
rewritten into functional options in [cli@7f207f3]. There was no mention
in the pull request of this method specifically, and this may have been
related to work being done elsewhere on compose-on-kubernetes or the
compose-cli plugin that may have needed options to modify the CLI config
after it was already initialized.
We should try to remove functions that mutate the CLI configuration after
initialization if possible (and likely remove the `Apply` method); currently
this function is used in docker compose, but as part of a hack that can
probably be avoided.
[cli@7f207f3]: 7f207f3f95
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Mirantis Container Runtime (MCR) 23.0 reached EOL, and the next LTS
version of MCR is 25.x
- Docker 26.x reached EOL and is no longer maintained
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This code was added in 846ecf59ff as an
escape hatch in case the new OAuth login flow would cause problems.
We have not received reports where the new flow caused problems, and
searching the internet shows no mentions of the env-var.
This env-var was not documented, so we can remove it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Configuration options for experimental CLI features were deprecated in
docker 19.03 (3172219932), and enabled by
default since docker 20.10 (977d3ae046).
This field was deprecated in c8f9187157,
which is part of the 28.x release, and is unused. This patch removes
the field.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This const was added in 846ecf59ff, but
only used internally; commit 18cdc25bb4
deprecated the const, which was included in the 28.4 release.
This patch removes the exported const, as it's unused.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This package was deprecated in 13010ba673,
and only used internally. The deprecation was included in the 28.4 release,
and there's no known external users, so we can remove it for v29.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Commit [moby@17d870b] (API v1.13, docker v1.1.0) changed the default to pause
containers during commit, keeping the behavior opt-in for older API versions.
This version-gate was removed in [moby@1b1147e] because API versions lower
than v1.23 were no longer supported.
This patch deprecates the `--pause` flag in favor of a `--no-pause` flag to
be more explicit on the default. The old `--pause` flag is marked deprecated
but still functional. Using the deprecated flag will print a warning, and an
error is produced when trying to use both the old and new flag;
docker commit --pause mycontainer
Flag --pause has been deprecated, and enabled by default. Use --no-pause to disable pausing during commit.
docker commit --pause=false mycontainer
Flag --pause has been deprecated, and enabled by default. Use --no-pause to disable pausing during commit.
docker commit --pause --no-pause mycontainer
Flag --pause has been deprecated, use --no-pause instead
conflicting options: --no-pause and --pause cannot be used together
[moby@17d870b]: 17d870bed5
[moby@1b1147e]: 1b1147e46b
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `--kernel-memory` flag was still included to allow it to be used with
old API versions, but it's no longer supported by the kernel, and no longer
handled by OCI runtimes, so deprecating the flags.
With this patch, a deprecation warning is now produced when trying to use
the option;
docker run --kernel-memory 123b busybox
Flag --kernel-memory has been deprecated, this option is deprecated in the kernel and no longer supported
docker container create --kernel-memory 123b busybox
Flag --kernel-memory has been deprecated, and no longer supported by the kernel
31fb57e2c6434490a2892031602be20d0206d3cf0fc281ea25654c46dcb62bac
Note that cobra does not _fail_ the command when using deprecated options;
we could make this a hard failure instead, but may not be worth the effort.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This code was missing a check for the ID field before truncating it to a
shorter length for presentation. This would result in a panic if an event
would either have an empty ID field or a shorter length ID;
panic: runtime error: slice bounds out of range [:12] with length 0
goroutine 82 [running]:
github.com/docker/cli/cli/command/container.RunStats.func2({{0x0, 0x0}, {0x0, 0x0}, {0x0, 0x0}, {0x40001fcba0, 0x9}, {0x40001fcba9, 0x5}, ...})
/go/src/github.com/docker/cli/cli/command/container/stats.go:146 +0x1d0
created by github.com/docker/cli/cli/command/container.(*eventHandler).watch in goroutine 6
/go/src/github.com/docker/cli/cli/command/container/stats.go:363 +0x1c8
We need to look at this code in general; the truncated ID is passed to
NewStats, which uses the ID to propagate the `Container` field in the
`StatsEntry` struct. which is not used in the default format used by
`docker stats` and, having the same content as the `ID` field on the
same struct, doesn't make it very useful, other than being able to
present it under a `CONTAINER` column (instead of `CONTAINER ID`);
we should consider deprecating it; there may be some subtle things
to look into here; the `Container` field originally held the container
name. This was changed in [moby@ef915fd], which introduced separate
`ID` and `Name` fields, renaming the old `Name` field to container.
Looking at [`Stats.SetStatistics()`] and related code in [stats_helpers.go],
the `Container` field is used as the "canonical" reference for the stats
record; this allows the stats _data_ to be refreshed when a new stats
sample arrives for the same container (also see [moby@929a77b], which
moved locking to the `Stats` wrapper struct). This construct allows to
account for intermediate states, where a stats sample was incomplete
or could produce an error; in that case, the reference to the container
for which the stats were sampled is kept to allow removing a container
from the list once the container was removed. We should consider removing
`Container` as a formatting option, and moving the `Container` field to
the outer struct; this makes the outer struct responsible for keeping a
reference to the container, allowing the `StatsEntry` as a whole to be
replaced atomically.
This patch only addresses the panic;
- It changes the logic to preserve the container ID verbatim instead
of truncating. This allows stats samples to be matched against the
`Actor.ID` as-is.
- Truncating the `Container` is moved to the presentation logic;
currently this does not take `--no-trunc` into account to keep
the existing behavior, but we can (should) consider adding this.
- Logging is improved to use structured logs, and an extra check is
added to prevent empty IDs from being added as watcher.
[`Stats.SetStatistics()`]: 82281087e3/cli/command/container/formatter_stats.go (L88-L94)
[moby@ef915fd]: ef915fd036
[moby@929a77b]: 929a77b814
[stats_helpers.go]: 82281087e3/cli/command/container/stats_helpers.go (L26-L51)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Use sub-tests
- Don't use un-named keys
- Add test-cases for 'Name', 'ID' and custom container names
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Name is a required argument for both "create" and "update", so better
to split it from the options struct.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's adding a slight indirection by constructing a function when called,
but makes the completion functions more consistent, the signature easier
to read, and making the return type a [cobra.CompletionFunc] makes it
more transparent what it's intended for, and helps discovery of functions
that provide completion.
[cobra.CompletionFunc]: https://pkg.go.dev/github.com/spf13/cobra#CompletionFunc
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's adding a slight indirection by constructing a function when called,
but makes the completion functions more consistent, the signature easier
to read, and making the return type a [cobra.CompletionFunc] makes it
more transparent what it's intended for, and helps discovery of functions
that provide completion.
[cobra.CompletionFunc]: https://pkg.go.dev/github.com/spf13/cobra#CompletionFunc
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's adding a slight indirection by constructing a function when called,
but makes the completion functions more consistent, the signature easier
to read, and making the return type a [cobra.CompletionFunc] makes it
more transparent what it's intended for, and helps discovery of functions
that provide completion.
[cobra.CompletionFunc]: https://pkg.go.dev/github.com/spf13/cobra#CompletionFunc
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's adding a slight indirection by constructing a function when called,
but makes the completion functions more consistent, the signature easier
to read, and making the return type a [cobra.CompletionFunc] makes it
more transparent what it's intended for, and helps discovery of functions
that provide completion.
[cobra.CompletionFunc]: https://pkg.go.dev/github.com/spf13/cobra#CompletionFunc
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This uses the DefaultShellCompDirective feature which was added
in cobra to override the default (which would complete to use
files for commands and flags).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This uses the DefaultShellCompDirective feature which was added
in cobra to override the default (which would complete to use
files for commands and flags).
Note that we set "cobra.NoFileCompletions" for many commands, which
is redundant with this change, so we could remove as well.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These commands accept two arguments; the first is a custom name,
the second is either a filename or "-" to create from STDIN.
With this patch:
# does not provide completion
docker secret create <tab>
# starts providing completion once a non-empty name is provided
docker secret create somename<tab>
file.txt other-file.txt
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The "docker inspect" command can inspect any type of object, which would
require all possible endpoints to be contacted. By default, we don't
provide completion, but if a `--type` is passed, we provide completion
for the given type.
For example, `docker inspect --type container` will complete container
names, `docker inspect --type volume` will complete volume names and
so on.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Similar to 323fbc485e - this code was added
in [moby@c127d96], but used string-matching to detect cases where a user
tried to install an image as plugin. However, this handling no longer matched
any error-strings, so no longer worked:
docker plugin install busybox
Error response from daemon: did not find plugin config for specified reference docker.io/library/busybox:latest
[moby@c127d96]: c127d9614f
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were deprecated in ad6ab189a6 and
were only used internally. Move them back inside the stack package.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were deprecated in ad6ab189a6 and
were only used internally. Move them back inside the stack package.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This warning was added in [moby@4a8b3ca] to print a warning when building
Linux images from a Windows client. Window's filesystem does not have an
"executable" bit, which mean that, for example, copying a shell script
to an image during build would lose the executable bit. So for Windows
clients, the executable bit would be set on all files, unconditionally.
Originally this was detected in the client, which had direct access to
the API response headers, but when refactoring the client to use a common
library in [moby@535c4c9], this was refactored into a `ImageBuildResponse`
wrapper, deconstructing the API response into an `io.Reader` and a string
field containing only the `OSType` header.
This was the only use and only purpose of the `OSType` field, and now that
BuildKit is the default builder for Linux images, this warning didn't get
printed unless BuildKit was explicitly disabled.
This patch removes the warning, so that we can potentially remove the
field, or the `ImageBuildResponse` type altogether.
[moby@4a8b3ca]: 4a8b3cad60
[moby@535c4c9]: 535c4c9a59
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This change updates the builder prune command to send the `ReservedSpace` parameter in preparation of `KeepStorage` deprecation in API v1.52.
Signed-off-by: Austin Vazquez <austin.vazquez@docker.com>
We may still change this, but in the client module, the signature
of the client.Opt changed to now include a non-exported type, which
means that we can't construct a custom option that is implemented
using client options:
#18 16.94 # github.com/docker/cli/cli/context/docker
#18 16.94 cli/context/docker/load.go:105:29: cannot use withHTTPClient(tlsConfig) (value of type func(*client.Client) error) as client.Opt value in argument to append
#18 16.94 cli/context/docker/load.go:152:6: cannot use c (variable of type *client.Client) as *client.clientConfig value in argument to client.WithHTTPClient(&http.Client{…})
We can consider exporting the `client.clientConfig` type (but keep its
fields non-exported), but for this use, we don't strictly need it, so
let's change the implementation to not having to depend on that.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
We may still change this, but in the client module, the signature
of the client.Opt changed to now include a non-exported type, which
means that we can't construct a custom option that is implemented
using client options:
#18 16.94 # github.com/docker/cli/cli/context/docker
#18 16.94 cli/context/docker/load.go:105:29: cannot use withHTTPClient(tlsConfig) (value of type func(*client.Client) error) as client.Opt value in argument to append
#18 16.94 cli/context/docker/load.go:152:6: cannot use c (variable of type *client.Client) as *client.clientConfig value in argument to client.WithHTTPClient(&http.Client{…})
We can consider exporting the `client.clientConfig` type (but keep its
fields non-exported), but for this use, we don't strictly need it, so
let's change the implementation to not having to depend on that.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This includes 1 security fix:
- net/http: CrossOriginProtection bypass patterns are over-broad
When passing patterns to CrossOriginProtection.AddInsecureBypassPattern,
requests that would have redirected to those patterns (e.g. without a trailing
slash) were also exempted, which might be unexpected.
Thanks to Marco Gazerro for reporting this issue.
This is CVE-2025-47910 and Go issue https://go.dev/issue/75054.
View the release notes for more information:
https://go.dev/doc/devel/release#go1.24.7
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
With this patch, completion is provided for images already present
in the local image cache to help pulling the latest version of the
same tag;
docker pull go<tab>
golang:1.12 golang:1.18.0 golang:1.21 golang:1.24 gopher:latest
golang:1.13 golang:1.20 golang:1.23 golang:latest
docker pull golang:<tab>
1.12 1.13 1.18.0 1.20 1.21 1.23 1.24 latest
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The restrictedNamePattern was a basic regular expression. Replace it
with a minimal utility to do the same, without having to use regular
expressions (or the "lazyregexp" package).
Some quick benchmarking (not committed) show that the non-regex approach
is ~18x faster:
BenchmarkIsValidName_Regex_Valid-10 8516511 119.4 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidName_Manual_Valid-10 172426240 6.964 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidName_Regex_Invalid-10 34824540 34.22 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidName_Manual_Invalid-10 550804021 2.173 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidName_Regex_Parallel-10 69289900 17.30 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidName_Manual_Parallel-10 1000000000 0.9296 ns/op 0 B/op 0 allocs/op
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The pluginNameRe was a basic regular expression, effectively only checking
if the name consisted of lowercase alphanumeric characters. Replace it
with a minimal utility to do the same, without having to use regular
expressions (or the "lazyregexp" package).
Some quick benchmarking (not committed) show that the non-regex approach
is ~25x faster:
BenchmarkIsValidPluginName_Regex_Valid-10 13956240 81.39 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidPluginName_Manual_Valid-10 360003060 3.318 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidPluginName_Regex_Invalid-10 35281794 33.74 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidPluginName_Manual_Invalid-10 906072663 1.320 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidPluginName_Regex_Parallel-10 96595677 12.04 ns/op 0 B/op 0 allocs/op
BenchmarkIsValidPluginName_Manual_Parallel-10 1000000000 0.4541 ns/op 0 B/op 0 allocs/op
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These consts and types were moved to a separate metadata package in commits
292713c887 and 4321293972,
and deprecated in 72f76f2720, 5876b2941c,
and 6fa7d18320.
This removes the deprecated aliases in `cli-plugins/manager` in favor of
their equivalent in `cli-plugins/manager/metadata`:
- `CommandAnnotationPlugin`
- `CommandAnnotationPluginVendor`
- `CommandAnnotationPluginVersion`
- `CommandAnnotationPluginInvalid`
- `CommandAnnotationPluginCommandPath`
- `NamePrefix`
- `MetadataSubcommandName`
- `HookSubcommandName`
- `Metadata`
- `ReexecEnvvar`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These errors satisfy errdefs.IsNotFound, which can be used instead. This
function was deprecated in 71460215d3 and
is no longer used.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This const was added in 846ecf59ff, but
only used internally. This patch deprecates the const, to be removed
in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This alias was added in 4321293972, which is
part of v28.0, but did not deprecate them. They are no longer used in the
CLI itself, but may be used by cli-plugin implementations.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These functions and types are shallow wrappers around the context
store and were intended for internal use as implementation for the
CLI itself.
They were exported in 3126920af1 to be
used by plugins and Docker Desktop. However, there's currently no public
uses of this, and Docker Desktop does not use these functions. These were
deprecated in 95eeafa551 and are no longer
used.
This patch removes the deprecated functions as they were meant to be
implementation specific for the CLI. If there's a need to provide
utilities for manipulating the context-store other than through the
CLI itself, we can consider creating an SDK for that purpose.
This removes:
- `RunCreate` and `CreateOptions`
- `RunExport` and `ExportOptions`
- `RunImport`
- `RunRemove` and `RemoveOptions`
- `RunUpdate` and `UpdateOptions`
- `RunUse`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were deprecated in 036d3a6bab and
30774ed1f2, and were originally in the
cli/command/stack package, but moved for the (now deprecated) Compose
on Kubernetes feature in 4d947de292.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This replaces the visitAll recursive function with a test that verifies that
the option is set for all commands and subcommands, so that it doesn't have
to be modified at runtime.
We currently still have to loop over all functions for the setValidateArgs
call, but that can be looked at separately.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These functions and types are shallow wrappers around the context
store and were intended for internal use as implementation for the
CLI itself.
They were exported in 3126920af1 to be
used by plugins and Docker Desktop. However, there's currently no public
uses of this, and Docker Desktop does not use these functions.
This patch deprecates the exported functions as they were meant to be
implementation specific for the CLI. If there's a need to provide
utilities for manipulating the context-store other than through the
CLI itself, we can consider creating an SDK for that purpose.
This deprecates:
- `RunCreate` and `CreateOptions`
- `RunExport` and `ExportOptions`
- `RunImport`
- `RunRemove` and `RemoveOptions`
- `RunUpdate` and `UpdateOptions`
- `RunUse`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Use a struct-literal for the IPAM config, and combine some of the checks.
Also use the Name field as a default, and only construct a scoped name
if the given name is empty (instead of the reverse).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function is currently only used within the package; create a non-exported
version of it, to make it clear it's not used elsewhere. This patch keeps
the exported function for now, but we can decide if we need to keep it
in future.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It was only used in a single place, and possibly incorrect. Let's inline
it to put the logic where it's used.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These are very small structs, so using pointers doesn't bring much
advantage and makes it slightly more cumbersome to use.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This intermediate slice was a left-over from the "Compose on Kubernetes"
feature, which required some conversions, but that code was removed in
193ede9b12, so the intermediate slice no
longer has a purpose.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Preserve the original order by avoiding the intermediate map[string] and
keeping an index for the first occurrence of a stack; this also avoids
looping multiple times.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Include name in test-table
- Don't use un-keyed values in struct
- Simplify test-table to take a format string instead of a whole formatter.Context
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Functions and types in this package were exported as part of the "compose
on kubernetes" feature, which was deprecated and removed. These functions
are meant for internal use, and will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Functions and types in this package were exported as part of the "compose
on kubernetes" feature, which was deprecated and removed. These functions
are meant for internal use, and will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Functions and types in this package were exported as part of the "compose
on kubernetes" feature, which was deprecated and removed. These functions
are meant for internal use, and will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Functions and types in this package were exported as part of the "compose
on kubernetes" feature, which was deprecated and removed. These functions
are meant for internal use, and will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Functions and types in this package were exported as part of the "compose
on kubernetes" feature, which was deprecated and removed. These functions
are meant for internal use, and will be removed in the next release.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This only impacts the JSON marshaled output; the "regular" output
of `docker info` already ignores these fields.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were just testing JSON marshaling fields that are deprecated, but
may be present in a response; these fields will be removed in future
API versions, so stop testing for them.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This special handling was added in [moby@9b6dcc8], and later updated in
[moby@c127d96], but it fully depended on string-matching, which is brittle.
Testing the original ticket that lead to this handling, it looks like the
string matching no longer works, and the daemon error is returned as-is:
With graphdrivers:
docker pull tiborvass/no-remove
Using default tag: latest
Error response from daemon: Encountered remote "application/vnd.docker.plugin.v0+json"(unknown) when fetching
With containerd snapshotters enabled:
docker pull tiborvass/no-remove
Using default tag: latest
latest: Pulling from tiborvass/no-remove
cf635291f7c9: Download complete
failed to unpack image on snapshotter overlayfs: mismatched image rootfs and manifest layers
The error-message for containerd can probably be improved, but as the special
handling in the CLI no longer works, we can remove it.
[moby@9b6dcc8]: 9b6dcc8b9d
[moby@c127d96]: c127d9614f
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was used to share it between "trust" and "image",
but was only a shallow wrapper, so split the implementations where
used.
It was deprecated in 7ad113ccc2 and is
no longer used.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Now that 3f5b1bdd32 removed DCT, which
needed some of the intermediate types (indexInfo), we can simplify the
auth code further and just get the base64-encoded AuthConfig to be set
as header.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were deprecated in a5f4ba08d9 and only
used internally.
This removes the deprecated types and functions:
- `RunConfigCreate` and `CreateOptions`
- `RunConfigInspect` and `InspectOptions`
- `RunConfigList` and `ListOptions`
- `RunConfigRemove` and `RemoveOptions`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were exported in f60369dfe6 to be
used in docker enterprise, but this never happened, and there's no
known consumers of these, so we should deprecate these. External
consumers can still call the API-client directly, which should've
been the correct thing to do in the first place.
This deprecates:
- `RunConfigCreate` and `CreateOptions`
- `RunConfigInspect` and `InspectOptions`
- `RunConfigList` and `ListOptions`
- `RunConfigRemove` and `RemoveOptions`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Relates to [cli@27b2797], which forked this type from the Moby API, and
[moby@6cfff7e], which made the same change on the API side.
The Email field was originally used to create a new Docker Hub account
through the `docker login` command. The `docker login` command could be
used both to log in to an existing account (providing only username and
password), or to create a new account (providing desired username and
password, and an e-mail address to use for the new account).
This functionality was confusing, because it was implemented when Docker
Hub was the only registry, but the same functionality could not be used
for other registries. This functionality was removed in Docker 1.11 (API
version 1.23) through [moby@aee260d], which also removed the Email field
([engine-api@9a9e468]) as it was no longer used.
However, this caused issues when using a new CLI connecting with an old
daemon, as the field would no longer be serialized, and the deprecation
may not yet be picked up by custom registries, so [engine-api@167efc7]
added the field back, deprecated it, and added an "omitempty". There
was no official "deprecated" format yet at the time, so let's make sure
the deprecation follows the proper format to make sure it gets noticed.
[cli@27b2797]: 27b2797f7d
[moby@6cfff7e]: 6cfff7e880
[moby@aee260d]: aee260d4eb
[engine-api@9a9e468]: 9a9e468f50
[engine-api@167efc7]: 167efc72bb
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These flags were added in 20a6ff32ee, and require
API version v1.32 or up, but they accidentally copied the flag-name from another
flag, so were not setting the annotation correctly.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Remove redundant intermediate variables
- Explicitly use an early return on error instead of combining with
other checks.
- Fix unhandled errors and combine defers
- Remove outstanding TODO that unlikely will be addressed
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `reference.TrimNamed` function strips both digests and tags; the
formatter function only wants to remove the digest, but preserve any
tags present.
Update the implementation to only trim the reference if there's a digest
present, otherwise use it as-is.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was exported to share it between "trust" and "image",
but was only a shallow wrapper, so split the implementations where
used.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were exported in 812f113685, but
while the function and options are exported, the option-fields were
all un-exported, so these were not usable.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch deprecates exported NewPruneCommand and moves the
implementation details to an unexported function.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Use a table-test to more easily allow adding test-cases
- Use the test-name itself as name for the golden file
- Use a struct-literal to create the fixture for formatting.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
This deprecates the following types and functions:
- `SignedTagInfo`
- `SignerInfo`
- `NewTrustTagFormat`
- `NewSignerInfoFormat`
- `TagWrite`
- `SignerInfoWrite`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The RegisterLegacy and Register functions register constructors for
commands, so we should expect them to be a fresh copy that is not
shared, which means that we can mutate the command in-place.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The Register and RegisterLegacy functions are designed to be used
in package init() functions, which are guaranteed to be run
sequentially.
From the documentation (https://go.dev/ref/mem#init);
> Program initialization runs in a single goroutine, but that goroutine
> may create other goroutines, which run concurrently. If a package `p`
> imports package `q`, the completion of `q`'s `init` functions happens
> before the start of any of `p`'s.
>
> The completion of all `init` functions is synchronized before the
> start of the function `main.main`.
This patch removes the synchonisation as no concurrency should happen
if these functions are used as intended.
As the internal queue is not expected to be mutated after use, we also
don't have to return a copy.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch removes the explicit `commands.AddCommands` function and
instead relies upon the `internal/commands` package which registers each
CLI command using `init()` instead.
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported trust commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- trust.NewTrustCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported plugin commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- plugin.NewPluginCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
- Parse/format the repository name when constructing and store the
result.
- Remove the Name() accessor, as this type is only used internally,
and no longer had any special handling.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch deprecates exported swarm commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- swarm.NewSwarmCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported registry commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- registry.NewLoginCommand
- registry.NewLogoutCommand
- registry.NewSearchCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported stack commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- stack.NewStackCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported context commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- context.NewContextCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported volume commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- volume.NewVolumeCommand
- volume.NewPruneCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported service commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- service.NewServiceCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported secret commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- secrets.NewSecretCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported manifest commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- manifest.NewManifestCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported node commands and moves the implementation
details to an unexported function.
Commands that are affected include:
- node.NewNodeCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This patch deprecates exported network commands and moves the
implementation details to an unexported function.
Commands that are affected include:
- network.NewNetworkCommand
- network.NewPruneCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This non-standard handling for these options was added in [moby@e4c1f07]
and [moby@abe32de] to work around a regression in Docker 1.13 that caused
`docker-machine` to fail. Preserving quotes in such cases is expected (and
standard behavior), but versions of Docker before 1.13 used a custom "mflag"
package for flag parsing, and that package contained custom handling for
quotes (added in [moby@0e9c40e]).
Given that Docker Machine reached EOL a long time ago and other options,
such as `docker context`, have been added to configure the CLI to connect
to a specific host (with corresponding TLS configuration), we can remove
the special handling for these flags, as it's inconsistent with all other
flags, and not worth maintaining for a tool that no longer exists.
[moby@e4c1f07]: e4c1f07729
[moby@abe32de]: abe32de6b4
[moby@0e9c40e]: 0e9c40eb82
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch deprecates exported system commands and moves the
implementation details to an unexported function.
Commands that are affected include:
- system.NewVersionCommand
- system.NewInfoCommand
- system.NewSystemCommand
- system.NewEventsCommand
- system.NewInspectCommand
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
- Use sentence-case to follow our docs guidelines.
- Add newlines to prevent these being rendered on a
single line.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Rewrite the test to not depend on registry.DecodeAuthConfig, which
may be moved internal to the daemon as part of the modules transition.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch unexports the `builder` and `bake` stub command and it adds
deprecation notices on the exported functions.
It also registers the commands using the new `cli/internal/commands`
package when the init function executes.
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
These aliases were added in 292713c887
(part of v28.0), but did not deprecate them. They are no longer used
in the CLI itself, but may be used by cli-plugin implementations.
This deprecates the aliases in `cli-plugins/manager` in favor of
their equivalent in `cli-plugins/manager/metadata`:
- `CommandAnnotationPlugin`
- `CommandAnnotationPluginVendor`
- `CommandAnnotationPluginVersion`
- `CommandAnnotationPluginInvalid`
- `CommandAnnotationPluginCommandPath`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These were deprecated in 6f0c66c152 and are
no longer used.
This removes the deprecated:
- `NewNamedListOptsRef`
- `NewNamedMapOpts`
- `NamedListOpts`
- `NamedMapOpts`
- `NamedOption`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `--tlscacert`, `--tlscert`, and `--tlskey` command-line flags had
non-standard behavior for handling values contained in quotes (`"` or `'`).
Normally, quotes are handled by the shell, for example, in the following
example, the shell takes care of handling quotes before passing the values
to the `docker` CLI:
docker --some-option "some-value-in-quotes" ...
However, when passing values using an equal sign (`=`), this may not happen
and values may be handled including quotes;
docker --some-option="some-value-in-quotes" ...
This caused issues with "Docker Machine", which used this format as part
of its `docker-machine config` output, and the CLI carried special, non-standard
handling for these flags.
Docker Machine reached EOL, and this special handling made the processing
of flag values inconsistent with other flags used, so this behavior is
deprecated. Users depending on this behavior are recommended to specify
the quoted values using a space between the flag and its value, as illustrated
above.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `NewNamedListOptsRef`, `NewNamedMapOpts` and related `NamedListOpts`,
`NamedMapOpts`, and `NamedOption` interface were added in [moby@677a6b3],
which added support for a `daemon.json` configuration file. That change
required a way to correlate command-line flags with their corresponding
fields in the `daemon.json` to detect conflicting options. At the time,
the CLI and daemon were produced from the same code, and shared packages
for command-line options, but when the CLI was moved to a separate
repository, these options were inherited.
[moby@677a6b3]: 677a6b3506
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
hostVar is used for the '--host' / '-H' flag to set [ClientOptions.Hosts].
The [ClientOptions.Hosts] field is a slice because it was originally shared
with the daemon config. However, the CLI only allows for a single host to
be specified.
hostVar presents itself as a "string", but stores the value in a string
slice. It produces an error when trying to set multiple values, matching
the check in [getServerHost].
[getServerHost]: 7eab668982/cli/command/cli.go (L542-L551)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The ClientOptions struct and related flags were inherited from the Moby
repository, where originally the CLI and Daemon used the same implementation
and had a "Common" options struct. When the CLI moved to a separate repository,
those structs were duplicated, but some daemon-specific logic remained. For
example, the daemon can be configured to listen on multiple ports and sockets
([moby@dede158]), but the CLI [can only connect to a single host][1]. The
daemon config also had to account for flags conflicting with `daemon.json`,
and use special flag-vars for this ([moby@677a6b3]).
Unfortunately, the `ClientConfig` struct became part of the public API and
is used as argument in various places, but we can remove the use of the
special flag var. This patch replaces the use of `NewNamedListOptsRef`
for a regular `StringArray`.
Unfortunately this changes the flag's type description from `list` to
`stringArray`, but we can look at changing that separately.
[moby@dede158]: dede1585ee
[1]: 0af135e906/docker/docker.go (L191-L193)
[moby@677a6b3]: 677a6b3506
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `ValidateHost` option was introduced in [moby@1ba1138] to be used
as validation func for the `--host` flag on the daemon in and CLI in
[moby@5e3f6e7], but is no longer used since [cli@6f61cf0]. which added
support for `ssh://` connections, and required validation elsewhere.
[moby@1ba1138]: 1ba11384bf
[moby@5e3f6e7]: 5e3f6e7023
[cli@6f61cf0]: 6f61cf053a
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `QuotedString` option was added in [moby@e4c1f07] and [moby@abe32de]
to work around a regression in Docker 1.13 that caused `docker-machine`
to fail. `docker-machine` produced instructions on how to set up a cli
to connect to the Machine it produced. These instructions used quotes
around the paths for TLS certificates, but with an `=` for the flag's
values instead of a space; due to this the shell would not handle
stripping quotes, so the CLI would now get the value including quotes.
Preserving quotes in such cases is expected (and standard behavior), but
versions of Docker before 1.13 used a custom "mflag" package for flag
parsing, and that package contained custom handling for quotes (added
in [moby@0e9c40e]).
For other flags, this problem could be solved by the user, but as these
instructions were produced by `docker-machine`'s `config` command, an
exception was made for the `--tls-xxx` flags. From [moby-29761]:
> The flag trimming behaviour is really unusual, and I would say unexpected.
> I think removing it is generally the right idea. Since we have one very
> common case where it's necessary for backwards compatibility we need to
> add a special case, but I don't think we should apply that case to every
> flag.
The `QuotedString` implementation has various limitations, as it doesn't
follow the same handling of quotes as a shell would.
Given that Docker Machine reached EOL a long time ago and other options,
such as `docker context`, have been added to configure the CLI to connect
to a specific host (with corresponding TLS configuration), we should remove
the special handling for these flags, as it's inconsitent with all other
flags, and not worth maintaining for a tool that no longer exists.
This patch deprecates the `QuotedString` option and removes its use. A
temporary, non-exported copy is added, but will be removed in the next
release.
[moby-29761]: https://github.com/moby/moby/issues/29761#issuecomment-270211265
[moby@e4c1f07]: e4c1f07729
[moby@abe32de]: abe32de6b4
[moby@0e9c40e]: 0e9c40eb82
[moby@c79a169]: c79a169a35
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These utilities were only used internally; create a local copy
where used, and deprecate the ones in cli.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It was only used in a single place and has no external consumers.
Move it to where it's used to keep things together.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The example shows that the `--volumes` option is used, which in current
versions of docker only removes "anonymous" volumes, but preserves named
volume:
$ docker system prune -a --volumes
...
- all anonymous volumes not used by at least one container
...
But the example output showed that a named volume ("named-vol") was
deleted;
Deleted Volumes:
named-vol
Co-authored-by: Roberto Villarreal <rrjjvv@yahoo.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The label validation steps now properly fail the workflow when required
labels are missing, instead of continuing with errors.
This change removes the `continue-on-error: true` directives and adds
`always()` conditions to ensure the validation steps run regardless of
previous step outcomes.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The PR validation workflow now enforces that every PR with an 'impact/*'
label must also have a corresponding 'kind/*' label, in addition to the
existing 'area/*' label requirement.
This change helps ensure proper categorization of pull requests by
requiring contributors to specify both the impact area and the kind of
change being made.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The logic used in this function was confusing; some errors were ignored,
but responses handled regardless. The intent here is to try to detect
whether the destination exists inside the container and is of the right
"type" (otherwise produce an error).
Failing to "stat" the path in the container means we can't produce a
nice error for the user, but we'll continue the request, which either
would succeed or produce an error returned by the daemon.
While working on this patch, I noticed that some error-handling on the
daemon side is incorrect. This patch does not fix those cases, but
makes the logic slightly easier to follow (we should consider extracting
the "stat" code to a separate function though).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- deprecate sockets.GetProxyEnv, sockets.DialerFromEnvironment
- add support for unix sockets on Windows
- remove legacy CBC cipher suites from client config
- align client and server defaults to be the same.
- remove support for encrypted TLS private keys.
- nat: optimize ParsePortSpec
full diff: https://github.com/docker/go-connections/compare/v0.5.0...v0.6.0
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This test was recently rewritten from testing plugin upgrade with
DCT enabled to just "plugin upgrade", but there's a fair amount of
complexity in the e2e tests that set up different daemons and registries.
It's possible that tests are affecting each-other, and some state (config)
is left behind.
Let's skip the test for now, and add a tracking ticket to dig deeper.
=== FAIL: e2e/global TestPromptExitCode/plugin_upgrade (7.55s)
cli_test.go:205: assertion failed:
Command: docker plugin push registry:5000/plugin-upgrade-test:latest
ExitCode: 1
Error: exit status 1
Stdout: The push refers to repository [registry:5000/plugin-upgrade-test]
459089aa5943: Preparing
adc41078d1d9: Preparing
d7bff979db13: Preparing
459089aa5943: Preparing
Stderr: error pushing plugin: failed to do request: Head "https://registry:5000/v2/plugin-upgrade-test/blobs/sha256:adc41078d1d937495df2f90444e5414a01db31e5a080f8aa4f163c64d41abd11": http: server gave HTTP response to HTTPS client
Failures:
ExitCode was 1 expected 0
Expected no error
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- https://github.com/golang/go/issues?q=milestone%3AGo1.24.6+label%3ACherryPickApproved
- full diff: golang/go@go1.24.5...go1.24.6
These minor releases include 2 security fixes following the security policy:
- os/exec: LookPath may return unexpected paths
If the PATH environment variable contains paths which are executables (rather
than just directories), passing certain strings to LookPath ("", ".", and ".."),
can result in the binaries listed in the PATH being unexpectedly returned.
Thanks to Olivier Mengué for reporting this issue.
This is CVE-2025-47906 and Go issue https://go.dev/issue/74466.
- database/sql: incorrect results returned from Rows.Scan
Cancelling a query (e.g. by cancelling the context passed to one of the query
methods) during a call to the Scan method of the returned Rows can result in
unexpected results if other queries are being made in parallel. This can result
in a race condition that may overwrite the expected results with those of
another query, causing the call to Scan to return either unexpected results
from the other query or an error.
We believe this affects most database/sql drivers.
Thanks to Spike Curtis from Coder for reporting this issue.
This is CVE-2025-47907 and https://go.dev/issue/74831.
View the release notes for more information:
https://go.dev/doc/devel/release#go1.24.6
Signed-off-by: Austin Vazquez <austin.vazquez@docker.com>
It was only used internally and has no external users. It should not be
used for new uses, because it also adds a minimum API version constraint
and a default from env-var, which must be evaluated for each individual
use of such flags.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Plugins are not widely used, and there's no known plugins that use
content-trust. We're working on updating the authentication stack
in the CLI, and the trust implementation hinders us in making
changes, so removing parts that are not high-priority (ahead of
full deprecation of DCT).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Use names for the plugin that don't refer to content-trust, as that's
not related to this test.
Make it slightly more clear which steps are preparation and which
are the actual test. The test sometimes fails in the preparation
step, and we could consider moving those separate and XFail the
test if the preparation fails;
Stderr: error pushing plugin: failed to do request: Head "https://registry:5000/v2/plugin-content-trust-upgrade/blobs/sha256:af932a31d4df3a2890f900bcf28e16cea87b2b440b8036ba86ab3418f3e50a35": http: server gave HTTP response to HTTPS client
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The gotest.tools `fs.NewDir` utility already sets up a `t.Cleanup`,
so we can treat it the same as `t.TempDir()` and let it handle
cleaning up by itself.
We should probably consider replacing some of this with `t.TempDir`.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- The `DisableInputEcho` and `PromptForInput` utilities were added in
c15ade0c64 as part of a bug-fix, which
was part of v28.x. [There are no (publicly visible) users][1] of either.
- The `ErrPromptTerminated` was added in v26.x (originally added in
10bf91a02d, later updated in commit
7c722c08d0. [It is not used][2]
- The `PromptForConfirmation` was added in [moby@280c872] (docker v1.13.0)
as part of the `docker <object> prune` subcommands. It was meant for
internal use but exported to allow re-using it in the `container`,
`image` (etc.) packages. However, a breaking change to its signature
was made in 10bf91a02d. It currently
does [not appear to have any (public) users][2].
This patch removes the `ErrPromptTerminated`, `DisableInputEcho`,
`PromptForInput`, and `PromptForConfirmation` utilities from the
`cli/command` package. The core functionality of these is still
available in the `internal/prompt` package, which we may make
public at some point, but still needs some refining / decoupling.
[moby@280c872]: 280c872366
[1]: https://grep.app/search?f.lang=Go®exp=true&q=%5C.%28DisableInputEcho%7CPromptForInput%29%5C%28
[2]: https://grep.app/search?f.lang=Go&q=%5C.ErrPromptTerminated
[3]: https://grep.app/search?f.lang=Go&q=.PromptForConfirmation%28
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This package will not be included in the api or client modules, and
we're currently only using a single function of it, and only the
unix implementation, so let's fork it for now (although the package
may be moved to moby/sys).
This removes the last dependency on github.com/docker/docker.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These aliases were added in 4321293972
(part of v28.0), but did not deprecate them. They are no longer used
in the CLI itself, but may be used by cli-plugin implementations.
This deprecates the aliases in `cli-plugins/manager` in favor of
their equivalent in `cli-plugins/manager/metadata`:
- `NamePrefix`
- `MetadataSubcommandName`
- `HookSubcommandName`
- `Metadata`
- `ReexecEnvvar`
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This was a pattern inheritted from pkg/errors.Wrapf, which ignored
nil errors for convenience. However, it is error-prone, as it is
not obvious when returning a nil-error.
All call-sites using `wrapAsPluginError` already do a check for
nil errors, so remove this code to prevent hard to find bugs.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It is for internal use, and no longer needed for testing, now that
the `Plugin` type handles marshalling errors.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Go does not by default marshal `error` type fields to JSON. The manager
package therefore implemented a `pluginError` type that implements
[encoding.TextMarshaler]. However, the field was marked as a regular
`error`, which made it brittle; assining any other type of error would
result in the error being discarded in the marshaled JSON (as used in
`docker info` output), resulting in the error being marshaled as `{}`.
This patch adds a custom `MarshalJSON()` on the `Plugin` type itself
so that any error is rendered. It checks if the error used already
implements [encoding.TextMarshaler], otherwise wraps the error in
a `pluginError`.
[encoding.TextMarshaler]: https://pkg.go.dev/encoding#TextMarshaler
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Experimental is always enabled (977d3ae046),
and the `Experimental` field in plugin metadata was deprecated in
977d3ae046 and removed in commit
6a50c4f700.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Move the version-check for pruners to the pruner, which can
return a [ErrNotImplemented] error to indicate they won't
be run with the API version that's used.
This helps separating concerns, and doesn't enforce knowledge
about what's supported by each content-type onto the system
prune command.
[ErrNotImplemented]: https://pkg.go.dev/github.com/docker/docker@v28.3.3+incompatible/errdefs#ErrNotImplemented
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This adds a "dry-run" / "pre-check" option for prune-functions,
which delegates constructing the confirmation message (what is
about to be pruned) and validation of the given options to the
prune-functions.
This helps separating concerns, and doesn't enforce knowledge
about what's supported by each content-type onto the system
prune command.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Introduce a "prune" package in which we maintain a list of prune
functions that are registered. Known prune "content-types" are
included in a pre-defined order, after which additional content
can be registered.
Using this approach no longer requires the "RunPrune" functions
to be exported, and allows additional content-types to be
introduced without having to import those packages into the
system package, so keeping things more decoupled.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Most places only use IndexInfo (and may not even need that), so replace
the use of ParseRepositoryInfo for NewIndexInfo, and move the RepositoryInfo
type to the trust package, which uses it as part of its ImageRefAndAuth
struct.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Use strong-typed switches for validating options
- Initialize defaults instead of setting them after
parsing the ports. Each option should be validated
as part of the parsing, so no invalid (or empty)
values should be set.
- Put variables closer to where they're used, and
pre-allocate slices.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This file is only used as default if no version is specified. We
should probably get rid of this, but let's update it to better
reflect the version that developer builds are building.
d48fb9f9f7/docker.Makefile (L22)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Rewrite the function to use strings.Cut instead of checking for,
and trimming prefixes for each option.
- More explicitly set the value, instead of setting an empty value,
then propagating the struct.
- Define a "type" to provide a more enum-like construct.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
notable changes;
- api/types/container: move StateStatus, NewStateStatus internal again
- daemon/server/httputils: remove ContainerDecoder interface
- runconfig: move to daemon/internal/runconfig
- pkg/stack: move to daemon/internal
- remove pkg/stringid as it has moved to the client module
- remove pkg/stdcopy as it has moved to the api module
- pkg/rootless: move to daemon/internal
- move api/types/plugins/logdriver to daemon/internal
- move api/types/plugins/logdriver to daemon/internal
- pkg/system: move to daemon/internal
- remove pkg/fileutils
full diff: 25e2b4d485...2574c2b2e9
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Removed the error return from the `ParseRepositoryInfo` function.
There are no validation steps inside `ParseRepositoryInfo` which
could cause an error, so we always returned a nil error.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
We transitioned most functionality of docker/errdefs to containerd
errdefs module, and the docker/errdefs package should no longer be
used.
Because of that, there will no longer be ambiguity, so we can remove
the aliases for this package, and use it as "errdefs".
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It it just a chain of `New("sometag").Parse(...)`, and most of our
uses don't use a tag for the template, so can call Parse.
There's no public users of this function, but deprecating it first
just in case.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- fix various unhandled errors
- remove some locally defined option-types in favor of option-types
defined by the client / api
- don't use unkeyed structs in tests, and add docs for some subtests
- fix some values in tests that triggered "spellcheck" warnings
- inline vars / functions that only had a single use.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Moves ParseVolume() to a new internal package to remove the dependency
on cli/compose/loader in cli/command/container/opts.go
refactor to keep types isolated
- rename the package to "volumespec" to reuse the name of the package
as part of the name (parsevolume.ParseVolume() -> volumespec.Parse())
- move the related compose types to the internal package as well,
and rename them to be more generic (not associated with "compose");
- ServiceVolumeConfig -> VolumeConfig
- ServiceVolumeBind -> BindOpts
- ServiceVolumeVolume -> VolumeOpts
- ServiceVolumeImage -> ImageOpts
- ServiceVolumeTmpfs -> TmpFsOpts
- ServiceVolumeCluster -> ClusterOpts
- alias the internal types inside cli/compose/types to keep backward
compatibility (for any external consumers); even though the implementation
is internal, Go allows aliasing types to use them externally.
Signed-off-by: Michael Tews <michael@tews.dev>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
internal/registry/errors.go:26:43: use-any: since Go 1.18 'interface{}' can be replaced by 'any' (revive)
func invalidParamf(format string, args ...interface{}) error {
^
internal/registry/registry_mock_test.go:52:51: use-any: since Go 1.18 'interface{}' can be replaced by 'any' (revive)
func writeResponse(w http.ResponseWriter, message interface{}, code int) {
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The registry.ServiceConfig struct in the API types was meant for the
registry configuration on the daemon side; it has variuos fields we
don't use, defines methods for (un)marshaling JSON, and a custom version
of `net.IPNet`, also to (un)marshal JSON.
None of that is needed, so let's change it to a local type, and implement
a constructor (as we now only have "insecure registries" to care
about).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It was written to be used as validate-func for command-line flags, which
we don't use it for (which for CLI-flags includes normalizing the value).
The validation itself didn't add much; it only checked the registry didn't
start or end with a hyphen (which would still fail when parsing).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
now that we no longer need to account for mirrors, these were
identical, so just use a single one.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The CLI does not have information about mirrors, and doesn't
configure them, so we can remove these parts.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This adds an internal fork of [github.com/docker/docker/registry], taken
at commit [moby@f651a5d]. Git history was not preserved in this fork,
but can be found using the URLs provided.
This fork was created to remove the dependency on the "Moby" codebase,
and because the CLI only needs a subset of its features. The original
package was written specifically for use in the daemon code, and includes
functionality that cannot be used in the CLI.
[github.com/docker/docker/registry]: https://pkg.go.dev/github.com/docker/docker@v28.3.2+incompatible/registry
[moby@49306c6]: 49306c607b/registry
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This was deprecated in 6f46cd2f4b,
which is part of v28.x, and no longer used, so we can remove it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This utility was only used in the CLI, but the implementation was
based on it being used on the daemon side, so included resolving
the host's IP-address, mirrors, etc.
The only reason it's used in the CLI is to provide credentials for
the registry that's being searched, so reduce it to just that.
There's more cleaning up to do in this area, so to make our lives
easier, it's implemented locally as non-exported functions; likely
to be replaced with something else.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Docker Content Trust is currently only implemented for the classic
builder, but is known to not work with multi-stage builds, and
requires rewriting the Dockerfile, which is brittle because the
Dockerfile syntax evolved with the introduction of BuildKit as
default builder.
Given that the classic builder is deprecated, and only used for
Windows images, which are not verified by content trust;
# docker pull --disable-content-trust=false mcr.microsoft.com/windows/servercore:ltsc2025
Error: remote trust data does not exist for mcr.microsoft.com/windows/servercore: mcr.microsoft.com does not have trust data for mcr.microsoft.com/windows/servercore
With content trust not implemented in BuildKit, and not implemented
in docker compose, this resulted in an inconsistent behavior.
This patch removes content-trust support for "docker build". As this
is a client-side feature, users who require this feature can still
use an older CLI to to start the build.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
gitutils_test.go:210: git-http-backend: git: 'http-backend' is not a git command. See 'git --help'.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Adds a local fork of this package for use in the classic builder.
Code was taken at commit [d33d46d01656e1d9ee26743f0c0d7779f685dd4e][1].
Migration was done using the following steps:
# install filter-repo (https://github.com/newren/git-filter-repo/blob/main/INSTALL.md)
brew install git-filter-repo
# create a temporary clone of docker
cd ~/Projects
git clone https://github.com/docker/docker.git build_context_temp
cd build_context_temp
# commit taken from
git rev-parse --verify HEAD
d33d46d01656e1d9ee26743f0c0d7779f685dd4e
git filter-repo --analyze
# remove all code, except for the remotecontext packages, and move to build/internal docs and previous locations of it
git filter-repo \
--path 'builder/remotecontext/git' \
--path 'builder/remotecontext/urlutil' \
--path-rename builder/remotecontext:cli/command/image/build/internal
# go to the target repository
cd ~/go/src/github.com/docker/cli
# create a branch to work with
git checkout -b fork_remotecontext
# add the temporary repository as an upstream and make sure it's up-to-date
git remote add build_context_temp ~/Projects/build_context_temp
git fetch build_context_temp
# merge the upstream code
git merge --allow-unrelated-histories --signoff -S build_context_temp/master
[1]: https://github.com/docker/docker/d33d46d01656e1d9ee26743f0c0d7779f685dd4e
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Removes direct imports of github.com/docker/docker/builder in
the image package, to be moved later.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- use consistent name for MountOpt vars
- cleanup some comments and make them a GoDoc
- remove import alias
- use subtests for tests that were prepared for it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Also rename "ctx" argument; we shouldn't use this as name for things
that are not a context.Context.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It's part of the presentation logic of the cli, and only used internally.
We can consider providing utilities for these, but better as part of
separate packages.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch deprecates the unused `RegistryAuthenticationPrivilegedFunc`.
The function would prompt the user when the registry returns a 403 after trying
the initial auth value set in `RegistryAuth`.
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
docker build --call outline .
TARGET: binary
BUILD ARG VALUE DESCRIPTION
BASE_VARIANT alpine
ALPINE_VERSION 3.21 sets the version of the alpine base image to use, including for the golang image.
GO_VERSION 1.24.5
XX_VERSION 1.6.1
GOVERSIONINFO_VERSION v1.4.1
GO_LINKMODE static defines if static or dynamic binary should be produced
GO_BUILDTAGS defines additional build tags
GO_STRIP strips debugging symbols if set
CGO_ENABLED manually sets if cgo is used
VERSION sets the version for the produced binary
PACKAGER_NAME sets the company that produced the windows binary
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This patch removes the interactive prompts from `docker push/pull`.
The prompt would only execute on a response status code 403 from the registry
after trying the value set in `RegistryAuth`. Docker Hub could return 404
instead or 429, which would never execute the prompt.
The UX regarding the prompt is also questionable since the user might
not actually want to authenticate with a registry and the CLI could fail fast
instead. The user can always run `docker login` or set the `DOCKER_AUTH_CONFIG`
environment variable to get authenticated.
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
client.ContainerDiff already validates the given container name/ID, and
produces an error when empty, so we don't have to check for this;
abba330bbf/client/container_diff.go (L13-L16)
While updating, also;
- remove the diffOptions type, as there were no other options, and make
the container name/ID a string argument.
- fix camelCase nameing of dockerCLI
Before this patch:
docker diff ""
Container name cannot be empty
With this patch:
docker diff ""
invalid container name or ID: value is empty
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
It was the only utility we consumed from the package, and it's trivial
to implement, so let's create local copies of it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This utility was only used for testing, and to generate a random
suffix for Dockerfiles. As we don't need the same contract as
pkg/stringid.GenerateRandomID() (not allow all-numeric IDs as they
would not be usable for hostnames), we can use a local test-utility,
and local implementation for the random suffix instead.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
We were depending on pkg/stringid to truncate IDs for presentation. While
traditionally, we used a fixed length for "truncated" IDs, this is not
a strict requirement (any ID-prefix should work, but conflicts may
happen on shorter IDs).
This patch adds a local `TruncateID()` utility in the formatter package;
it's currently using the same implementation and length as the
`stringid.TruncateID` function, but may diverge in future.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `AllowOverwriteDirWithFile` option was added when reimplementing the
CLI using the API Client lib in [moby@1b2b91b]. Before that refactor, the
`noOverwriteDirNonDir` query argument [would be set unconditionally][1]
by the CLI, with no options to control the behavior.
It's unclear why the `noOverwriteDirNonDir` was implemented as opt-in (not
opt-out), as overwriting a file with a directory (or vice-versa) would
generally be unexpected behavior.
We're considering making `noOverwriteDirNonDir` unconditional on the daemon
side, and to deprecate the `AllowOverwriteDirWithFile` option. This patch
removes its use, as it was set to the default either way, and there's no
options to configure it from the CLI.
[1]: 8c9ad7b818/api/client/cp.go (L345-L346)
[moby@1b2b91b]: 1b2b91ba43
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Make the test slightly more permissive; we're looking for a trailing
newline, not necessarily an empty line.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- https://github.com/golang/go/issues?q=milestone%3AGo1.24.5+label%3ACherryPickApproved
- full diff: https://github.com/golang/go/compare/go1.24.4...go1.24.5
This minor releases include 1 security fixes following the security policy:
- cmd/go: unexpected command execution in untrusted VCS repositories
Various uses of the Go toolchain in untrusted VCS repositories can result in
unexpected code execution. When using the Go toolchain in directories fetched
using various VCS tools (such as directly cloning Git or Mercurial repositories)
can cause the toolchain to execute unexpected commands, if said directory
contains multiple VCS configuration metadata (such as a '.hg' directory in a Git
repository). This is due to how the Go toolchain attempts to resolve which VCS
is being used in order to embed build information in binaries and determine
module versions.
The toolchain will now abort attempting to resolve which VCS is being used if it
detects multiple VCS configuration metadata in a module directory or nested VCS
configuration metadata (such as a '.git' directoy in a parent directory and a
'.hg' directory in a child directory). This will not prevent the toolchain from
building modules, but will result in binaries omitting VCS related build
information.
If this behavior is expected by the user, the old behavior can be re-enabled by
setting GODEBUG=allowmultiplevcs=1. This should only be done in trusted
repositories.
Thanks to RyotaK (https://ryotak.net) of GMO Flatt Security Inc for reporting
this issue.
This is CVE-2025-4674 and https://go.dev/issue/74380.
View the release notes for more information:
https://go.dev/doc/devel/release#go1.24.5
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
I noticed this in a ticket in the compose issue tracker; with debug logging
enabled, the OTEL error-logger may be logging even if there's no error;
DEBU[0000] Executing bake with args: [bake --file - --progress rawjson --metadata-file /tmp/compose-build-metadataFile-1203980021.json --allow fs.read=/home/user/dev/project --allow fs.read=/home/user/dev/project --allow fs.read=/home/user/dev/project/nginx --allow fs.read=/home/user/dev/project]
TRAC[0000] Plugin server listening on @docker_cli_d8df486f78df3b7357995be71bf0cef6
DEBU[0005] otel error error="<nil>"
^CTRAC[0055] Closing plugin server
TRAC[0055] Closing plugin server
DEBU[0055] otel error error="<nil>"
DEBU[0055] otel error error="<nil>"
Update the error-handler to not log if there's no error.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Make it a hidden file. From the [CodeCov docs][1]:
> Can I name the file .codecov.yml?
>
> Yes, you can name the file `codecov.yml` or `.codecov.yml`. However, the
> file must still be located in the repository root, `dev/`, or `.github/`
> directories
[1]: https://docs.codecov.com/docs/codecov-yaml#can-i-name-the-file-codecovyml
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The strslice.StrSlice type is a string-slice with a custom JSON Unmarshal
function to provide backward-compatibility with older API requests (see
[moby@17d6f00] and [moby@ea4a067]).
Given that the type is assigned implicitly through the fields on HostConfig,
we can just use a regular []string instead.
[moby@17d6f00]: 17d6f00ec2
[moby@ea4a067]: ea4a06740b
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Adding image tags that follow the semver major and minor versions (e.g., `28`
and `28.3`) for the moby-bin images.
This makes it easier for users to reference the latest build within a
major or minor version series without having to know the exact
minor/patch version.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
When connecting to a remote daemon through an ssh:// connection,
the CLI connects with the remote host using ssh, executing the
`docker system dial-stdio` command on the remote host to connect
to the daemon API's unix socket.
By default, the `docker system dial-stdio` command connects with the
daemon using the default location (/var/run/docker.sock), or the
location as configured on the remote host.
Commit 25ebf0ec9c (included in docker
CLI v24.0.0-rc.2 and higher) introduced a feature to allow the location
of the socket to be specified through the host connection string, for
example:
DOCKER_HOST='ssh://example.test/run/custom-docker.sock'
The custom path is included as part of the ssh command executed from
the client machine to connect with the remote host. THe example above
would execute the following command from the client machine;
ssh -o ConnectTimeout=30 -T -- example.test docker --host unix:///run/custom-docker.sock system dial-stdio
ssh executes remote commands in a shell environment, and no quoting
was in place, which allowed for a connection string to include additional
content, which would be expanded / executed on the remote machine.
For example, the following example would execute `echo hello > /hello.txt`
on the remote machine;
export DOCKER_HOST='ssh://example.test/var/run/docker.sock $(echo hello > /hello.txt)'
docker info
# (output of docker info from the remote machine)
While this doesn't allow the user to do anything they're not already
able to do so (by directly using the same SSH connection), the behavior
is not expected, so this patch adds quoting to prevent such URLs from
resulting in expansion.
This patch updates the cli/connhelper and cli/connhelper/ssh package to
quote parameters used in the ssh command to prevent code execution and
expansion of variables on the remote machine. Quoting is also applied to
other parameters that are obtained from the DOCKER_HOST url, such as username
and hostname.
- The existing `Spec.Args()` method inthe cli/connhelper/ssh package now
quotes arguments, and returns a nil slice when failing to quote. Users
of this package should therefore check the returned arguments before
consuming. This method did not provide an error-return, and adding
one would be a breaking change.
- A new `Spec.Command` method is introduced, which (unlike the `Spec.Args()`
method) provides an error return. Users are recommended to use this new
method instead of the `Spec.Args()` method.
Some minor additional changes in behavior are included in this patch;
- Connection URLs with a trailing slash (e.g. `ssh://example.test/`)
would previously result in `unix:///` being used as custom socket
path. After this patch, the trailing slash is ignored, and no custom
socket path is used.
- Specifying a remote command is now required. When passing an empty
remote command, `Spec.Args()` now results in a `nil` value to be
returned (or an `no remote command specified` error when using
`Spec.Comnmand()`.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cli/connhelper/internal/syntax/parser.go:31:2: Duplicate words (the) found (dupword)
// Note that it shares some features with Bash, due to the the shared
^
cli/connhelper/internal/syntax/quote.go:48:1: cyclomatic complexity 35 of func `Quote` is high (> 16) (gocyclo)
func Quote(s string, lang LangVariant) (string, error) {
^
cli/connhelper/internal/syntax/quote.go:103:3: shadow: declaration of "offs" shadows declaration at line 56 (govet)
offs := 0
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This was introduced in 9b54d860cd,
which added `docker container remove` as alias for `docker container rm`.
However, due to the `NewRmCommand` being used both for adding the top-level
`docker rm` command and for adding the `docker container rm` command, it
also introduced a (hidden) top-level `docker remove` command;
docker remove --help | head -n1
Usage: docker rm [OPTIONS] CONTAINER [CONTAINER...]
The command was not documented, and did not appear in `--help` output,
nor was auto-complete provided;
docker --help | grep remove
docker r<TAB>
rename (Rename a container) rm (Remove one or more containers) run (Create and run a new container from an image)
restart (Restart one or more containers) rmi (Remove one or more images)
This patch adds a dedicated, non-exported `newRemoveCommand` to add sub-
commands for `docker container`, taking a similar approach as was done in
[moby@b993609d5a] for `docker image rm`.
With this patch applied, the hidden command is no longer there, but
the `docker rm`, `docker container rm`, and `docker container remove`
commands stay functional as intended;
docker remove foo
docker: unknown command: docker remove
Run 'docker --help' for more information
docker rm --help | head -n1
Usage: docker rm [OPTIONS] CONTAINER [CONTAINER...]
docker container rm --help | head -n1
Usage: docker container rm [OPTIONS] CONTAINER [CONTAINER...]
docker container remove --help | head -n1
Usage: docker container rm [OPTIONS] CONTAINER [CONTAINER...]
[moby@b993609d5a]: b993609d5a
Reported-by: Lorenzo Buero <138243046+LorenzoBuero@users.noreply.github.com>
Co-authored-by: Lorenzo Buero <138243046+LorenzoBuero@users.noreply.github.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
When pulling or pushing images, the CLI could prompt for a password
if the push/pull failed and the registry returned a 401 (Unauthorized)
Ironically, this feature did not work when using Docker Hub (and possibly
other registries using basic auth), due to some custom error handling added
in [moby@19a93a6e3d42], which also discards the registry's status code,
changing it to a 404;
curl -v -XPOST --unix-socket /var/run/docker.sock 'http://localhost/v1.50/images/create?fromImage=docker.io%2Fexample%2Fprivate&tag=latest'
...
< HTTP/1.1 404 Not Found
< Content-Type: application/json
...
{"message":"pull access denied for example/private, repository does not exist or may require 'docker login'"}
And due to a bug, other registries (not using basic auth) returned a generic
error, which resulted in a 500 Internal Server Error. That bug was fixed in
docker 28.2, now returning the upstream status code and trigger an interactive
prompt;
docker pull icr.io/my-ns/my-image:latest
Please login prior to pull:
Username:
This prompt would be triggered unconditionally, also if the CLI was run
non-interactively and no TTY attached;
docker pull icr.io/my-ns/my-image:latest < /dev/null
Please login prior to pull:
Username:
With this PR, no prompt is shown ;
# without STDIN attached
docker pull icr.io/my-ns/my-image:latest < /dev/null
Error response from daemon: error from registry: Authorization required. See https://cloud.ibm.com/docs/Registry?topic=Registry-troubleshoot-auth-req - Authorization required. See https://cloud.ibm.com/docs/Registry?topic=Registry-troubleshoot-auth-req
For now, the prompt is still shown otherwise;
docker pull icr.io/my-ns/my-image:latest
Login prior to pull:
Username: ^C
[moby@19a93a6e3d42]: 19a93a6e3d
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Currently, image tree visualization doesn't properly detect chips for
parent images, only looking at child images. This patch fixes the issue
by checking both parent and child images when determining which chips to
display in the tree view.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This patch enables the CLI to natively pick up the `DOCKER_AUTH_CONFIG`
environment variable and use it as a credential store.
The `DOCKER_AUTH_CONFIG` value should be a JSON object and must store
the credentials in a base64 encoded string under the `auth` key.
Specifying additional fields will cause the parser to fail.
For example:
`printf "username:pat" | openssl base64 -A`
`export DOCKER_AUTH_CONFIG='{
"auths": {
"https://index.docker.io/v1/": {
"auth": "aGk6KTpkY2tyX3BhdF9oZWxsbw=="
}
}
}'`
Credentials stored in `DOCKER_AUTH_CONFIG` would take precedence over any
credential stored in the file store (`~/.docker/config.json`) or native store
(credential helper).
Destructive actions, such as deleting a credential would result in a noop if
found in the environment credential. Credentials found in the file or
native store would get removed.
Signed-off-by: Alano Terblanche <18033717+Benehiko@users.noreply.github.com>
This field was added in [moby@5130fe5d38837302e], which
added it for use as intermediate struct when parsing CLI flags (through
`runconfig.ParseExec`) in [moby@c786a8ee5e9db8f5f].
Commit [moby@9d9dff3d0d9e92adf] rewrote the CLI to use
Cobra, and as part of this introduced a separate `execOptions` type in
`api/client/container`, however the ExecOptions.Detach field was still
used as intermediate field to store the flag's value.
Given that the client doesn't use this field, let's remove its use to
prevent giving the impression that it's used anywhere.
[moby@5130fe5d38837302e]: 5130fe5d38
[moby@c786a8ee5e9db8f5f]: c786a8ee5e
[moby@9d9dff3d0d9e92adf]: 9d9dff3d0d
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The RegistryAuthenticationPrivilegedFunc has some conditional logic to
add additional hints when logging in to the default (Docker Hub) registry.
Commit 9f4165ccb8 inadvertently passed the
wrong variable to PromptUserForCredentials, which caused it to show the
additional hints for Docker Hub.
Before this patch, hints were printed for the default (docker hub) registry;
docker pull icr.io/my-ns/my-image:latest
Login prior to pull:
Log in with your Docker ID or email address to push and pull images from Docker Hub. If you don't have a Docker ID, head over to https://hub.docker.com/ to create one.
You can log in with your password or a Personal Access Token (PAT). Using a limited-scope PAT grants better security and is required for organizations using SSO. Learn more at https://docs.docker.com/go/access-tokens/
Username:
With this patch, those hints are omitted;
docker pull icr.io/my-ns/my-image:latest
Login prior to pull:
Username:
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This file is only used as default if no version is specified. We
should probably get rid of this, but let's update it to better
reflect the version that developer builds are building.
d48fb9f9f7/docker.Makefile (L22)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Image config fields like Cmd, Entrypoint, Env, etc. will be omitted from
/images/{name}/json response when empty, starting in v29.0.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
- https://github.com/golang/go/issues?q=milestone%3AGo1.24.4+label%3ACherryPickApproved
- full diff: https://github.com/golang/go/compare/go1.24.3...go1.24.4
This release includes 3 security fixes following the security policy:
- net/http: sensitive headers not cleared on cross-origin redirect
Proxy-Authorization and Proxy-Authenticate headers persisted on cross-origin redirects potentially leaking sensitive information.
Thanks to Takeshi Kaneko (GMO Cybersecurity by Ierae, Inc.) for reporting this issue.
This is CVE-2025-4673 and Go issue https://go.dev/issue/73816.
- os: inconsistent handling of O_CREATE|O_EXCL on Unix and Windows
os.OpenFile(path, os.O_CREATE|O_EXCL) behaved differently on Unix and Windows systems when the target path was a dangling symlink. On Unix systems, OpenFile with O_CREATE and O_EXCL flags never follows symlinks. On Windows, when the target path was a symlink to a nonexistent location, OpenFile would create a file in that location.
OpenFile now always returns an error when the O_CREATE and O_EXCL flags are both set and the target path is a symlink.
Thanks to Junyoung Park and Dong-uk Kim of KAIST Hacking Lab for discovering this issue.
This is CVE-2025-0913 and Go issue https://go.dev/issue/73702.
- crypto/x509: usage of ExtKeyUsageAny disables policy validation
Calling Verify with a VerifyOptions.KeyUsages that contains ExtKeyUsageAny unintentionally disabledpolicy validation. This only affected certificate chains which contain policy graphs, which are rather uncommon.
Thanks to Krzysztof Skrzętnicki (@Tener) of Teleport for reporting this issue.
This is CVE-2025-22874 and Go issue https://go.dev/issue/73612.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
These comments were added to enforce using the correct import path for
our packages ("github.com/docker/docker", not "github.com/moby/moby").
However, when working in go module mode (not GOPATH / vendor), they have
no effect, so their impact is limited.
Remove these imports in preparation of migrating our code to become an
actual go module.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This change eliminates the automatic creation of image tags in the
format `dockereng/cli-bin:sha-ad132f5` for every push.
They're not too useful, produce noise and use a lot of space.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Also renamed some vars for clarity, renamed a error-returns to prevent
shadowing, and fixed some linter warnings about unhandled errors.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
If we have an error type that we're checking a substring against, we
should really be checking using ErrorContains to indicate the right
semantics to assert.
Mostly done using these transforms:
find . -type f -name "*_test.go" | \
xargs gofmt -w -r 'assert.Assert(t, is.ErrorContains(e, s)) -> assert.ErrorContains(t, e, s)'
find . -type f -name "*_test.go" | \
xargs gofmt -w -r 'assert.Assert(t, is.Contains(err.Error(), s)) -> assert.ErrorContains(t, err, s)'
find . -type f -name "*_test.go" | \
xargs gofmt -w -r 'assert.Check(t, is.Contains(err.Error(), s)) -> assert.Check(t, is.ErrorContains(err, s))'
As well as some small fixups to helpers that were doing
strings.Contains explicitly.
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
builder/remotecontext/git/gitutils_test.go:116:3: The copy of the 'for' variable "tc" can be deleted (Go 1.22+) (copyloopvar)
tc := tc
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This is a partial revert of 389ada7188, which
switched from os/exec to the golang.org/x/sys/execabs package to mitigate
security issues (mainly on Windows) with lookups resolving to binaries in the
current directory.
from the go1.19 release notes https://go.dev/doc/go1.19#os-exec-path
> ## PATH lookups
>
> Command and LookPath no longer allow results from a PATH search to be found
> relative to the current directory. This removes a common source of security
> problems but may also break existing programs that depend on using, say,
> exec.Command("prog") to run a binary named prog (or, on Windows, prog.exe) in
> the current directory. See the os/exec package documentation for information
> about how best to update such programs.
>
> On Windows, Command and LookPath now respect the NoDefaultCurrentDirectoryInExePath
> environment variable, making it possible to disable the default implicit search
> of “.” in PATH lookups on Windows systems.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cmd.Environ() is new in go1.19, and not needed for this specific case.
Without this, trying to use this package in code that uses go1.18 will fail;
builder/remotecontext/git/gitutils.go:216:23: cmd.Environ undefined (type *exec.Cmd has no field or method Environ)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Setting cmd.Env overrides the default of passing through the parent
process' environment, which works out fine most of the time, except when
it doesn't. For whatever reason, leaving out all the environment causes
git-for-windows sh.exe subprocesses to enter an infinite loop of
access violations during Cygwin initialization in certain environments
(specifically, our very own dev container image).
Signed-off-by: Cory Snider <csnider@mirantis.com>
While it is undesirable for the system or user git config to be used
when the daemon clones a Git repo, it could break workflows if it was
unconditionally applied to docker/cli as well.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Prevent git commands we run from reading the user or system
configuration, or cloning submodules from the local filesystem.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Keep It Simple! Set the working directory for git commands by...setting
the git process's working directory. Git commands can be run in the
parent process's working directory by passing the empty string.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Make the test more debuggable by logging all git command output and
running each table-driven test case as a subtest.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Older versions of Go don't format comments, so committing this as
a separate commit, so that we can already make these changes before
we upgrade to Go 1.19.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Simplify some of the logic, and add documentation about the package,
as well as warnings that this package should not be used as a general-
purpose utility.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
pkg/urlutil (despite its poorly chosen name) is not really intended as a generic
utility to handle URLs, and should only be used by the builder to handle (remote)
build contexts.
- IsURL() only does a very rudimentary check for http(s):// prefixes, without any
other validation, but due to its name may give incorrect expectations.
- IsGitURL() is written specifically with docker build remote git contexts in
mind, and has handling for backward-compatibility, where strings that are
not URLs, but start with "github.com/" are accepted.
Because of the above, this patch:
- moves the package inside builder/remotecontext, close to where it's intended
to be used (ideally this would be part of build/remotecontext itself, but this
package imports many other dependencies, which would introduce those as extra
dependencies in the CLI).
- deprecates pkg/urlutil, but adds aliases as there are some external consumers.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The io/ioutil package has been deprecated in Go 1.16. This commit
replaces the existing io/ioutil functions with their new definitions in
io and os packages.
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
urlutil.IsUrl() was merely checking if the url had a http(s)://
prefix, which is just as well handled through using url.Parse()
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was only used inside gitutils,
and is written specifically for the requirements
there.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
`docker build` accepts remote repositories
using either the `git://` notation, or `git@`.
Docker attempted to parse both as an URL, however,
`git@` is not an URL, but an argument to `git clone`.
Go 1.7 silently ignored this, and managed to
extract the needed information from these
remotes, however, Go 1.8 does a more strict
validation, and invalidated these.
This patch adds a different path for `git@` remotes,
to prevent them from being handled as URL (and
invalidated).
A test is also added, because there were no
tests for handling of `git@` remotes.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
# - windows-2022 # FIXME: some tests are failing on the Windows runner, as well as on Appveyor since June 24, 2018: https://ci.appveyor.com/project/docker/cli/history
steps:
-
name:Checkout
uses:actions/checkout@v4
uses:actions/checkout@v5
with:
path:${{ env.GOPATH }}/src/github.com/docker/cli
-
name:Set up Go
uses:actions/setup-go@v5
uses:actions/setup-go@v6
with:
go-version:"1.24.3"
go-version:"1.25.4"
-
name:Test
run:|
go test -coverprofile=/tmp/coverage.txt $(go list ./... | grep -vE '/vendor/|/e2e/')
go test -coverprofile=/tmp/coverage.txt $(go list ./... | grep -vE '/vendor/|/e2e/|/cmd/docker-trust')
os-chdir: false # FIXME(thaJeztah):Disable `os.Chdir()` detections; should be automatically disabled on Go < 1.24; see https://github.com/docker/cli/pull/5835#issuecomment-2665302478
@ -221,6 +223,14 @@ linters:
linters:
- staticcheck
# Ignore deprecation linting for cli/command/stack/*.
#
# FIXME(thaJeztah): remove exception once these functions are un-exported or internal; see https://github.com/docker/cli/pull/6389
test-unit:## run unit tests, to change the output format use: GOTESTSUM_FORMAT=(dots|short|standard-quiet|short-verbose|standard-verbose) make test-unit
gotestsum -- $${TESTDIRS:-$(shell go list ./... | grep -vE '/vendor/|/e2e/')}$(TESTFLAGS)
gotestsum -- $${TESTDIRS:-$(shell go list ./... | grep -vE '/vendor/|/e2e/|/cmd/docker-trust')}$(TESTFLAGS)
.PHONY:test-coverage
test-coverage:## run test coverage
mkdir -p $(CURDIR)/build/coverage
gotestsum -- $(shell go list ./... | grep -vE '/vendor/|/e2e/') -coverprofile=$(CURDIR)/build/coverage/coverage.txt
gotestsum -- $(shell go list ./... | grep -vE '/vendor/|/e2e/|/cmd/docker-trust') -coverprofile=$(CURDIR)/build/coverage/coverage.txt
.PHONY:lint
lint:## run all the lint tools
@ -52,7 +52,7 @@ shellcheck: ## run shellcheck validation
.PHONY:fmt
fmt:## run gofumpt (if present) or gofmt
@if command -v gofumpt > /dev/null;then\
gofumpt -w -d -lang=1.23 . ;\
gofumpt -w -d -lang=1.24 . ;\
else\
go list -f {{.Dir}} ./... | xargs gofmt -w -s -d ;\
{name:"builtin alias",c:&fakeCandidate{path:builtinAlias},invalid:`plugin "alias" duplicates an alias of builtin command "builtin"`},
{name:"fetch failure",c:&fakeCandidate{path:goodPluginPath,exec:false},invalid:fmt.Sprintf("failed to fetch metadata: faked a failure to exec %q",goodPluginPath)},
{name:"metadata not json",c:&fakeCandidate{path:goodPluginPath,exec:true,meta:`xyzzy`},invalid:"invalid character"},
{name:"empty schemaversion",c:&fakeCandidate{path:goodPluginPath,exec:true,meta:`{}`},invalid:`plugin SchemaVersion "" is not valid`},
{name:"invalid schemaversion",c:&fakeCandidate{path:goodPluginPath,exec:true,meta:`{"SchemaVersion": "xyzzy"}`},invalid:`plugin SchemaVersion "xyzzy" is not valid`},
{name:"no vendor",c:&fakeCandidate{path:goodPluginPath,exec:true,meta:`{"SchemaVersion": "0.1.0"}`},invalid:"plugin metadata does not define a vendor"},
{name:"empty vendor",c:&fakeCandidate{path:goodPluginPath,exec:true,meta:`{"SchemaVersion": "0.1.0", "Vendor": ""}`},invalid:"plugin metadata does not define a vendor"},
// ReexecEnvvar is the name of an ennvar which is set to the command
// used to originally invoke the docker CLI when executing a
// plugin. Assuming $PATH and $CWD remain unchanged this should allow
// the plugin to re-execute the original CLI.
ReexecEnvvar=metadata.ReexecEnvvar
// ResourceAttributesEnvvar is the name of the envvar that includes additional
// resource attributes for OTEL.
//
// Deprecated: The "OTEL_RESOURCE_ATTRIBUTES" env-var is part of the OpenTelemetry specification; users should define their own const for this. This const will be removed in the next release.
assert.DeepEqual(t,config.MacAddress,tc.expectedCfg.MacAddress)//nolint:staticcheck // ignore SA1019: field is deprecated, but still used on API < v1.44.
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
