Commit Graph

6385 Commits

Author SHA1 Message Date
Vincent Demeester f2123b3fe4
Merge pull request #1749 from thaJeztah/compose_3.8_compose_credential_spec_fix
Update compose 3.8 to disallow additional properties
2019-03-18 16:02:47 +01:00
Silvin Lubecki 0d922266e2
Merge pull request #1397 from thaJeztah/fix-subscription-filter
[master] fix subscription filter
2019-03-18 15:35:08 +01:00
Silvin Lubecki d80e023382
Merge pull request #1750 from thaJeztah/fix_circle_vendor_flakiness
CircleCI: Increase no-output timeout to 15 minutes for vendoring
2019-03-18 15:22:48 +01:00
Sebastiaan van Stijn 69f1727248
Merge pull request #1524 from thaJeztah/fix_system_prune_output
Fix "warning" output on docker system prune
2019-03-18 15:11:22 +01:00
Mason Fish e7176e8dd0
fix subscription filter
Signed-off-by: Mason Fish <mason.fish@docker.com>
(cherry picked from commit a31b20d7db)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 14:57:30 +01:00
Sebastiaan van Stijn 81b319aa5f
Bump Golang 1.12.1
go1.12.1 (released 2019/03/14) includes fixes to cgo, the compiler, the go
command, and the fmt, net/smtp, os, path/filepath, sync, and text/template
packages. See the Go 1.12.1 milestone on our issue tracker for details.

For the relase notes of Go 1.12.0, see: https://golang.org/doc/go1.12

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 14:47:24 +01:00
Silvin Lubecki 7f09b9d8e2
Merge pull request #1742 from thaJeztah/bump_golang_1.11.6
Bump Golang 1.11.6
2019-03-18 14:46:51 +01:00
Sebastiaan van Stijn 26e004797b
Make system prune warning filters human-readable
The warning, printed before running `docker system prune` was printing the
filters in JSON format.

This patch attempts to make the output human readable;

- updating the code, and template to print filters individually
- reducing the indentation (which was quite deep)

Before this patch was applied;

```
docker system prune --filter until=24h --filter label=hello-world --filter label!=foo=bar --filter label=bar=baz

WARNING! This will remove:
        - all stopped containers
        - all networks not used by at least one container
        - all dangling images
        - all dangling build cache
        - Elements to be pruned will be filtered with:
        - label={"label":{"bar=baz":true,"hello-world":true},"label!":{"foo=bar":true},"until":{"24h":true}}
Are you sure you want to continue? [y/N]
```

With this patch applied;

```
WARNING! This will remove:
  - all stopped containers
  - all networks not used by at least one container
  - all dangling images
  - all dangling build cache

  Items to be pruned will be filtered with:
  - label!=foo=bar
  - label!=never=remove-me
  - label=bar=baz
  - label=hello-world
  - label=remove=me
  - until=24h

Are you sure you want to continue? [y/N]
```

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 13:57:18 +01:00
Sebastiaan van Stijn 4a0218bb11
Fix system prune warning missing filters from config-file
The warning, printed before runing docker system prune was missing any filter
that was set in the configuration file. In addition, the warning prefixes the
filters with `label=`, which is no longer accurate, now that the prune command
also supports "until" as a filter.

Before this change, only the filters set on the command-line were shown,
and any filter set in the configuration file was missing;

```
mkdir -p ./test-config
echo '{"pruneFilters": ["label!=never=remove-me", "label=remove=me"]}' > test-config/config.json
docker --config=./test-config system prune --filter until=24h --filter label=hello-world --filter label!=foo=bar --filter label=bar=baz

WARNING! This will remove:
        - all stopped containers
        - all networks not used by at least one container
        - all dangling images
        - all dangling build cache
        - Elements to be pruned will be filtered with:
        - label={"label":{"bar=baz":true,"hello-world":true},"label!":{"foo=bar":true},"until":{"24h":true}}
Are you sure you want to continue? [y/N]
```

With this patch applied, both options from the commandline and options set
in the configuration file are shown;

```
mkdir -p ./test-config
echo '{"pruneFilters": ["label!=never=remove-me", "label=remove=me"]}' > test-config/config.json
docker --config=./test-config system prune --filter until=24h --filter label=hello-world --filter label!=foo=bar --filter label=bar=baz

WARNING! This will remove:
        - all stopped containers
        - all networks not used by at least one container
        - all dangling images
        - all dangling build cache
        - Elements to be pruned will be filtered with:
        - filter={"label":{"bar=baz":true,"hello-world":true,"remove=me":true},"label!":{"foo=bar":true,"never=remove-me":true},"until":{"24h":true}}
```

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 13:57:09 +01:00
Sebastiaan van Stijn 80918147ff
Merge pull request #1748 from silvin-lubecki/cosmetics
Add some spaces in docs for cosmetics and readability reasons.
2019-03-18 13:50:32 +01:00
Silvin Lubecki 9e9fbf0699
Merge pull request #1719 from jcsirot/fix-default-context
Makes the default docker context behavior consistent
2019-03-18 13:48:38 +01:00
Sebastiaan van Stijn dba90e4999
CircleCI: Increase no-output timeout to 15 minutes for vendoring
Vendoring can take some time, depending on network-speed, so
reduce flakiness by increasing the default timeout, to prevent:

    make[1]: Entering directory '/go/src/github.com/docker/cli'
    rm -rf vendor
    bash -c 'vndr |& grep -v -i clone'
    2019/03/18 11:38:26 Collecting initial packages
    Too long with no output (exceeded 10m0s)

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 13:07:22 +01:00
Sebastiaan van Stijn 70846619a9
Update compose 3.8 to disallow additional properties
This was added in other schemas in 1e99ed3ca3,
but not coppied to version 3.8

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-18 12:56:16 +01:00
Silvin Lubecki 8401c81b46 Add some spaces for cosmetics and readability reasons.
Signed-off-by: Silvin Lubecki <silvin.lubecki@docker.com>
2019-03-18 12:32:48 +01:00
Silvin Lubecki cfd5b16ae0
Merge pull request #1745 from ijc/disable-dial-stdio-for-plugins
cli-plugins: disable use of dial-stdio
2019-03-18 12:29:41 +01:00
Ian Campbell ff2ed6efa8 cli-plugins: disable use of dial-stdio
Since #1654 so far we've had problems with it not working on Windows (npipe
lacked the `CloseRead` method) and problems with using tcp with tls (the tls
connection also lacks `CloseRead`). Both of these were workedaround in #1718
which added a nop `CloseRead` method.

However I am now seeing hangs (on Windows) where the `system dial-stdio`
subprocess is not exiting (I'm unsure why so far).

I think the 3rd problem found with this is an indication that `dial-stdio` is
not quite ready for wider use outside of its initial usecase (support for
`ssh://` URLs to connect to remote daemons).

This change simply disables the `dial-stdio` path for all plugins. However
rather than completely reverting 891b3d953e ("cli-plugins: use `docker system
dial-stdio` to call the daemon") I've just disabled the functionality at the
point of use and left in a trap door environment variable so that those who
want to experiment with this mode (and perhaps fully debug it) have an easier
path do doing so.

The e2e test for this case is disabled unless the trap door envvar is set. I
also renamed the test to clarify that it is about cli plugins.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-18 10:58:51 +00:00
Jean-Christophe Sirot b3aa17187f Make default context behaves like a real context:
- when using "--context default" parameter
- when printing the list of contexts
- when exporting the default context to a tarball

Signed-off-by: Jean-Christophe Sirot <jean-christophe.sirot@docker.com>
(+1 squashed commit)
Squashed commits:
[20670495] Fix CLI initialization for the `docker stack deploy --help` command and ensure that the dockerCli.CurrentContext() always returns a non empty context name (default as a fallback)
Remove now obsolete code handling empty string context name
Minor code cleanup

Signed-off-by: Jean-Christophe Sirot <jean-christophe.sirot@docker.com>
2019-03-18 11:45:46 +01:00
Sebastiaan van Stijn 86a5a489f7
Merge pull request #1690 from jcsirot/fix-contextstore-for-plugins
Always initialize context store
2019-03-18 11:27:07 +01:00
Silvin Lubecki 33c8c0543b
Merge pull request #1629 from thaJeztah/update_rmi_docs
Improve `docker image rm` reference docs
2019-03-18 10:39:45 +01:00
Sebastiaan van Stijn 1500105975
Bump Golang 1.11.6
go1.11.6 (released 2019/03/14) includes fixes to cgo, the compiler, linker,
runtime, go command, and the crypto/x509, encoding/json, net, and net/url
packages. See the Go 1.11.6 milestone on our issue tracker for details:

https://github.com/golang/go/issues?q=milestone%3AGo1.11.6

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
2019-03-17 16:01:58 +01:00
Silvin Lubecki 651ccc0711
Merge pull request #1713 from thaJeztah/fix_plugin_test
Fix: plugin-tests discarding current environment
2019-03-14 16:22:05 +01:00
Sebastiaan van Stijn 7f1176b8aa
Merge pull request #1737 from ijc/plugins-using-PersistentPreRunE
Fix regression for CLI plugins using PersistentPreRunE
2019-03-14 15:51:13 +01:00
Ian Campbell 3af168c7df Ensure plugins can use PersistentPreRunE again.
I got a bit carried away in d4ced2ef77 ("allow plugins to have argument
which match a top-level flag.") and broke the ability of a plugin to use the
`PersistentPreRun(E)` hook on its top-level command (by unconditionally
overwriting it) and also broke the plugin framework if a plugin's subcommand
used those hooks (because they would shadow the root one). This could result in
either `dockerCli.Client()` returning `nil` or whatever initialisation the
plugin hoped to do not occuring.

This change revert the relevant bits and reinstates the requirement that a
plugin calls `plugin.PersistentPreRunE` if it uses that hook itself.

It is at least a bit nicer now since we avoid the need for the global struct
since the interesting state is now encapsulated in `tcmd` (and the closure).

In principal this could be done even more simply (by calling `tcmd.Initialize`
statically between `tcmd.HandleGlobalFlags` and `cmd.Execute`) however this has
the downside of _always_ initialising the cli (and therefore dialing the
daemon) even for the `docker-cli-plugin-metadata` command but also for the
`help foo` and `foo --help` commands (Cobra short-circuits the hooks in this
case).

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-14 14:29:09 +00:00
Ian Campbell 0b794e0620 e2e/cli-plugins: explicitly check that PersistentPreRunE works
I regressed this in d4ced2ef77 ("allow plugins to have argument which match a
top-level flag.") by unconditionally overwriting any `PersistentRunE` that the
user may have supplied.

We need to ensure two things:

1. That the user can use `PersistentRunE` (or `PersistentRun`) for their own
   purposes.
2. That our initialisation always runs, even if the user has used
   `PersistentRun*`, since that will shadow the root.

To do this add a `PersistentRunE` to the helloworld plugin which logs (covers 1
above) and then use it when calling the `apiversion` subcommand (which covers 2
since that uses the client)

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-14 14:20:19 +00:00
Tibor Vass c3fc547cc9
Merge pull request #1712 from thaJeztah/fix_test_for_go_1.12
Fix test for Go 1.12.x
2019-03-13 13:38:57 -07:00
Vincent Demeester c748c850f6
Merge pull request #1674 from ericcurtin/patch-1
Add exit status to docker exec manpage
2019-03-13 18:56:48 +01:00
Silvin Lubecki 237bdbf5f6
Merge pull request #1722 from ijc/plugins-global-arg-clash
Allow plugins to have --flags which are the same as the top-level
2019-03-13 15:55:02 +01:00
Jean-Christophe Sirot a1af6e261f Cover the changes with unit test
Signed-off-by: Jean-Christophe Sirot <jean-christophe.sirot@docker.com>
2019-03-13 14:18:41 +01:00
Jean-Christophe Sirot 37fcaf7a29 Resolve the docker Endpoint even if the client already exists. In that case the `TestDialStdio` e2e test had to be modified: the `--tls` option triggers an error since the endpoint resolution tries to read the `${DOCKER_CERT_PATH}/ca.pem` file which does not exist.
Signed-off-by: Jean-Christophe Sirot <jean-christophe.sirot@docker.com>
2019-03-13 14:18:41 +01:00
Jean-Christophe Sirot 3b26cfce8b Always initialize context store
Signed-off-by: Jean-Christophe Sirot <jean-christophe.sirot@docker.com>
2019-03-13 14:18:41 +01:00
Ian Campbell e824bc86f3 Use a copy of root flagset in `HandleGlobalFlags`
This makes things more idempotent, rather than relying on undoing the
interspersed settings.

Note that the underlying `Flag`s remain shared, it's just the `FlagSet` which
is duplicated.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-13 11:28:17 +00:00
Ian Campbell 2c624e8984 Add e2e test for handling of `version`/`-v`/`--version` with plugins
Previous commits fixed the first issue on #1661, this simply adds a test for
it. Note that this is testing the current behaviour, without regard for the
second issue in #1661 which proposes a different behaviour.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-13 11:28:17 +00:00
Ian Campbell d4ced2ef77 allow plugins to have argument which match a top-level flag.
The issue with plugin options clashing with globals is that when cobra is
parsing the command line and it comes across an argument which doesn't start
with a `-` it (in the absence of plugins) distinguishes between "argument to
current command" and "new subcommand" based on the list of registered sub
commands.

Plugins breaks that model. When presented with `docker -D plugin -c foo` cobra
parses up to the `plugin`, sees it isn't a registered sub-command of the
top-level docker (because it isn't, it's a plugin) so it accumulates it as an
argument to the top-level `docker` command. Then it sees the `-c`, and thinks
it is the global `-c` (for AKA `--context`) option and tries to treat it as
that, which fails.

In the specific case of the top-level `docker` subcommand we know that it has
no arguments which aren't `--flags` (or `-f` short flags) and so anything which
doesn't start with a `-` must either be a (known) subcommand or an attempt to
execute a plugin.

We could simply scan for and register all installed plugins at start of day, so
that cobra can do the right thing, but we want to avoid that since it would
involve executing each plugin to fetch the metadata, even if the command wasn't
going to end up hitting a plugin.

Instead we can parse the initial set of global arguments separately before
hitting the main cobra `Execute` path, which works here exactly because we know
that the top-level has no non-flag arguments.

One slight wrinkle is that the top-level `PersistentPreRunE` is no longer
called on the plugins path (since it no longer goes via `Execute`), so we
arrange for the initialisation done there (which has to be done after global
flags are parsed to handle e.g. `--config`) to happen explictly after the
global flags are parsed. Rather than make `newDockerCommand` return the
complicated set of results needed to make this happen, instead return a closure
which achieves this.

The new functionality is introduced via a common `TopLevelCommand` abstraction
which lets us adjust the plugin entrypoint to use the same strategy for parsing
the global arguments. This isn't strictly required (in this case the stuff in
cobra's `Execute` works fine) but doing it this way avoids the possibility of
subtle differences in behaviour.

Fixes #1699, and also, as a side-effect, the first item in #1661.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-13 11:28:17 +00:00
Ian Campbell 8289ae03f8 Add e2e tests for plugin vs global argument issues
These won't pass right now due to https://github.com/docker/cli/issues/1699
("Plugins can't re-use the same flags as cli global flags") and the change in
935d47bbe9 ("Ignore unknown arguments on the top-level command."), but the
intention is to fix them now.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-13 10:54:31 +00:00
Ian Campbell 4e5f0af9cd e2e: start a new file for cli plugins arguments tests
`TestRunGoodArgument` fits here.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-13 10:54:31 +00:00
Sebastiaan van Stijn d6a230606c
Merge pull request #1718 from ijc/dial-stdio-npipe-on-windows
dial-stdio: handle connections which lack CloseRead method.
2019-03-13 11:49:25 +01:00
Stephane Jeandeaux 05674a5096 [pretty print] pretty print and healthcheck
fixes #117

Print healthcheck information in pretty mode.

Signed-off-by: Stephane Jeandeaux <stephane.jeandeaux@gmail.com>
2019-03-12 22:00:46 -04:00
Tibor Vass 81ac432cc2
Merge pull request #1700 from thaJeztah/update_engine
Update docker/docker 8aca18d, containerd v1.2.4
2019-03-12 10:41:51 -07:00
Ian Campbell 0449ad8d06 Revert "Disable `docker system dial-stdio` on Windows"
This reverts commit c41c23813c.

This case is now handled due to the previous commit.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-12 14:57:52 +00:00
Ian Campbell 186e7456ac dial-stdio: Close the connection
This was leaking the fd.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-12 14:57:40 +00:00
Ian Campbell 8919bbf04d dial-stdio: handle connections which lack CloseRead method.
This happens on Windows when dialing a named pipe (a path which is used by CLI
plugins), in that case some debugging shows:

    DEBU[0000] conn is a *winio.win32MessageBytePipe
    DEBU[0000] conn is a halfReadCloser: false
    DEBU[0000] conn is a halfWriteCloser: true
    the raw stream connection does not implement halfCloser
In such cases we can simply wrap with a nop function since closing for read
isn't too critical.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-12 14:52:42 +00:00
Sebastiaan van Stijn bf4a96e564
Merge pull request #1688 from luoyunpeng/optimize-blockIOTypecheck
use char to check blockIO type
2019-03-12 10:32:31 +01:00
Ian Campbell e1a7b56308 Allow control over dirs passed to e2e and unit tests
Allows e.g.:

    $ make -f docker.Makefile TESTDIRS=./cli/command/trust/... test-unit

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-11 16:02:55 +00:00
Ian Campbell 023559b98c Allow flags to be passed to e2e and unit tests
This allows e.g.

    $ make -f docker.Makefile fmt test-e2e-non-experimental TESTFLAGS="-test.run TestRunGoodArgument|TestRunGood"

As well as adding the var to `$(ENVVARS)` we also need to use that when
invocking the e2e image, the existing `$(DOCKER_RUN)` is not used here because
the bind mounts differ. The other variables included in `$(ENVVARS)` are
harmless when running the e2e tests.

The `${TESTFLAGS}` envvar is already understood by `scripts/test/e2e/run`.

Note that since this modifies `$(ENVVARS)` this is now also available to the unit
test target too, so add the use to the invocation so it takes effect.

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-11 15:59:27 +00:00
Silvin Lubecki 6d2d597a6d
Merge pull request #1728 from ijc/e2e-handle-alpine-bump
Fixes for e2e testing after Alpine bump
2019-03-11 15:59:17 +01:00
Ian Campbell 0b0c57871a e2e: avoid `usermod -p` by using `useradd`'s `--password` option
Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-11 14:25:41 +00:00
Ian Campbell e854a9cf96 e2e: Expand `useradd`'s `-m` otion into `--create-home`
... for improved readability

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-11 14:25:41 +00:00
Ian Campbell 5de2d9e8a9 e2e Use `useradd`'s `--shell` option
... in preference to `chsh`, since in recent alpine 3.9.2 images that can fail
with:

    Password: chsh: PAM: Authentication token manipulation error

Which seems to relate to the use of `!` as the password for `root` in `/etc/shadow`gq

Signed-off-by: Ian Campbell <ijc@docker.com>
2019-03-11 14:25:19 +00:00
Silvin Lubecki b86bff84b6
Merge pull request #1710 from ijc/no-dial-stdio-on-windows
Disable `docker system dial-stdio` on Windows
2019-03-11 15:14:23 +01:00
Elliot Luo 0bb397f9ef use char to check blockIO type
Signed-off-by: Elliot Luo <956941328@qq.com>
2019-03-11 10:01:22 +08:00