* Print immutable action package details in set up job logs
* "Source commit SHA" instead of "Commit SHA" for immutable actions logs
---------
Co-authored-by: Thomas Boop <[email protected]>
* fix: add `git` and `curl` to default packages installed
Hi 👋🏼
These packages are used in a ton of actions on the marketplace. It would be nice if they were installed and ready for use instead of having to install them with `apt-get` on every single Github workflow.
* Update Dockerfile
* Update images/Dockerfile
Co-authored-by: Guillermo Caracuel <[email protected]>
* Update images/Dockerfile
Co-authored-by: Tingluo Huang <[email protected]>
---------
Co-authored-by: Guillermo Caracuel <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* redirect supported OS doc to public docs
* Anchor to appropriate OS heading
Co-authored-by: Patrick Ellis <[email protected]>
---------
Co-authored-by: Patrick Ellis <[email protected]>
* Upgrade node20: 20.8.1 → 20.13.1
* Call out the release process for `alpine_nodejs` in a comment
* move the comment to the end of the line so it's more obvious which variable it's talking about
* backoff if we retried polling for more than 50 times in less than 30minutes
* run dotnet format
* move delay to after no message trace
* run dotnet format
* Add new SessionConflict return code
* formatting
* Change return type of CreateSessionAsync to new enum
* Update entry scripts to handle new exit code
* Move enum
* consume pipelines service url if present
updates how the `ACTIONS_RUNTIME_URL` variable is set to utilize a new value, `PipelinesServiceUrl` if present in the endpoint. if this value is not present then the existing system connection endpoint is used to retain backward compatibility.
* consume pipelines url
updates how the `ACTIONS_RUNTIME_URL` variable is set to utilize a new value, `PipelinesServiceUrl` if present in the endpoint. if this value is not present then the existing system connection endpoint is used to retain backward compatibility.
This is required in order to ssh into a codespace for the repo:
> error getting ssh server details: failed to start SSH server:
Please check if an SSH server is installed in the container.
If the docker image being used for the codespace does not have an SSH server,
install it in your Dockerfile or, for codespaces that use Debian-based images,
you can add the following to your devcontainer.json:
>
> "features": {
> "ghcr.io/devcontainers/features/sshd:1": {
> "version": "latest"
> }
> }
* add content-type to block blob upload
* Add content-type for sdk path
* fix spacing
* merge headers and only when file extension is .txt
* add conditions
* tweak conditions and path matching
* pass in headers
* add content-type for appendblob
* Set an explicit timeout on ResultsHttpClient
* Hook up retry and standard message handler to ResultsHttpClient
* Remove explicit timeout constructor
* Fix linter
* Bump directly dotnet vulnerable packages
* Use just single file sdk upgrade for vulnerable dependencies
* Save with UTF8-BOM
* Trim down sdk to only Sdk.csproj upgrade
---------
Co-authored-by: Ferenc Hammerl <[email protected]>
* Mark action download failures as infra failures
* Fix Windows variable name
* Update src/Runner.Worker/ActionManager.cs
Co-authored-by: Thomas Boop <[email protected]>
* Apply suggestions from code review
Co-authored-by: Thomas Boop <[email protected]>
---------
Co-authored-by: Thomas Boop <[email protected]>
* Add in workflow to automatically generate dotnet SDK patches
* Update workflow to move the creation of the PR until all the hashes have been compiled and added to the path
* Update upgrade to run not in parallel to prevent git command overlaps
* Update for tings feedback and do better user error handling
* JSON spec doesn't have comments
* Has to use the outputs to proxy variables between the runs
* Wrong output variable
* Be more explicit on the branch check
* Fix comments and json output
* Missed variable name in rename
* Fix race condition on jq write
* Update for comments and more optimized hash updates
* Need to forcibly exit for state to be detected
* Fixed the failure to use a variable instead
* hash files translation works with host context translated
* Change file encoding to utf8 for PipelineTemplateConstants.cs
* infer the feature flag on hash files based on the existence of the context
* encoded utf8 with bom
* rename host-workspace to host-work-directory and feature flag the new context var
* Added comment to explain ignore host work directory resolution
Co-authored-by: fhammerl <[email protected]>
* trigger pipeline
* Move to the github host-workspace context instead of the runner host-work-directory
* remove unused imports
---------
Co-authored-by: fhammerl <[email protected]>
* Add 'http://' to http(s)_proxy if there is no protocol
Before this commit, these URIs used to be ignored (nullproxy).
The new behaviour aligns with go style http(s)_proxy
* Update src/Runner.Sdk/RunnerWebProxy.cs
Co-authored-by: Nikola Jokic <[email protected]>
* Assert that original protocol is preserved
* Use startsWith for testing protocol
* Only modify proxy if useful
---------
Co-authored-by: Nikola Jokic <[email protected]>
* Fix `uses: docker://image:tag` steps when container hook is used
* Update src/Runner.Worker/ActionManager.cs
---------
Co-authored-by: Ferenc Hammerl <[email protected]>
* Set runner environment in runner context and env
Extract runner_environment from the global context and expose in the
`github.runner` context and env as `RUNNER_ENVIRONMENT`.
Signed-off-by: Philip Harrison <[email protected]>
* encoding.
---------
Signed-off-by: Philip Harrison <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* [1742] Ensure multiple composite annoations are correctly written.
This implementation uses a collector pattern to allow embedded ExecutionContexts to stash Issue objects for later processing by a non-embedded ancestor ExecutionContext.
Also:
- Provide explicit constructor implementations for ExecutionContext
- Leverage explicit constructors to solidify immutability of several ExecutionContext class members.
- Fixed erroneous call to ExecutionContext.Complete in CompositeActionHandler.cs
- Use a consistent timestamp for FinishTime in ExecutionContext::Complete
* Ensure collected issues are processed only by a non-embedded ExecutionContext.
This was already implicit. Now, just making it explicit.
* Provide a clear mechanism that allows callers to opt-in/opt-out of ExecutionContext::AddIssue's logging behavior.
* Addressed deserialization inconsistencies in TimelineRecord.cs
* Added TimelineRecord unit tests.
* Refined unit tests related to TimelineRecord::Variables case-insensitivity
* Add a unit test that verifies ExecutionContextLogOptions::LogMessageOverride has the desired effect.
* Responded to PR feedback.
* Don't allow embedded ExecutionContexts to add Issues to a TimelineRecord
* Only upload to Results with new job message type
* No need to have separate websocketFeedServer
* Linting fix
* Update src/Runner.Common/JobServerQueue.cs
Co-authored-by: Tingluo Huang <[email protected]>
* add connection timeout
* Consolidate initializing webclient to result client
* Retry websocket delivery for console logs
* Linter fix
* Do not give up for good, reconnect again in 10 minutes
* Has to reset delivered
* Only first time retry 3 times to connect to websocket
---------
Co-authored-by: Tingluo Huang <[email protected]>
* Rename AcquireJobRequest::StreamID to AcquireJobRequest::JobMessageID to match corresponding server-side update.
* appeased the linter
* Added unit tests to prove AcquireJobRequest serialization/deserialization is as-expected.
* Distinguish unit test variations.
* Incorporated PR Feedback.
* which handles broken symlink & unit test added (#2150)
* Update src/Runner.Sdk/Util/WhichUtil.cs
Co-authored-by: Ferenc Hammerl <[email protected]>
* fix pr comments
* trace log added, in case we find a symlink that is broken
* add check in case the target of the symlink is a relative path; added test that check symlink that targets a full path; added test that check symlink that targets a relative path;
* fix tests
* fix tests for linux
---------
Co-authored-by: Eli Entelis <[email protected]>
Co-authored-by: Eli Entelis <[email protected]>
Co-authored-by: Ferenc Hammerl <[email protected]>
* Also send Steps update to Results service
* Refactor to separate results server from current job server
* If hit any error while uploading to Results, skip Results upload
* Add proxy authentication and buffer request for WinHttpHandler
* Remove unnecessary null guard
* Also send Results telemetry when step update fails
* IResultsServer is not disposable
* Guard against NullReference while creating HostContext
* Throw on IOUtil loading object if content is empty or object is null
* removed unused dependency
* Changed exceptions to ArgumentNullException and ArgumentException
* fixed my mistake on LoadObject
* fixed tests
* trigger workflows
* trigger workflows
* encode files using utf8
* terminate the runner on deprication message
* added TaskAgentVersion exception to catch deprication
* AccessDenied exception with inner exception InvalidTaskAgent
* Access denied exception in program and in message listener
* Fixed copy
* remove trace message from message listener
* Bypass top level domain even if no_proxy specified it with leading '.'
E.g. no_proxy='.github.com' will now bypass github.com.
* Bypass proxy on all hosts when no_proxy is * (wildcard)
* Undo '.' stripping
* Simplify unit tests
* Respect wildcard even if it's one of many no_proxy items
* Runner service exit after consecutive re-try exits
* Rename failure counts and include reset count on runner listening for jobs
* Changed from graceful shutdown to stopping=true
* Revert "Revert "Uploading step logs to Results as well (#2422)" (#2437)"
This reverts commit 8c096baf49.
* Properly guard the upload to results feature
* Delete skipped file if deletesource is true
* Update lint.yml
Don't ignore the formatting errors
* Add formatting made by @cory-miller
* Use dotnet format
* Format only modified files
* Add instruction to contribute.md
* Use git status instead of git diff
* Rename queue to results queue
* Add results contracts
* Add Results logging handling
* Adding calls to create and finalize append blob
* Modifications for azurite upload
* Only call upload complete on final section and remove size
* Make method specific to step log so we can support job log later
* Change contract for results
* Add totalline count to the result log upload file
* Actually pass lineCount to Results Service
* Fix typos
* Code cleanup
* Fixing typos
* Apply suggestions from code review
Co-authored-by: Konrad Pabjan <[email protected]>
---------
Co-authored-by: Brittany Ellich <[email protected]>
Co-authored-by: Konrad Pabjan <[email protected]>
* Sanitize OS Desc for UserAgents
* Only drop brackets if needed, refactoring
* Add missing ')'
* Readd missing brackets around '(header)'
* Add comments
* Use bracket solution from SDK
* Rename tests
* send more stuff as part of job completed
* feedback
* set only once
* feedback
* feedback
* fix test
* feedback
* nit: spacing
* nit: line
Co-authored-by: Tingluo Huang <[email protected]>
---------
Co-authored-by: Tingluo Huang <[email protected]>
* Defer evaluation of a step's DisplayName until its condition is evaluated.
* Formalize TryUpdateDisplayName and EvaluateDisplayName as members of interface `IStep` (#2374)
* start calling run service for job completion
* cleanup
* nit: lines
Co-authored-by: Tingluo Huang <[email protected]>
* clean up
* give sanity back to thboop
Co-authored-by: Thomas Boop <[email protected]>
* add clean up back
* clean up
* clean up more
* oops
* copied from existing, but :thumb:
Co-authored-by: Thomas Boop <[email protected]>
---------
Co-authored-by: Tingluo Huang <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
* Remove unnecessary timelineId and timelineRecordId and use Guid stepId
* Log upload error to kusto
* Remove try-catch
* Using a well known telmetry record to avoid replacing issues
* fix Guid format
* Use results service for uploading step summaries
* Use results summary over generic results naming convention
* Apply suggestions from code review
Co-authored-by: Tingluo Huang <[email protected]>
* Addressing feedback
* Fix merge issue
* Remove empty line
* Update Results json objects to use snake case
* Adding the reference
Co-authored-by: Yang Cao <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* Added env variable to control wether the terminal is silent
* Log to stdout if PrintLogToStdout is enabled
* Extracted console logging to stdouttracelistener
* Remove useless usings
* Rewrite TraceListener as superclass
* Only print to stdout if env is set
* Add comment for Console.Out
* Format Listener
* Revert var name in terminal
* Check env in hostcontext instead of Tracing constructor
* Remove superclass & dupe logging code
* Log hostType
* Readonly '_' prefix 'hostType'
* Fix test
* Revert Terminal change
Co-authored-by: Ferenc Hammerl <[email protected]>
* Workflow
* add back github-token to id:image
* added handle to image name
* removed core require
* multi line added :latest
* added label
* added repository_owner in IMAGE_NAME
* with the release
* release
* markdown label description
* Remove markdown desciprion
* Remove double quotes in labels
* Reverted back releaseVersion
* run.sh installs SIGINT and SIGTERM traps to gracefully stop runner
* replaced trap to force wait for child processes and to send kill to run-helper instead of Runner.Listener
* run with signal handling if RUNNER_MANUALLY_TRAP_SIG is set
* Update src/Misc/layoutroot/run.sh
Co-authored-by: Tingluo Huang <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* Allow container.image to be null/empty
* Remove unnecessary workflow type
* Update src/Sdk/DTPipelines/Pipelines/ObjectTemplating/PipelineTemplateConverter.cs
Co-authored-by: Ferenc Hammerl <[email protected]>
* Rename ConvertToJobContainer to ConvertToContainer
It converts serviceContainers too
* Remove unused parameter
AllowExpressions was always false
* Allow service in yaml with empty string for value
* Don't attempt to start services without an image
* Improve error messages
* Revert "Remove unused parameter"
This reverts commit f4ab2d50d0.
* Revert "Rename ConvertToJobContainer to ConvertToContainer"
This reverts commit ffc62f8343.
* Fix revert merge conflict
* Removed info logs
* Added info log for services without container images
Co-authored-by: moonblade <[email protected]>
Co-authored-by: Ferenc Hammerl <[email protected]>
Co-authored-by: Ava Stancu <[email protected]>
For runners that are already configured but need to add systemd services after the fact, a new command option is added to generate the file only. Once generated, ./svc.sh install and other commands will be available.
This also adds support for falling back to the tenant name in the Actions URL in cases when the GitHub URL is not provided, such as for our hosted larger runners.
* Avastancu/joannaakl/service container error log (#2110)
* adding support for a service container docker logs
* Adding Unit test to ContainerOperationProvider
* Adding another test to ContainerOperationProvider
* placed the docker logs output in dedicated ##group section
* Removed the exception thrown if the service container was not healthy
* Removed duplicated logging to the executionContext
* Updated the container logs sub-section message
* Print service containers only if they were healthy
Unhealthy service logs are printed in ContainerHealthCheckLogs called prior to this step.
* Removed recently added method to inspect docker logs
The method was doing the same thing as the existing DockerLogs method.
* Added execution context error
This will make a failed health check more visible in the UI without disrupting the execution of the program.
* Removing the section 'Waiting for all services to be ready'
Since nested subsections are not being displayed properly and we already need one subsection per service error.
* Update src/Runner.Worker/Container/DockerCommandManager.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/TestHostContext.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Change the logic for printing Service Containers logs
Service container logs will be printed in the 'Start containers' section only if there is an error.
Healthy services will have their logs printed in the 'Stop Containers' section.
* Removed unused import
* Added back section group.
* Moved service containers error logs to separate group sections
* Removed the test testing the old logic flow.
* Remove unnecessary 'IsAnyUnhealthy' flag
* Remove printHello() function
* Add newline to TestHostContext
* Remove unnecessary field 'UnhealthyContainers'
* Rename boolean flag indicating service container failure
* Refactor healthcheck logic to separate method to enable unit testing.
* Remove the default value for bool variable
* Update src/Runner.Worker/ContainerOperationProvider.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Runner.Worker/ContainerOperationProvider.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Rename Healthcheck back to ContainerHealthcheck
* Make test sequential
* Unextract the container error logs method
* remove test asserting thrown exception
* Add configure await
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Add back test asserting exception
* Check service exit code if there is no healtcheck configured
* Remove unnecessary healthcheck for healthy service container
* Revert "Check service exit code if there is no healtcheck configured"
This reverts commit fec24e8341.
Co-authored-by: Ava S <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* Do not fail service containers without the healthcheck
Co-authored-by: JoannaaKL <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* adding support for a service container docker logs
* Adding Unit test to ContainerOperationProvider
* Adding another test to ContainerOperationProvider
* placed the docker logs output in dedicated ##group section
* Removed the exception thrown if the service container was not healthy
* Removed duplicated logging to the executionContext
* Updated the container logs sub-section message
* Print service containers only if they were healthy
Unhealthy service logs are printed in ContainerHealthCheckLogs called prior to this step.
* Removed recently added method to inspect docker logs
The method was doing the same thing as the existing DockerLogs method.
* Added execution context error
This will make a failed health check more visible in the UI without disrupting the execution of the program.
* Removing the section 'Waiting for all services to be ready'
Since nested subsections are not being displayed properly and we already need one subsection per service error.
* Update src/Runner.Worker/Container/DockerCommandManager.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/TestHostContext.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Change the logic for printing Service Containers logs
Service container logs will be printed in the 'Start containers' section only if there is an error.
Healthy services will have their logs printed in the 'Stop Containers' section.
* Removed unused import
* Added back section group.
* Moved service containers error logs to separate group sections
* Removed the test testing the old logic flow.
* Remove unnecessary 'IsAnyUnhealthy' flag
* Remove printHello() function
* Add newline to TestHostContext
* Remove unnecessary field 'UnhealthyContainers'
* Rename boolean flag indicating service container failure
* Refactor healthcheck logic to separate method to enable unit testing.
* Remove the default value for bool variable
* Update src/Runner.Worker/ContainerOperationProvider.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Runner.Worker/ContainerOperationProvider.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Rename Healthcheck back to ContainerHealthcheck
* Make test sequential
* Unextract the container error logs method
* remove test asserting thrown exception
* Add configure await
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Test/L0/Worker/ContainerOperationProviderL0.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Add back test asserting exception
* Check service exit code if there is no healtcheck configured
* Remove unnecessary healthcheck for healthy service container
* Revert "Check service exit code if there is no healtcheck configured"
This reverts commit fec24e8341.
Co-authored-by: Ava S <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
* Adding a new vars context for non-secret variables
* Fix test case
* Trigger checks
* Remove variables from env context and environment varibale
* remove extra references
* Add prefix handling to configuration variables
* Fix test cases
* Consume variables using vars in context data
* removed action_yaml changes
* docker: escape key-value pair as -e KEY and VALUE being environment var
* removed code duplication, removed unused method and test
* add release notes
Co-authored-by: Nikola Jokic <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
Co-authored-by: Nikola Jokic <[email protected]>
* update.sh new logic with creating a file and waiting for it for 30 sec
* pr refactor
* removing file when it's detected bu run-helper
* Update src/Misc/layoutroot/run-helper.sh.template
Co-authored-by: Ferenc Hammerl <[email protected]>
* clean up update.finished in Runner.Listener right before it starts a process _update.sh
* pr fix IOUtil.DeleteFile
* self update widnows version - cmd
* same logic for returnCode 4 - ephemeral
* Init var so cmd doesn't run into syntax errors
* Use constants, setting a var messed up ERRORLEVEL
* Use var for updatefile
Co-authored-by: stefanruvceski <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
Co-authored-by: Ferenc Hammerl <[email protected]>
* Created env var for forcing node12 actions to run on node16
* get value of hostContext environment variable
* changing location of forced node version check
* small code refactoring
Co-authored-by: Ferenc Hammerl <[email protected]>
* more of small code refactoring
Co-authored-by: Ferenc Hammerl <[email protected]>
* refactoring of conditions for getting internal node version
* changing expected value for node version env var
* Adding empty line between two methods
* Created method GetNodeVersion
* GetNodeVersion from function to inline call and PR fixes
Co-authored-by: Ferenc Hammerl <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
* Jsut simple solution without additional funcs
* Delete old comment
* resolve
* Refactor retry function, make it more common
* Make retry function generic, get rid of extra params
* delete extra using
* Add cancellation token and limit of attempts
* Add some additional logging
* Rework condition
* replace to do..while
* return `while (true)` to simplify code structure
* Add other cancelling token, add TODO comment
* Added ability to run Dockerfile.SUFFIX ContainerAction
* Extracted IsDockerFile method
* reformatted, moved from index to Last()
* extracted IsDockerfile to DockerUtil with L0
* added check for IsDockerfile to account for docker://
* updated test to clearly show path/dockerfile:tag
* fail if Data.Image is not Dockerfile or docker://[image]
* Setup noops for JobPrepare and JobCleanup hooks
* Add container jobstarted and jobcomplete hooks
* Run 'index.js' instead of specific command hooks
* Call jobprepare with command arg
* Use right command name (hardcoded)
Co-authored-by: Nikola Jokic <[email protected]>
* Invoke hooks with arguments
* Add PrepareJob hook to work with jobcontainers
Co-authored-by: Nikola Jokic <[email protected]>
* Rename methods
* Use new hookcontainer to run prep and clean hooks
* Get path from ENV
* Use enums
* Use IOUtils.cs
* Move container files to folder
* Move namespaces
* Store "state" between hooks
* Remove stdin stream in containerstephosts
* Update Constants.cs
* Throw if stdin fails
* Cleanup obvious nullrefs and unused vars
* Cleanup containerhook directory
* Call step exec hook
* Fix windows build
* Remove hook from hookContainer
* Rename file
* More renamings
* Add TODOs
* Fix env name
* Fix missing imports
* Fix imports
* Run script step on jobcontainer
* Enable feature if env is set
* Update ContainerHookManager.cs
* Update ContainerHookManager.cs
* Hooks allowed to work even when context isn't returned
* Custom hooks enabled flag and additional null checks
* New line at the end of the FeatureFlagManager.cs
* Code refactoring
* Supported just in time container building or pulling
* Try mock-build for osx
* Build all platforms
* Run mock on self-hosted
* Remove GITHUB prefix
* Use ContainerHooksPath instead of CustomHooksPath
* Null checks simplified
* Code refactoring
* Changing condition for image builing/pulling
* Code refactoring
* TODO comment removed
Co-authored-by: Ferenc Hammerl <[email protected]>
* Call container step if FF is on
* Rename run script function
* Use JToken instead of dynamic
* Add TODO
* Small refactoring + renames + TODOs
* Throw on DetermineNodeRuntimeVersion
* Fix formatting
* Add run-container-step
* Supported nodeJS in Alpine containers
* Renamed Alpine to IsAlpine in HookResponse
* Method for checking platform for alpine container
* Added container hooks feature flag check
* Update IsHookFeatureEnabled with new params
* Rename featureflag method
* Finish rename
* Set collection null values to empty arrays when JSON serialising them
* Disable FF until we merge
* Update src/Runner.Worker/Container/ContainerHooks/HookContainer.cs
* Fix method name
* Change hookargs to superclass from interface
* Using only Path.Combine in GenerateResponsePath
* fix merge error
* EntryPointArgs changed to list of args instead of one args string
* Changed List to IEnumerable for EntryPointArgs and MountVolumes
* Get ContainerRuntimePath for JobContainers from hooks
* Read ContainerEnv from response file
* Port mappings saved after creating services
* Support case when responseFile doesn't exist
* Check if response file exists
* Logging in ExecuteHookScript
* Save hook state after all 4 hooks
* Code refactoring
* Remove TODO
Co-authored-by: Ferenc Hammerl <[email protected]>
* Remove second TODO
Co-authored-by: Ferenc Hammerl <[email protected]>
* Removing container env changes
* Removing containerEnv and dockerManager
* Delete mock-build.yml
* Update IOUtil.cs
* Add comment about containerhooks
* Fix merge mistake
* Remove solved todo
* Determine which shell to use for hooks scenario
* Overload for method ExecuteHookScript with prependPath as arg
* Adding HostContext to the GetDefaultShellForScript call
* prependPath as a mandatory parameter
* Improve logging for hooks
* Small changes in logging
* Allow null for ContainerEntryPointArgs
* Changed log messages
* Skip setting EntryPoint and EntryPointArgs if hooks are enabled
* Throw if IsAlpine is null in PrepareJob
* Code refactoring - added GetAndValidateResponse method
* Code refactoring
* Changes in exception message
* Only save hookState if returned
* Use FF from server
* Empty line
* Code refactoring
Co-authored-by: Ferenc Hammerl <[email protected]>
* Send null instead of string empty
* Remove TODO
* Code refactoring and some small changes
* Allow Globals to be null to pass L0
* Fix setup in StepHostL0
* Throw exception earlier if response file doesn't exist and prepare_job hook is running
* Refactoring GetResponse method
* Changing exception message if response file is not found
Co-authored-by: Ferenc Hammerl <[email protected]>
* Chaning exception message if isAlpine is null for prepare_job hook
* Rename hook folder
* Fail if compatible hookfile not found
* Use .Value instead of casting bool? to bool
* Format spacing
* Formatting
* User user and system mvs
* Use variables instead of entire context in featuremanager
* Update stepTelemetry if step uses containerhooks
* Restore import
* Remove unneccessary field from HookContainer
* Refactor response context and portmappings
* Force allow hooks if FF is on
* Code refactoring
* Revert deleting usings
* Better hookContainer defaults and use correct portmapping list
* Make GetDefaultShellForScript a HostContext extension method
* Generic hookresponse
* Code refactoring, unnecessary properties removed - HookContainer moved to the HookInput.cs
* Remove empty line
* Code refactoring and better exception handling
* code refactor, removing unnecessary props
* Move hookstate to global ContainerHookState
* Trace exception before we throw it for not losing information
* Fix for null ref exception in GetResponse
* Adding additional check for null response in prepareJob hook
* Refactoring GetResponse with additional check
* Update error messages
* Ports in ResponseContainer changed from IList to IDictionary
* Fix port format
* Include dockerfile
* Send null Registry obj if there's nothing in it
* Minor formatting
* Check if hookIndexPath exists relocated to the ContainerHookManager
* Code refactoring - ValidateHookExecutable added to the ContainerHookManager
* check if ContainerHooksPath when AllowRunnerContainerHooks is on
* Submit JSON telemetry instead of boolean
* Prefix step hooks with "run"
* Rename FeatureManager
* Fix flipped condition
* Unify js shell path getter with ps1 and sh getter
* Validate on run, not on instantiation of manager
* Cleanup ExecuteAsync methods
* Handle exception in executeHookScript
* Better exception types
* Remove comment
* Simplify boolean
* Allow jobs without jobContainer to run
* Use JObject instead of JToken
* Use correct Response type
* Format class to move cleanupJobAsync to the end of public methods
* Rename HookIndexPath to HookScriptPath
* Refactor methods into expression bodies
* Fix args class hierarchy
* Fix argument order
* Formatting
* Fix nullref and don't swallow stacktrace
* Whilelist HookArgs
* Use FF in FeatureManager
* Update src/Runner.Worker/ContainerOperationProvider.cs
Co-authored-by: Tingluo Huang <[email protected]>
* Update src/Runner.Worker/ActionRunner.cs
Co-authored-by: Thomas Boop <[email protected]>
* Update src/Runner.Worker/ActionRunner.cs
Co-authored-by: Thomas Boop <[email protected]>
* Only mount well known dirs to job containers
* Get trace from hostcontext
* Use hook execution for setting telemetry
Co-authored-by: Nikola Jokic <[email protected]>
Co-authored-by: Nikola Jokic <[email protected]>
Co-authored-by: Nikola Jokic <[email protected]>
Co-authored-by: Stefan Ruvceski <[email protected]>
Co-authored-by: ruvceskistefan <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
Co-authored-by: stefanruvceski <[email protected]>
Co-authored-by: Tingluo Huang <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
* Allow admins to fail jobs without container
* Make method static
* Update src/Runner.Common/Constants.cs
Co-authored-by: Thomas Boop <[email protected]>
* Update src/Runner.Worker/JobExtension.cs
Co-authored-by: Thomas Boop <[email protected]>
* Update src/Runner.Worker/JobExtension.cs
Co-authored-by: Thomas Boop <[email protected]>
* Rename env
* Add test for throwing when no container but required
* Update src/Runner.Worker/JobExtension.cs
* Update src/Test/L0/Worker/JobExtensionL0.cs
* Update src/Runner.Common/Constants.cs
Co-authored-by: Thomas Boop <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
It generated invalid arguments for `Process()` when the `bash` command itself was an argument as well, for example:
```
_proc.StartInfo.FileName = "/usr/bin/docker";
_proc.StartInfo.Arguments = "exec -i --workdir /__w/container-hook-e2e/container-hook-e2e 47105c66144d8809d9fa2bce9a58ea0564cd14def0ae7952cd6231fba3576db1 sh -e '/__w/_temp/fd086560-cb92-4f3b-a99c-35a6b7b1bbdb.sh'";
```
* Job hook provider now sets shell name for script handler
* fixed script handler and job hook provider to work with the name without fail
* returned used import by osx
* fixed order of imports
* added quotes around resolved script path allowing space in script path
* added quotes around bash and sh _defaultArguments
* Changed double quotes to single quotes in sh -e
Co-authored-by: Ferenc Hammerl <[email protected]>
* Changed double quotes to single quotes in bash
Co-authored-by: Ferenc Hammerl <[email protected]>
Co-authored-by: Ferenc Hammerl <[email protected]>
* Added ability to run Dockerfile.SUFFIX ContainerAction
* Extracted IsDockerFile method
* reformatted, moved from index to Last()
* extracted IsDockerfile to DockerUtil with L0
* added check for IsDockerfile to account for docker://
* updated test to clearly show path/dockerfile:tag
* fail if Data.Image is not Dockerfile or docker://[image]
* Added warning in case of invalid combination of command and flags and/or arguments
* Deleting unnecessary comments
* Added separate list for generic options
* Added PAT to the valid remove options
* Added command name to the error message
* composite action github.action_path set based on the StepHost
* in progress on updating github context for input template
* Fixed updating the context data for evaluation
* refactored logic so it is a little cleaner
* removed resolving the action_path in CompositeActionHandler
* removed added DeepClone
* added feature flag and modified the dict in place
* refactored step host to change context data. Added L0
* repaired spaces
* moved logic from step host to execution context, added recursive translation
* removed empty lines
* moved to extension methods
* Update src/Test/L0/Worker/StepHostL0.cs
Co-authored-by: Ferenc Hammerl <[email protected]>
* Added continue on error to composite action
* changed from boolean-strategy-context -> boolean-steps-context for action_yaml
* refactored composite handler to always set outcome
* retrigger checks
* fixed typo in ??= operator
* boolean-steps-context accepts the same context as string-steps-context
* setting the outcome only on continue-on-error
* moved continue on error logic to the execution context
* Added L0 table tests for continue-on-error ExecutionContext
* Added missing mocks on StepsRunnerL0 for this update
* removed empty line and added one line separating the call
* Removed empty line
Co-authored-by: Ferenc Hammerl <[email protected]>
Co-authored-by: Ferenc Hammerl <[email protected]>
* Fixing null reference exception when configuring runner with invalid repo URL or token
* Throw exception instead of ConvertFromJson
* Storing the response code
* Add annotations if node 12 action is found
* Better placeholder
* Only warn if FF is on
* Move annotation logic
* Pass in the LTS Url
* Raise annotation right before executing the action
* Match server side FF name
* Change name back to features
* Better warning text
* Update src/Runner.Common/Constants.cs
Co-authored-by: Thomas Boop <[email protected]>
Co-authored-by: Thomas Boop <[email protected]>
* feed via websocket
* feed via websocket
* feedback
* ensure right schema is used
* fix resiliency
* some fixes
* fix sending message
* chunk data
* let's abort, which will also dispose
* close gracefully
* use safe_sleep executable in bash scripts
* new line at the end of safe_sleep bash script
* Replacing relative paths with absolute paths and changing location of safe_sleep
Co-authored-by: Ferenc Hammerl <[email protected]>
* Set GH actions Node version to 16 if FF is on unless user opted out
* Add L0s (WIP)
* Wrap tests into theory
* Only check for node12 actions
* Refactor node version picking
* Lets allow up to 150 characters on linux/mac, just to avoid some issues with runner naming
* Add 4 randomized digits on mac/linux
* fix pragma issue
* fix test
* Address pr feedback
* reduce complexity
* lets make it cleaner!
* fix test
* fix logic
* Issue 1596: runner throws nullref exception when writting env var
* Adding tests for missing new line after EOF marker
* Changing newline to new line
* Issue 1662: Adding retry policy for methods GetTenantCredential and GetJITRunnerTokenAsync
* Adding HttpClient creation to the retry
* Random backoff time
* Use 16 to run RunnerService.js
* Execute hashfiles using node16
* Run downloadCert.js using node16
* Run makeWebRequest.js using node16
* Run macos-run-invoker.js using node16
* Run hashFiles.js using node16
* Update tests to use node16
* Update documentation to recommend node16
* Duplicate macos service js fix for 16
* Add PR link
* Revert ADR node change
* Merge node12/16 retainment IFs
* Try both node12 and node16
* Close if
* Revert "Update tests to use node16"
This reverts commit bbca7b9f1c.
* Fix condition
* Unfurl if condition
* Allow user to force a node version
* Format update template
* Comment env var
* Rename vars
* Fix naming
* Fix rename
* Set node ver override if job message has it
* Format executionContext
* Can only receive 'forceNode12' or nothing from FF
No specific node version from server
Co-authored-by: Ferenc Hammerl <[email protected]>
* Solving issue with debugging on macOS M1
* Fixing problem with debugging on macOS M1
* Adding targetArchitecture in launch.json configs
* Code refactor
* Kill node process to recover handle
So we can print to the console in Runner.Listener once again
* Revert testing changes
Co-authored-by: Ferenc Hammerl <[email protected]>
* First prototype of step summary environment variable
* Fix file contention issue
* Try to simplify cleaning up file references
* use step id as md file name, queue file attachment
* separate logic into attachment summary func
* Fix indentation
* Add (experimental) feature flag support
* reorganize summary upload determination logic
* file i/o exception handling + pr feedback
* Revert changes for now to reintroduce them later
* Add skeleton SetStepSummaryCommand
* Update step summary feature flag name
* Port ShouldUploadAttachment from previous iteration
* Port QueueStepSummaryUpload from previous iteration
* Improve exception handling when uploading attachment
* Add some minor logging improvements
* Refuse to upload files larger than 128k
* Implement secrets scrubbing
* Add TODO comment to remove debugging temp files
* Add first tests
* Add test for secret masking
* Add some naming/style fixes suggested in feedback
* inline check for feature flag
* Inline method for style consistency
* Make sure that scrubbed file doesn't exist before creating it
* Rename SetStepSummaryCommand to CreateStepSummaryCommand
* Fix error handling messages
* Fix file command name when registering extension
* Remove unnecessary file deletion
Co-authored-by: Rob Herley <[email protected]>
* Only execute post for actions that have one
* Working container runner update with run.sh
* Revert "Only execute post for actions that have one"
This reverts commit 9675941fdb.
* Relaunch the listener without quitting run.cmd
* Fix typo
* Extract most os run.sh logic so we can update it
* Add bash line endings
* Extract the logic from run.cmd
* Add EoF lines
* Add unexpected ERRORLEVEL messages to cmd
* Simplify contract between run and helper
* Remove unused exit
* WIP: run a copy of the helper so it's safe to update
* Throw NonRetryableException if not configured
* Log and format
* Fix typo
* Fix typo
* Use helper template system for bash as well
* Update run.sh
* Remove unnecessary comments
* Use ping instead of timeout
* Use localhost in ping-timeout (n times, w timeout)
Co-authored-by: Ferenc Hammerl <[email protected]>
* Revert "Make `run.sh|cmd` handle update without quitting so containers using them as entrypoints don't exit on update (#1494)"
d8251bf912
* update runnerversion as well
* Only execute post for actions that have one
* Working container runner update with run.sh
* Revert "Only execute post for actions that have one"
This reverts commit 9675941fdb.
* Relaunch the listener without quitting run.cmd
* Fix typo
* Extract most os run.sh logic so we can update it
* Add bash line endings
* Extract the logic from run.cmd
* Add EoF lines
* Add unexpected ERRORLEVEL messages to cmd
* Simplify contract between run and helper
* Remove unused exit
* WIP: run a copy of the helper so it's safe to update
* Throw NonRetryableException if not configured
* Log and format
* Fix typo
* Fix typo
* Use helper template system for bash as well
* Update run.sh
* Remove unnecessary comments
* Use ping instead of timeout
* Use localhost in ping-timeout (n times, w timeout)
Co-authored-by: Ferenc Hammerl <[email protected]>
Many people open bug reports or feature requests in the `actions/runner`
repository that are more generally about GitHub Actions. Often changes
in GitHub Actions are cross-cutting across multiple teams or feature
areas, so it's best if we direct people to the more general areas
(Actions Community Support or GitHub Feedback) so that we can get the
most eyes on the problem and give the quickest response.
* conditional support for composite actions
* Fix Conditional function evaluation
* Push launch.json temporarily
* Revert "Push launch.json temporarily"
* rename context
* Cleanup comments
* fix success/failure functions to run based on pre/main steps
* idea of step_status
* change to use steps context, WIP
* add inputs to possible if condition expressions
* use action_status
* pr cleanup
* Added right stages
* Test on stage in conditional functions
* Fix naming and formatting
* Fix tests
* Add success and failure L0s
* Remove comment
* Remove whitespace
* Undo formatting
* Add L0 for step-if parsing
* Add ADR
Co-authored-by: Thomas Boop <[email protected]>
* Always register post steps for local actions
* Register post steps along with their conditions
* remove debug code
Co-authored-by: Ferenc Hammerl <[email protected]>
Without creating a session, the service is not able to access the keychains for the user specified under `UserName`. This causes any workflow that deals with code signing to fail as the only keychain loaded with be the system one. This should fix#350
* Handle macOS upgrade smoothly
* cleanup
* misc cleanup
* final updates
* Update src/Misc/layoutbin/update.sh.template
Co-authored-by: Patrick Ellis <[email protected]>
* Update src/Misc/layoutbin/update.sh.template
Co-authored-by: Patrick Ellis <[email protected]>
* Upload telemetry and default to old method as needed
* minor fix
* add one more bit of logging
* some more telemetry
* quote variables to handle spaces
* tiny fix for ubuntu
* remove version and move telemetry to diag
* use full path
Co-authored-by: Patrick Ellis <[email protected]>
* Prevent stopTokens that are workflow commands
Co-authored-by: Thomas Boop <[email protected]>
* Check context for env var too
* Accept true, 1 and $true instead of just "true"
* Setup ExpressionValues in tests
* Update src/Runner.Common/Constants.cs
Co-authored-by: Thomas Boop <[email protected]>
* Separate success and fail tests for invalid token
* Fix envcontext for tests
Co-authored-by: Thomas Boop <[email protected]>
* Ignore agentpool when unconfiguring the runner
Runner names and IDs are unique within a ServiceHost
They don't need to be included when unconfiguring the runner.
* Use -1 instead of 0 to highlight how it is ignored
* Use overloads and 0 instead of -1
Using 0 seems to be the convention
* Fix typo calling the wrong method
* Add generateIdTokenUrl as an env var
* Add generateIdTokenUrl to env vars
* Add basic telemetry class and submit it on jobcompleted
* Use constructor overload
* Rename telemetry to jobTelemetry
* Rename telemetry file
* Make JobTelemetryType a string
* Collect telemetry
* Remove debugger
* Update src/Runner.Worker/ActionCommandManager.cs
Co-authored-by: Thomas Boop <[email protected]>
* Use same JobTelemetry for all contexts
* Mask telemetry data
* Mask in JobRunner instead
* Empty line
* Change method signature
Returning with a List suggests we clone it and that the
original doesn't change..
* Update launch.json
Co-authored-by: Thomas Boop <[email protected]>
* Add generateIdTokenUrl as an env var
* Add generateIdTokenUrl to env vars
* Add name and runner_group to context
* No longer add runner-group
* Update runner name if needed
* Get interface instead of concrete class
* Check for nulls on ReservedAgent
* Avoid loading setting file unnecesseraly
* Only check agentName once
* Use Trace.Error when can't update settings
* Better equals and exception handling
* Update JobDispatcher.cs
* Add tests and null check
* Avoid white, it is unreadable on light themes (#1295)
* remove ', ConsoleColor.White' from banner
* remove ', ConsoleColor.White' from prompt
* cleanups
* Update error to say 'uninstall' not 'unconfigure'
* Say uninstall service in *nix config error msgs
Co-authored-by: Ferenc Hammerl <[email protected]>
* Trim pwsh special chars when masking secrets
* Add pwsh valueEncoder
* Explain regex
* Update ValueEncoders.cs
* Add tests for pwsh color codes in secrets
* Formatting
* Group tests into theories
* Split secret on PS chars and mask for them
* Clean up comments
* Remove unused unittest
* Rename escape methods
* Use flags in svc creation script
* Refactor regex and add comments
* Fix indentation and typo in user matching
* Consistency use flags in automation scripts
* Update documentation to reflect new usage
* Make example more readable
* Remove test echos from script
* Remove test echo
* Format scripts and remove test script
* Remove tar
* Use getopts and single letter flags
* Update docs to show flag usage
* Update usage of create svc
* Revert svc to not use flags
* Revert delete script
* Update docs
* Readd deleted comments
* Add Notice Command
* Add Feature Flag For Enhanced Annotations
* Fix tests
* Add validation for columns and lines
* Fix order to match service
* Remove console.write
* Make Validation Better
* Cleanup
* Handle empty/whitespace strings
* Add more validation for line/column ranges
* Make Validation Debug, Not Throw
* Change casing to 🐫 from 🐍
* Give notice a well known tag
* Cleanup
* Sanitize invalid commands rather than fail
* fix some typos in docs of JobDispatcher
* one more typo
* Update src/Runner.Listener/JobDispatcher.cs
Co-authored-by: Ferenc Hammerl <[email protected]>
* Stop ignoring .vscode (launch scripts)
* Check in launch scripts for config and run
This can cause an issue with existing launch configuration on the machines of contributors.
* Improve error msg when runner is not configured
* Unignore .vscode/launch and tasks only
* Remove stopAtEntry and add eof newline
* Remove Runner.Listener from error message
* Rename tasks and run configs
* Ignore BuildConstants.cs
* Use better error msg
* Explain development steps in depth
* Add launch config to directly debug worker
* Update docs with VS Code tips
* Remove auto-generated comments
* Fix link to quickstart in vscode.md
* Remove ':' from link to quickstart
* Revert "Ignore BuildConstants.cs"
This reverts commit 0f13922a87.
* Replace `.sh` with `.(sh/cmd)` in docs
* Wait for debugger to attach in Worker
Only if GITHUB_ACTIONS_RUNNER_ATTACH_DEBUGGER is set
* Only wait if env variable parses to boolean 'true'
* Add 30s timeout to the wait
* Clean up leftover line
* Decrease wait to 20s
* Use ConvertToBoolean isntead of TryParse
From February 2021, in order to provide feedback on pull requests, Code Scanning workflows must be configured with both `push` and `pull_request` triggers. This is because Code Scanning compares the results from a pull request against the results for the base branch to tell you only what has changed between the two.
Early in the beta period we supported displaying results on pull requests for workflows with only `push` triggers, but have discontinued support as this proved to be less robust.
See https://docs.github.com/en/free-pro-team@latest/github/finding-security-vulnerabilities-and-errors-in-your-code/configuring-code-scanning#scanning-pull-requests for more information on how best to configure your Code Scanning workflows.
During job run we may fail to resolve actions download info, and this
stack is fully controlled by GitHub actions so it should be counted as
infrastructure failure instead of user failure.
* Use FIPS compliant crypto when required
* Comment cleanup
* Store OAuth signing scheme in credentialData instead of runner setting
Add encryption scheme for job message encyption key to session
Further cleanup of unused crypto code
* Update windows rsa key manager to use crossplat dotnet RSA api
* Undo unneeded ConfigurationManager change
* add `workflow_dispatch`
* Add an environment variable to indicate which repository the currently running Action came from.
* Expose the Action ref as well.
* Move setting `github.action_repository` and `github.action_ref` to `ActionRunner.cs`.
* Don't set `action_repository` and `action_ref` for local Actions.
Co-authored-by: Tingluo Huang <[email protected]>
remove 3 "redundant" text and put one text for DRY.
and developers always forget `sudo` and annoying `Need to run with sudo privilege` message.
so, add first.
* Change `ping .. > nul` to `sleep`
The filename `nul` is a Windows-ism that causes the update script to
create such a file in the current working directory. The `ping`
utility is also an dependency not installed by
`installdependencies.sh`, so it seemed easier to change it to the
standard `sleep` command.
* Update dotnet-install script as requested by test
* Update dotnet-install.ps1
Co-authored-by: Tingluo Huang <[email protected]>
* Unify apt/apt-get logic
The previous logic was buggy in that it tried to use `apt` in the `apt-get` branch after deciding that `apt` was unavailable...
* Prefer apt-get over apt
apt does not have a stable cli and using it from scripts yields annoying messages
* Improve English for missing apt-get & apt case
* Fix apt-get/apt fallback behavior for $ patterns
If there's a `$` in the apt install pattern, it will not fail if it selects a thing and decides it isn't interested in installing it.
* Fix spelling of libssl
This change updates the .NET Core SDK used by the Actions Runner to
version 3.1.302 to address the issues that are caused by the following issue:
https://github.com/dotnet/runtime/issues/13475
See #574 for more information.
Fixes#574
* Log in with container credentials if given
* Stub in registry aware auth for later
* Fix hang if password is empty
* Remove default param to fix build
* PR Feedback. Add some tests and fix parse
* Add environment variable for GITHUB_ACTION_PATH
* ah
* Remove debugging messages
* Set github action path at step level instead of global scope to avoid necessary removal
* Remove set context for github action
* Set github action path before and after composite action
* Copy GitHub Context, use this copied context for each composit step, and then set the action_path for each one (to avoid stamping over parent pointer GitHubContext
* Explicitly define what is allowed for an action
* Add step-env
* Remove secrets + defaults
* new line
* Add safety check to prevent from checking defaults in ScriptHandler for composite action
* Revert "Add safety check to prevent from checking defaults in ScriptHandler for composite action"
This reverts commit aeae15de7b.
* Need to explictly use ActionStep type since we need the .Inputs attribute which is only found in the ActionStep not IStep
* Fix ActionManifestManager
* Remove todos
* Revert "Revert "Add safety check to prevent from checking defaults in ScriptHandler for composite action""
This reverts commit a22fcbc036.
* revert
* Remove needs in env
* Make shell required + add inputs
* Remove passing context to all composite steps attribuyte
* Remove redundant code (display name is already evaluated in ActionRunner beforehand for each step)
* remove
* Remove nesting information for composite steps.
* put messages in debug logs if composite. if not, put these messages as outputs
* Fix group issue
* Fix end group issue
* Exploring child Linked Cancellation Tokens
* Preliminary Timeout-minutes fix
* Final Solution for resolving cancellation token's timeout vs. cancellation
* Clean up + Fix error handling
* Use linked tokens instead
* Clean up
* one liner
* Remove JobExecutionContext => Replace with public Root accessor
* Move CreateLinkedTokenSource in the CreateCompositeStep Function
This PR changes GITHUB_ACTION to use the step ContextName, instead of refname. The behavior is behind a feature flag. Refname is an otherwise deprecated property.
Primary motivation: For composite actions, we need a distinct GITHUB_ACTION for each nested step. This PR adds code to generate a default context name for nested steps.
For nested steps, GITHUB_ACTION will be set to "{ScopeName}.{ContextName}" to ensure no collisions.
A corresponding change will be made on the server so context name is never empty. Generated context names will start with "__".
A follow-up PR is required to avoid tracking "step" context values (outputs/conclusion/result) for generated context names. Waiting on telemetry from the server to confirm it's safe to assume leading "__" is a generate context name.
* Add basic framework for baby steps runner
* Basic logic for adding steps / invoking composite action steps
* Composite Steps Runner MVP
* Fix null object reference error
* intialize composiute
* Comment out code that is handled by stepsrunner
* Add composite clean up step
* Remove previous 'workarounds' from StepsRunner. Clean Up PR
* Remove todo
* Remove todo
* Fix using unitialized object yikes
* Remove time delay
* Format handler
* Move output handler into action handler
* Add try to evaluate display name
* Remove while loop yikes
* Abstract away the windows encoding check during step running
* Github context set to {ScopeName}.{ContextName} or {ContextName} if ScopeName is null
* Remove setting result to sucess since result defaults to sucess
* Fix windows error
* Fix windows
* revert:
* Windows fix
* Fix Windows Error in Abstraction
* Remove Composite Steps Runner => consolidate into Composite Steps Runner
* Remove unn. attribute in ExecutionContext
* Change protection levels, plus change function name to more clear meaning
* Remove location param
* location pt.2 fix
* Remove outputs step
* Remove temp directory
* new line
* Add arguitl not null
* better comment
* Change encoding name
* Check count > 0 for composite steps, import System.Threading
* Change function header encodingutil
* Add TODO
* Add await
* Handle Failed Step
* Move over SetAllCompositeOutputs to the handler
* Remove timeout-minutes setting in steps-level
* Use only ExecutionContext
* Move using to the top
* Remove redundant check
* Change function name
* Remove testing code
* Consolidate error code
* Consolidate code
* Change HandleOutput => ProcessCompositeActionOutputs
* Remove set the timeout comment
* Add Cancelling functionality + Remove unn. parameter
* Composite Action Run Steps
* Env Flow => Able to get env variables and overwrite current env variables => but it doesn't 'stick'
* clean up
* Clean up trace messages + add Trace debug in ActionManager
* Add debugging message
* Optimize runtime of code
* Change String to string
* Add comma to Composite
* Change JobSteps to a List, Change Register Step function name
* Add TODO, remove unn. content
* Remove unnecessary code
* Fix unit tests
* Fix env format
* Remove comment
* Remove TODO message for context
* Add verbose trace logs which are only viewable by devs
* Initial Start for FileTable stuff
* Progress towards passing FileTable or FileID or FileName
* Sort usings in Composite Action Handler
* Change 0 to location
* Update context variables in composite action yaml
* Add helpful error message for null steps
* Pass fileID to all children token of root action token
* Change confusing term context => templateContext, Eliminate _fileTable and only use ExecutionContext.FileTable + update this table when need be
* Remove unnessary FileID attribute from CompositeActionExecutionData
* Clean up file path for error message
* Remove todo
* Initial start/framework for output handling
* Outline different class vs Handler approach
* Remove InitializeScope
* Remove InitializeScope
* Fix Workflow Step Env overiding Parent Env
* First Approach for Attaching ID + Group ID to each Composite Action Step
* Add GroupID to the ActionDefinitionData
* starting foundation for handling clean up outputs step
* Pass outputs data to each composite action step to enable set-output functionality
* Create ScopeName for whole composite action.
This will enable us to add to the StepsContext[ScopeName] for the composite action which will allow us to use all these outputs in the cleanup step
* Hook up composite output step to handler => tmmrw implement composite output handler
* Add post composite action step to cleanup outputs => triggers composite output cleanup handler
* Fix Outputs Token handling start. Add individual step scope names.
* Set up Scope Name and Context Name naming system{
* Figured out how to pass Parent Execution Context to clean up step
* Figured out how to pass Parent Execution Context and scope names to
clean up step
* Add GetOutput function for StepsContext
* Generate child scope name correctly if parent scope name is null
* Simplify InitializeScope()
* Outputs are set correctly and able to get all final outputs in handler
* Parse through Action Outputs
* Fix null ScopeName + ContextName in CompositeOutputHandler
* Shift over handling of Action Outputs to output handler
* First attempt to fix null retrievals for output variables
* Basic Support for Outputs Done.
* Clean up pt.1
* Refactor outputs to avoid using Action Reference
* Clean up code
* Clean up part 2
* Add clarifying comments for the output handler
* Remove TODO
* Remove env in composite action scope
* Clean up
* Revert back
* revert back
* add back envToken
* Remove unnecessary code
* Add file length check
* Clean up
* Fix logging issue
* Figure out how to handle set-env edge cases
* formatting
* fix unit tests
* Fix windows unit test syntax error
* Fix period
* Sanity check for fileTable add + remove unn. code
* revert back
* Add back line break
* Fix null errors
* Address situation if FileTable is null + add sanity check for adding file to fileTable
* add line
* Revert
* Fix unit tests to instantiate a FileTable
* Fix logic for trimming manifestfile path
* Add null check
* Revert
* Revert
* revert
* spacing
* Add filetable to testing file, remove ? since we know filetable should never be non null
* Fix Throw logic
* Clarify template outputs token
* Add another type support for outputs to avoid container unit tests errors
* Add mapping for parity
* Build support for new outputs format
* Build support for new outputs format
* Refactor to avoid duplication of action yaml for workflow yaml
* revert
* revert
* revert
* spacing
* Composite Action Run Steps
* Env Flow => Able to get env variables and overwrite current env variables => but it doesn't 'stick'
* clean up
* Clean up trace messages + add Trace debug in ActionManager
* Add debugging message
* Optimize runtime of code
* Change String to string
* Add comma to Composite
* Change JobSteps to a List, Change Register Step function name
* Add TODO, remove unn. content
* Remove unnecessary code
* Fix unit tests
* Fix env format
* Remove comment
* Remove TODO message for context
* Add verbose trace logs which are only viewable by devs
* Initial Start for FileTable stuff
* Progress towards passing FileTable or FileID or FileName
* Sort usings in Composite Action Handler
* Change 0 to location
* Update context variables in composite action yaml
* Add helpful error message for null steps
* Pass fileID to all children token of root action token
* Change confusing term context => templateContext, Eliminate _fileTable and only use ExecutionContext.FileTable + update this table when need be
* Remove unnessary FileID attribute from CompositeActionExecutionData
* Clean up file path for error message
* Remove todo
* Initial start/framework for output handling
* Outline different class vs Handler approach
* Remove InitializeScope
* Remove InitializeScope
* Fix Workflow Step Env overiding Parent Env
* First Approach for Attaching ID + Group ID to each Composite Action Step
* Add GroupID to the ActionDefinitionData
* starting foundation for handling clean up outputs step
* Pass outputs data to each composite action step to enable set-output functionality
* Create ScopeName for whole composite action.
This will enable us to add to the StepsContext[ScopeName] for the composite action which will allow us to use all these outputs in the cleanup step
* Hook up composite output step to handler => tmmrw implement composite output handler
* Add post composite action step to cleanup outputs => triggers composite output cleanup handler
* Fix Outputs Token handling start. Add individual step scope names.
* Set up Scope Name and Context Name naming system{
* Figured out how to pass Parent Execution Context to clean up step
* Figured out how to pass Parent Execution Context and scope names to
clean up step
* Add GetOutput function for StepsContext
* Generate child scope name correctly if parent scope name is null
* Simplify InitializeScope()
* Outputs are set correctly and able to get all final outputs in handler
* Parse through Action Outputs
* Fix null ScopeName + ContextName in CompositeOutputHandler
* Shift over handling of Action Outputs to output handler
* First attempt to fix null retrievals for output variables
* Basic Support for Outputs Done.
* Clean up pt.1
* Refactor outputs to avoid using Action Reference
* Clean up code
* Clean up part 2
* Add clarifying comments for the output handler
* Remove TODO
* Remove env in composite action scope
* Clean up
* Revert back
* revert back
* add back envToken
* Remove unnecessary code
* Add file length check
* Clean up
* Figure out how to handle set-env edge cases
* formatting
* fix unit tests
* Fix windows unit test syntax error
* Fix period
* Sanity check for fileTable add + remove unn. code
* revert back
* Add back line break
* Fix null errors
* Address situation if FileTable is null + add sanity check for adding file to fileTable
* add line
* Revert
* Fix unit tests to instantiate a FileTable
* Fix logic for trimming manifestfile path
* Add null check
* Revert
* Revert
* revert
* spacing
* Add filetable to testing file, remove ? since we know filetable should never be non null
* Fix Throw logic
* Clarify template outputs token
* Add another type support for outputs to avoid container unit tests errors
* Add mapping for parity
* Build support for new outputs format
* Refactor to avoid duplication of action yaml for workflow yaml
* Move SDK work in ActionManifestManager, Condense Code
* Defer runs evaluation till after for loop to ensure order doesn't matter
* Fix logic error in setting scope and context names
* Add Regex + Add Child Context name null resolution
* move private function to bottom of class
* Composite Action Run Steps
* Env Flow => Able to get env variables and overwrite current env variables => but it doesn't 'stick'
* clean up
* Clean up trace messages + add Trace debug in ActionManager
* Add debugging message
* Optimize runtime of code
* Change String to string
* Add comma to Composite
* Change JobSteps to a List, Change Register Step function name
* Add TODO, remove unn. content
* Remove unnecessary code
* Fix unit tests
* Fix env format
* Remove comment
* Remove TODO message for context
* Add verbose trace logs which are only viewable by devs
* Initial Start for FileTable stuff
* Progress towards passing FileTable or FileID or FileName
* Sort usings in Composite Action Handler
* Change 0 to location
* Update context variables in composite action yaml
* Add helpful error message for null steps
* Pass fileID to all children token of root action token
* Change confusing term context => templateContext, Eliminate _fileTable and only use ExecutionContext.FileTable + update this table when need be
* Remove unnessary FileID attribute from CompositeActionExecutionData
* Clean up file path for error message
* Remove todo
* Fix Workflow Step Env overiding Parent Env
* Remove env in composite action scope
* Clean up
* Revert back
* revert back
* add back envToken
* Remove unnecessary code
* Add file length check
* Clean up
* Figure out how to handle set-env edge cases
* formatting
* fix unit tests
* Fix windows unit test syntax error
* Fix period
* Sanity check for fileTable add + remove unn. code
* revert back
* Add back line break
* Fix null errors
* Address situation if FileTable is null + add sanity check for adding file to fileTable
* add line
* Revert
* Fix unit tests to instantiate a FileTable
* Fix logic for trimming manifestfile path
* Add null check
* Add filetable to testing file, remove ? since we know filetable should never be non null
* Composite Action Run Steps
* Env Flow => Able to get env variables and overwrite current env variables => but it doesn't 'stick'
* clean up
* Clean up trace messages + add Trace debug in ActionManager
* Add debugging message
* Optimize runtime of code
* Change String to string
* Add comma to Composite
* Change JobSteps to a List, Change Register Step function name
* Add TODO, remove unn. content
* Remove unnecessary code
* Fix unit tests
* Fix env format
* Remove comment
* Remove TODO message for context
* Add verbose trace logs which are only viewable by devs
* Sort usings in Composite Action Handler
* Change 0 to location
* Update context variables in composite action yaml
* Add helpful error message for null steps
* Fix Workflow Step Env overiding Parent Env
* Remove env in composite action scope
* Clean up
* Revert back
* revert back
* add back envToken
* Remove unnecessary code
* Figure out how to handle set-env edge cases
* formatting
* fix unit tests
* Fix windows unit test syntax error
* Composite Action Run Steps
* Clean up trace messages + add Trace debug in ActionManager
* Change String to string
* Add comma to Composite
* Change JobSteps to a List, Change Register Step function name
* Add TODO, remove unn. content
* Remove unnecessary code
* Fix unit tests
* Add verbose trace logs which are only viewable by devs
* Sort usings in Composite Action Handler
* Change 0 to location
* Update context variables in composite action yaml
* Add helpful error message for null steps
about:If you have issues with GitHub Actions, please follow the "support for GitHub Actions" link, below.
title:''
labels:bug
assignees:''
---
<!--
👋 You're opening a bug report against the GitHub Actions **runner application**.
🛑 Please stop if you're not certain that the bug you're seeing is in the runner application - if you have general problems with actions, workflows, or runners, please see the [GitHub Community Support Forum](https://github.community/c/code-to-cloud/52) which is actively monitored. Using the forum ensures that we route your problem to the correct team. 😃
-->
**Describe the bug**
A clear and concise description of what the bug is.
about:If you have feedback or suggestions about GitHub Actions, please open a discussion (or add to an existing one) in the GitHub Actions Feedback. GitHub Actions Product Managers and Engineers monitor the feedback forum.
Thank you 🙇♀ for wanting to create a feature in this repository. Before you do, please ensure you are filing the issue in the right place. Issues should only be opened on if the issue **relates to code in this repository**.
* If you have found a security issue [please submit it here](https://hackerone.com/github)
* If you have questions or issues with the service, writing workflows or actions, then please [visit the GitHub Community Forum's Actions Board](https://github.community/t5/GitHub-Actions/bd-p/actions)
* If you are having an issue or question about GitHub Actions then please [contact customer support](https://help.github.com/en/actions/automating-your-workflow-with-github-actions/about-github-actions#contacting-support)
If you have a feature request that is relevant to this repository, the runner, then please include the information below:
**Describe the enhancement**
A clear and concise description of what the features or enhancement you need.
**Code Snippet**
If applicable, add a code snippet.
**Additional information**
Add any other context about the feature here.
NOTE: if the feature request has been agreed upon then the assignee will create an ADR. See docs/adrs/README.md
Whenever possible, changes should be accompanied by non-trivial tests that meaningfully exercise the core functionality of the new code being introduced.
All tests are in the `Test/` directory at the repo root. Fast unit tests are in the `Test/L0` directory and by convention have the suffix `L0.cs`. For example: unit tests for a hypothetical `src/Runner.Worker/Foo.cs` would go in `src/Test/L0/Worker/FooL0.cs`.
Run tests using this command:
```sh
cd src && ./dev.sh test
```
### Formatting
After editing .cs files, always format the code using this command:
```sh
cd src && ./dev.sh format
```
### Feature Flags
Wherever possible, all changes should be safeguarded by a feature flag; `Features` are declared in [Constants.cs](src/Runner.Common/Constants.cs).