gh-actions-ps-profiling

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

gh-actions-ps-profiling

★ 15 Open GitHub ↗

GitHub Action for Continuous Profiling which you can run to profile your CI/CD. It uses parca and Polar Signals cloud.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# GitHub Action for Continuous Profiling

GitHub Action for Continuous Profiling which you can run to profile your workloads in CI/CD. It uses [Polar Signals Cloud](https://www.polarsignals.com/) (or any other [Parca](https://www.parca.dev/) compatible API).

## How to use

You will need access to Polar Signals Cloud or another way to send Parca profiles to remote store.

This project is also a demo of how to use this action, view the [.github/workflows](.github/workflows) directory to view the example usage.

If you are using Polar Signals Cloud, the only thing required to configure is the `polarsignals_cloud_token` which is the API token for Polar Signals Cloud, where it sends the profiling data. You can find the docs on how to obtain a token [here](https://www.polarsignals.com/docs/generating-tokens).

## PR Comments

On pull request events, this action automatically creates or updates a single PR comment with a link to your profiling data. The comment shows the latest run prominently and maintains a history of previous runs in a collapsible section.

For this feature to work, the `project_uuid` (your Polar Signals Cloud project UUID) is required. This UUID is not sensitive and can be provided directly in the workflow file, not as a secret - it ends up in the URL either way.

If any of the required parameters are missing, the PR comment will be skipped without failing the workflow, and log messages will indicate the reason.

### Required Permissions

```yaml
permissions:
  pull-requests: write
  contents: read
```

### Example Configuration

```yaml
name: Profiling Workflow
on: [pull_request]

permissions:
  pull-requests: write
  contents: read

jobs:
  profile:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3

### Matrix Workflows

When using matrix strategies, multiple parallel jobs run for the same commit. Use the `job_name` input to distinguish between them in PR comments:

```yaml
jobs:
  profile:
    strategy:
      matrix:
        node: [18, 20]
        os: [ubuntu-latest, macos-latest]
    runs-on: ${{ matrix.os }}
    steps:
      - uses: actions/checkout@v3

- name: Run Continuous Profiling
        uses: polarsignals/gh-actions-ps-profiling@main
        with:
          polarsignals_cloud_token: ${{ secrets.POLARSIGNALS_CLOUD_TOKEN }}
          project_uuid: 'your-project-uuid-here'
          job_name: node-${{ matrix.node }}-${{ matrix.os }}
          labels: "ref_name=${{ github.ref_name }};workflow=${{ github.workflow }};gh_run_id=${{ github.run_id }};gh_job=${{ github.job }};gh_job_index=${{ strategy.job-index }};node=${{ matrix.node }};os=${{ matrix.os }}"
```

Each matrix job will have its own row in the PR comment, tracked by the combination of commit SHA and job name.

## Labels

By default, the following labels are attached to profiling data:

- `ref_name` - The branch or tag name
- `workflow` - The workflow name
- `gh_run_id` - The GitHub Actions run ID
- `gh_job` - The GitHub Actions job name
- `gh_job_index` - The job index in matrix workflows (0-based)

You can customize labels using the `labels` input with a semicolon-separated list of `key=value` pairs.

### Example Profiling Data

Profiling data from one CI run looks [like this](https://pprof.me/475d1cc/).

gh-actions-ps-profiling

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README