Continuous Integration
devsy ci starts a devcontainer workspace and runs a command in it. Use it to run tests inside the container and check the build of a newly created workspace.
If you omit the workspace argument and select an existing workspace interactively, Devsy can delete that workspace after the run, including its workspace data. Pass --keep when selecting a workspace interactively to prevent cleanup.
How it works
- Devsy resolves a workspace from the given source. With no source, it prompts for a workspace in an interactive terminal.
- Devsy builds and starts the devcontainer like
devsy workspace up, without IDE or SSH config changes. - Devsy runs the command inside the container.
- Unless you pass
--keep, Devsy attempts cleanup when its initial lookup finds no existing workspace, even when the command fails. This can delete an existing workspace selected interactively after that lookup.
A failing command fails the CI job. Workspace setup or cleanup errors also cause devsy ci to fail.
Usage
devsy ci [flags] [workspace-path|workspace-name] -- <cmd> [args...]Run the test suite inside the devcontainer for the current project:
devsy ci . -- make testTarget a specific project folder and pass environment variables:
devsy ci ./service -- npm test
devsy ci --remote-env CI=true --remote-env NODE_ENV=test . -- npm testSelect a named config, such as a .devcontainer/<name> profile:
devsy ci --devcontainer id:ci . -- make testKeep the workspace after the run, for debugging:
devsy ci --keep . -- ./run-integration-tests.shCommon flags
Run devsy ci --help for the full list. Common flags:
--keepkeeps the workspace after the run, whether the command succeeds or fails.--remote-env KEY=VALUEsets an environment variable in the container at run time. Repeatable.--devcontainerpicks the config:none,image:<ref>,id:<name>, or a path to adevcontainer.json.--cache-fromand--no-cachecontrol the image build cache.--platformruns the container under another platform, such aslinux/amd64.--secret,--env,--build-secret, and--git-tokeninject stored Devsy secrets and variables. See Secrets.
Control progress output
--no-status hides Devsy's live progress messages. It does not filter
diagnostics or the command's own output.
| Output | Effect of --no-status |
|---|---|
| Devsy progress | Hidden, including progress from the nested workspace command |
| Devsy diagnostics | Still controlled by --quiet, --log-level, and diagnostic format |
| Command stdout and stderr | Preserved on their original streams |
| Recorded events | Still recorded where the command already records them |
For example, keep warnings and errors while hiding progress:
devsy ci --no-status --result-format=plain --log-level=warn . -- bash .tools/test.shTo show only errors, use --quiet instead. It takes precedence over
--log-level=warn and suppresses warnings too:
devsy ci --no-status --result-format=plain --quiet . -- bash .tools/test.shWhen stdout is redirected, use --result-format=plain to keep the success
result JSON off stderr. Use --result-format=json when the caller needs the
structured result.
Setup history after cleanup
Devsy keeps the build and lifecycle hook history in the local store after workspace cleanup. Read it by run ID after the job ends, or from an artifact step on the same machine:
devsy workspace logs --source setup --list
devsy workspace logs --source setup --run RUN_ID --context CONTEXT --result-format=plainRetention is limited. See Setup history. The history covers the build and lifecycle hooks. The stdout and stderr of your command after -- stay on their normal streams. Devsy does not wait for hooks still running when the command ends.
Hooks cannot ask for input
CI jobs have no terminal to answer a prompt. devsy ci has no --interactive-post-start option. Write hooks that take their values from the job instead.
Desktop launches, --detach, JSON output, and management or platform tasks also stay noninteractive. If you pass --interactive-post-start to devsy workspace up in one of these runs, Devsy rejects it with an error. It does not wait for input that nobody can see.
Pass values to the workspace in one of these ways:
- Environment variables:
devsy ci --workspace-env NAME=value . -- make test. Use--workspace-envfor values that hooks need. The--remote-envflag applies only to the command that CI runs. - Secrets:
--secretinjects a stored Devsy secret. See Secrets. - Files: write the value to a file that the hook reads.
- Arguments: put a value that is not secret in the hook command, such as
"postStartCommand": ["./start.sh", "example-name"].
Moving the prompt to postAttachCommand does not provide input. postAttachCommand runs without input in every mode.
Build and push images
To build and publish a devcontainer image for reuse as a build cache, run devsy workspace build:
devsy workspace build --repository ghcr.io/my-org/my-devcontainer --tag latest --pushLater CI jobs can pass that image to --cache-from.
GitHub Actions
The workflow installs the CLI, sets up the Docker provider, and runs devsy ci:
name: Devcontainer CI
on:
push:
branches: [main]
pull_request:
jobs:
devcontainer:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Install Devsy
run: |
sudo curl -fsSL https://github.com/devsy-org/devsy/releases/latest/download/devsy-linux-amd64 -o /usr/local/bin/devsy
sudo chmod +x /usr/local/bin/devsy
- name: Run CI in devcontainer
run: |
devsy provider add docker && devsy provider use docker
devsy ci . -- make test