Phase 2: Implement Execution Environment Abstraction (v0.3.0)

This commit implements Phase 2 of the CHORUS Task Execution Engine development plan,
providing a comprehensive execution environment abstraction layer with Docker
container sandboxing support.

## New Features

### Core Sandbox Interface
- Comprehensive ExecutionSandbox interface with isolated task execution
- Support for command execution, file I/O, environment management
- Resource usage monitoring and sandbox lifecycle management
- Standardized error handling with SandboxError types and categories

### Docker Container Sandbox Implementation
- Full Docker API integration with secure container creation
- Transparent repository mounting with configurable read/write access
- Advanced security policies with capability dropping and privilege controls
- Comprehensive resource limits (CPU, memory, disk, processes, file handles)
- Support for tmpfs mounts, masked paths, and read-only bind mounts
- Container lifecycle management with proper cleanup and health monitoring

### Security & Resource Management
- Configurable security policies with SELinux, AppArmor, and Seccomp support
- Fine-grained capability management with secure defaults
- Network isolation options with configurable DNS and proxy settings
- Resource monitoring with real-time CPU, memory, and network usage tracking
- Comprehensive ulimits configuration for process and file handle limits

### Repository Integration
- Seamless repository mounting from local paths to container workspaces
- Git configuration support with user credentials and global settings
- File inclusion/exclusion patterns for selective repository access
- Configurable permissions and ownership for mounted repositories

### Testing Infrastructure
- Comprehensive test suite with 60+ test cases covering all functionality
- Docker integration tests with Alpine Linux containers (skipped in short mode)
- Mock sandbox implementation for unit testing without Docker dependencies
- Security policy validation tests with read-only filesystem enforcement
- Resource usage monitoring and cleanup verification tests

## Technical Details

### Dependencies Added
- github.com/docker/docker v28.4.0+incompatible - Docker API client
- github.com/docker/go-connections v0.6.0 - Docker connection utilities
- github.com/docker/go-units v0.5.0 - Docker units and formatting
- Associated Docker API dependencies for complete container management

### Architecture
- Interface-driven design enabling multiple sandbox implementations
- Comprehensive configuration structures for all sandbox aspects
- Resource usage tracking with detailed metrics collection
- Error handling with retryable error classification
- Proper cleanup and resource management throughout sandbox lifecycle

### Compatibility
- Maintains backward compatibility with existing CHORUS architecture
- Designed for future integration with Phase 3 Core Task Execution Engine
- Extensible design supporting additional sandbox implementations (VM, process)

This Phase 2 implementation provides the foundation for secure, isolated task
execution that will be integrated with the AI model providers from Phase 1
in the upcoming Phase 3 development.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
anthonyrawlins
2025-09-25 14:28:08 +10:00
parent d1252ade69
commit 8d9b62daf3
653 changed files with 88039 additions and 3766 deletions

View File

@@ -90,6 +90,10 @@ git push <YOUR_FORK> <YOUR_BRANCH_NAME>
Open a pull request against the main `opentelemetry-go` repo. Be sure to add the pull
request ID to the entry you added to `CHANGELOG.md`.
Avoid rebasing and force-pushing to your branch to facilitate reviewing the pull request.
Rewriting Git history makes it difficult to keep track of iterations during code review.
All pull requests are squashed to a single commit upon merge to `main`.
### How to Receive Comments
* If the PR is not ready for review, please put `[WIP]` in the title,
@@ -105,10 +109,9 @@ A PR is considered **ready to merge** when:
This is not enforced through automation, but needs to be validated by the
maintainer merging.
* The qualified approvals need to be from [Approver]s/[Maintainer]s
affiliated with different companies. Two qualified approvals from
[Approver]s or [Maintainer]s affiliated with the same company counts as a
single qualified approval.
* At least one of the qualified approvals need to be from an
[Approver]/[Maintainer] affiliated with a different company than the author
of the PR.
* PRs introducing changes that have already been discussed and consensus
reached only need one qualified approval. The discussion and resolution
needs to be linked to the PR.
@@ -177,25 +180,76 @@ patterns in the spec.
For a deeper discussion, see
[this](https://github.com/open-telemetry/opentelemetry-specification/issues/165).
## Tests
Each functionality should be covered by tests.
Performance-critical functionality should also be covered by benchmarks.
- Pull requests adding a performance-critical functionality
should have `go test -bench` output in their description.
- Pull requests changing a performance-critical functionality
should have [`benchstat`](https://pkg.go.dev/golang.org/x/perf/cmd/benchstat)
output in their description.
## Dependencies
This project uses [Go Modules] for dependency management. All modules will use
`go.mod` to explicitly list all direct and indirect dependencies, ensuring a
clear dependency graph. The `go.sum` file for each module will be committed to
the repository and used to verify the integrity of downloaded modules,
preventing malicious tampering.
This project uses automated dependency update tools (i.e. dependabot,
renovatebot) to manage updates to dependencies. This ensures that dependencies
are kept up-to-date with the latest security patches and features and are
reviewed before being merged. If you would like to propose a change to a
dependency it should be done through a pull request that updates the `go.mod`
file and includes a description of the change.
See the [versioning and compatibility](./VERSIONING.md) policy for more details
about dependency compatibility.
[Go Modules]: https://pkg.go.dev/cmd/go#hdr-Modules__module_versions__and_more
### Environment Dependencies
This project does not partition dependencies based on the environment (i.e.
`development`, `staging`, `production`).
Only the dependencies explicitly included in the released modules have be
tested and verified to work with the released code. No other guarantee is made
about the compatibility of other dependencies.
## Documentation
Each non-example Go Module should have its own `README.md` containing:
Each (non-internal, non-test) package must be documented using
[Go Doc Comments](https://go.dev/doc/comment),
preferably in a `doc.go` file.
- A pkg.go.dev badge which can be generated [here](https://pkg.go.dev/badge/).
- Brief description.
- Installation instructions (and requirements if applicable).
- Hyperlink to an example. Depending on the component the example can be:
- An `example_test.go` like [here](exporters/stdout/stdouttrace/example_test.go).
- A sample Go application with its own `README.md`, like [here](example/zipkin).
- Additional documentation sections such us:
- Configuration,
- Contributing,
- References.
Prefer using [Examples](https://pkg.go.dev/testing#hdr-Examples)
instead of putting code snippets in Go doc comments.
In some cases, you can even create [Testable Examples](https://go.dev/blog/examples).
[Here](exporters/jaeger/README.md) is an example of a concise `README.md`.
You can install and run a "local Go Doc site" in the following way:
Moreover, it should be possible to navigate to any `README.md` from the
root `README.md`.
```sh
go install golang.org/x/pkgsite/cmd/pkgsite@latest
pkgsite
```
[`go.opentelemetry.io/otel/metric`](https://pkg.go.dev/go.opentelemetry.io/otel/metric)
is an example of a very well-documented package.
### README files
Each (non-internal, non-test, non-documentation) package must contain a
`README.md` file containing at least a title, and a `pkg.go.dev` badge.
The README should not be a repetition of Go doc comments.
You can verify the presence of all README files with the `make verify-readmes`
command.
## Style Guide
@@ -208,6 +262,10 @@ For a non-comprehensive but foundational overview of these best practices
the [Effective Go](https://golang.org/doc/effective_go.html) documentation
is an excellent starting place.
We also recommend following the
[Go Code Review Comments](https://go.dev/wiki/CodeReviewComments)
that collects common comments made during reviews of Go code.
As a convenience for developers building this project the `make precommit`
will format, lint, validate, and in some cases fix the changes you plan to
submit. This check will need to pass for your changes to be able to be
@@ -475,8 +533,33 @@ documentation are allowed to be extended with additional methods.
> Warning: methods may be added to this interface in minor releases.
These interfaces are defined by the OpenTelemetry specification and will be
updated as the specification evolves.
Otherwise, stable interfaces MUST NOT be modified.
#### How to Change Specification Interfaces
When an API change must be made, we will update the SDK with the new method one
release before the API change. This will allow the SDK one version before the
API change to work seamlessly with the new API.
If an incompatible version of the SDK is used with the new API the application
will fail to compile.
#### How Not to Change Specification Interfaces
We have explored using a v2 of the API to change interfaces and found that there
was no way to introduce a v2 and have it work seamlessly with the v1 of the API.
Problems happened with libraries that upgraded to v2 when an application did not,
and would not produce any telemetry.
More detail of the approaches considered and their limitations can be found in
the [Use a V2 API to evolve interfaces](https://github.com/open-telemetry/opentelemetry-go/issues/3920)
issue.
#### How to Change Other Interfaces
If new functionality is needed for an interface that cannot be changed it MUST
be added by including an additional interface. That added interface can be a
simple interface for the specific functionality that you want to add or it can
@@ -531,32 +614,110 @@ functionality should be added, each one will need their own super-set
interfaces and will duplicate the pattern. For this reason, the simple targeted
interface that defines the specific functionality should be preferred.
See also:
[Keeping Your Modules Compatible: Working with interfaces](https://go.dev/blog/module-compatibility#working-with-interfaces).
### Testing
We allow using [`testify`](https://github.com/stretchr/testify) even though
it is seen as non-idiomatic according to
the [Go Test Comments](https://go.dev/wiki/TestComments#assert-libraries) page.
The tests should never leak goroutines.
Use the term `ConcurrentSafe` in the test name when it aims to verify the
absence of race conditions. The top-level tests with this term will be run
many times in the `test-concurrent-safe` CI job to increase the chance of
catching concurrency issues. This does not apply to subtests when this term
is not in their root name.
### Internal packages
The use of internal packages should be scoped to a single module. A sub-module
should never import from a parent internal package. This creates a coupling
between the two modules where a user can upgrade the parent without the child
and if the internal package API has changed it will fail to upgrade[^3].
There are two known exceptions to this rule:
- `go.opentelemetry.io/otel/internal/global`
- This package manages global state for all of opentelemetry-go. It needs to
be a single package in order to ensure the uniqueness of the global state.
- `go.opentelemetry.io/otel/internal/baggage`
- This package provides values in a `context.Context` that need to be
recognized by `go.opentelemetry.io/otel/baggage` and
`go.opentelemetry.io/otel/bridge/opentracing` but remain private.
If you have duplicate code in multiple modules, make that code into a Go
template stored in `go.opentelemetry.io/otel/internal/shared` and use [gotmpl]
to render the templates in the desired locations. See [#4404] for an example of
this.
[^3]: https://github.com/open-telemetry/opentelemetry-go/issues/3548
### Ignoring context cancellation
OpenTelemetry API implementations need to ignore the cancellation of the context that are
passed when recording a value (e.g. starting a span, recording a measurement, emitting a log).
Recording methods should not return an error describing the cancellation state of the context
when they complete, nor should they abort any work.
This rule may not apply if the OpenTelemetry specification defines a timeout mechanism for
the method. In that case the context cancellation can be used for the timeout with the
restriction that this behavior is documented for the method. Otherwise, timeouts
are expected to be handled by the user calling the API, not the implementation.
Stoppage of the telemetry pipeline is handled by calling the appropriate `Shutdown` method
of a provider. It is assumed the context passed from a user is not used for this purpose.
Outside of the direct recording of telemetry from the API (e.g. exporting telemetry,
force flushing telemetry, shutting down a signal provider) the context cancellation
should be honored. This means all work done on behalf of the user provided context
should be canceled.
## Approvers and Maintainers
### Approvers
- [Evan Torrie](https://github.com/evantorrie), Verizon Media
- [Sam Xie](https://github.com/XSAM), Cisco/AppDynamics
- [David Ashpole](https://github.com/dashpole), Google
- [Robert Pająk](https://github.com/pellared), Splunk
- [Chester Cheung](https://github.com/hanyuancheung), Tencent
- [Damien Mathieu](https://github.com/dmathieu), Elastic
### Maintainers
- [Aaron Clawson](https://github.com/MadVikingGod), LightStep
- [Anthony Mirabella](https://github.com/Aneurysm9), AWS
- [Tyler Yahn](https://github.com/MrAlias), Splunk
- [Damien Mathieu](https://github.com/dmathieu), Elastic ([GPG](https://keys.openpgp.org/search?q=5A126B972A81A6CE443E5E1B408B8E44F0873832))
- [David Ashpole](https://github.com/dashpole), Google ([GPG](https://keys.openpgp.org/search?q=C0D1BDDCAAEAE573673085F176327DA4D864DC70))
- [Robert Pająk](https://github.com/pellared), Splunk ([GPG](https://keys.openpgp.org/search?q=CDAD3A60476A3DE599AA5092E5F7C35A4DBE90C2))
- [Sam Xie](https://github.com/XSAM), Splunk ([GPG](https://keys.openpgp.org/search?q=AEA033782371ABB18EE39188B8044925D6FEEBEA))
- [Tyler Yahn](https://github.com/MrAlias), Splunk ([GPG](https://keys.openpgp.org/search?q=0x46B0F3E1A8B1BA5A))
For more information about the maintainer role, see the [community repository](https://github.com/open-telemetry/community/blob/main/guides/contributor/membership.md#maintainer).
### Approvers
- [Flc](https://github.com/flc1125), Independent
For more information about the approver role, see the [community repository](https://github.com/open-telemetry/community/blob/main/guides/contributor/membership.md#approver).
### Triagers
- [Alex Kats](https://github.com/akats7), Capital One
- [Cheng-Zhen Yang](https://github.com/scorpionknifes), Independent
For more information about the triager role, see the [community repository](https://github.com/open-telemetry/community/blob/main/guides/contributor/membership.md#triager).
### Emeritus
- [Gustavo Silva Paiva](https://github.com/paivagustavo), LightStep
- [Josh MacDonald](https://github.com/jmacd), LightStep
- [Aaron Clawson](https://github.com/MadVikingGod)
- [Anthony Mirabella](https://github.com/Aneurysm9)
- [Chester Cheung](https://github.com/hanyuancheung)
- [Evan Torrie](https://github.com/evantorrie)
- [Gustavo Silva Paiva](https://github.com/paivagustavo)
- [Josh MacDonald](https://github.com/jmacd)
- [Liz Fong-Jones](https://github.com/lizthegrey)
For more information about the emeritus role, see the [community repository](https://github.com/open-telemetry/community/blob/main/guides/contributor/membership.md#emeritus-maintainerapprovertriager).
### Become an Approver or a Maintainer
See the [community membership document in OpenTelemetry community
repo](https://github.com/open-telemetry/community/blob/main/community-membership.md).
repo](https://github.com/open-telemetry/community/blob/main/guides/contributor/membership.md).
[Approver]: #approvers
[Maintainer]: #maintainers
[gotmpl]: https://pkg.go.dev/go.opentelemetry.io/build-tools/gotmpl
[#4404]: https://github.com/open-telemetry/opentelemetry-go/pull/4404