Skip to main content

Developer-focused logging library for Python with OpenSearch integration.

Project description

devlogs

devlogs logo

A developer-focused logging library for Python based on OpenSearch.

Step 1: Get OpenSearch running

If you don't have OpenSearch running and you want to stand one up:

  1. Install Docker (and Docker Compose) for your OS.
  2. Copy docker-compose.example.yaml to docker-compose.yaml.
  3. Change the OpenSearch password in the copied file.
  4. Start OpenSearch:
    docker compose up -d
    

Step 2: Copy/paste these instructions into your coding agent

Please do the following in this project:

  1. Install devlogs as a dev dependency (pip install devlogs or add it to requirements-dev.txt/pyproject.toml optional dependencies).
  2. Add DEVLOGS_OPENSEARCH_HOST, DEVLOGS_OPENSEARCH_PORT, DEVLOGS_OPENSEARCH_USER, DEVLOGS_OPENSEARCH_PASS, and DEVLOGS_INDEX to the project .env file (or equivalent config). Defaults are localhost, 9200, admin, and devlogs-<projectname>.
  3. Add devlogs hooks at the beginning of the application (main entrypoint/startup module), wrapped in an environment check so it only runs in development:
    import os
    import logging
    if os.getenv("ENVIRONMENT") != "production":
        from devlogs.handler import DevlogsHandler
        from devlogs.build_info import resolve_build_info
    
        # Resolve build info (reads .build.json or generates)
        build_info = resolve_build_info(write_if_missing=True)
    
        handler = DevlogsHandler(
            application="my-app",  # Required: your app name
            component="api",       # Required: component name
            level=logging.INFO,
            version=build_info.build_id,
        )
        logging.getLogger().addHandler(handler)
        logging.getLogger().setLevel(logging.INFO)
    
        # Log startup
        logging.info("App started")
    
  4. Run devlogs init (inside the virtualenv if one is set up) and verify the index is healthy.
  5. Ask the user if they want MCP set up; if yes, state which agent you are (copilot, claude, or codex) and run devlogs initmcp <agent>.

Step 3: Use devlogs

  1. Run devlogs initmcp <agent> to set up the MCP server.
  2. Then run devlogs tail to see the last logs, or devlogs tail -f to follow along
  3. Finally, ask your agent to query devlogs for errors. Watch it solve problems on its own!

If you want to install it by hand

  1. Install devlogs:

    pip install devlogs
    
  2. Start OpenSearch:

    docker-compose up -d opensearch
    

    Or point DEVLOGS_OPENSEARCH_* at an existing cluster.

  3. Initialize indices/templates:

    devlogs init
    
  4. Use in Python code (development only):

    import os
    import logging
    
    # Only enable devlogs in development
    if os.getenv("ENVIRONMENT") != "production":
        from devlogs.handler import DevlogsHandler
        from devlogs.build_info import resolve_build_info
    
        # Get build info (reads .build.json or generates)
        build_info = resolve_build_info(write_if_missing=True)
    
        handler = DevlogsHandler(
            application="my-app",
            component="default",
            level=logging.DEBUG,
            version=build_info.build_id,
        )
        logging.getLogger().addHandler(handler)
        logging.getLogger().setLevel(logging.DEBUG)
    
        logging.info("Hello from devlogs!")
    
  5. Tail logs from CLI:

    devlogs tail --area web --follow
    
  6. Search logs from CLI:

    devlogs search --q "error" --area web
    
  7. Run the web UI:

    uvicorn devlogs.web.server:app --port 8088
    # Then open http://localhost:8088/ui/
    

MCP Agent Setup

If you want MCP set up, identify your agent type and run the matching command from your project root:

devlogs initmcp copilot
devlogs initmcp claude
devlogs initmcp codex
devlogs initmcp all

This writes MCP config files in the standard locations:

  • Claude: .mcp.json
  • Copilot (VS Code): .vscode/mcp.json
  • Codex: ~/.codex/config.toml

Features

  • DevlogsHandler - Standard logging.Handler for OpenSearch with v2.0 schema
  • HTTP Collector Service for centralized log ingestion
  • Devlogs Record Format v2.0 - Standardized schema with application, component, top-level message/level/area
  • Context manager for operation_id/area
  • Structured custom fields on log entries (extra={"features": {...}} stored as fields)
  • CLI and Linux shell wrapper
  • Minimal embeddable web UI
  • Robust tests (pytest)

Note: Version 2.0.0 introduces breaking changes. See MIGRATION-V2.md for upgrade instructions.

HTTP Collector

The devlogs collector is a standalone HTTP service for centralized log ingestion. It supports two modes:

  • Forward mode: Proxy requests to an upstream collector
  • Ingest mode: Write directly to OpenSearch

Quick Start

# Start collector in ingest mode
DEVLOGS_OPENSEARCH_HOST=localhost DEVLOGS_INDEX=devlogs-myapp devlogs-collector serve

# Start collector in forward mode
DEVLOGS_FORWARD_URL=https://central-collector.example.com devlogs-collector serve

Using the Python Client

from devlogs.devlogs_client import create_client

client = create_client(
    collector_url="http://localhost:8080",
    application="my-app",
    component="api-server",
)

client.emit(
    message="Request processed",
    level="info",
    fields={"user_id": "123", "duration_ms": 45}
)

Docker

docker build -f Dockerfile.collector -t devlogs-collector .
docker run -p 8080:8080 -e DEVLOGS_OPENSEARCH_URL=https://admin:pass@opensearch:9200/devlogs devlogs-collector

See HOWTO-COLLECTOR.md for complete collector documentation.

Jenkins Integration

Option 1: Jenkins Plugin (Recommended)

Install the Devlogs Jenkins plugin for native integration:

pipeline {
    agent any
    stages {
        stage('Build') {
            steps {
                devlogs(url: credentials('devlogs-url')) {
                    sh 'make build'
                }
            }
        }
    }
}

See jenkins-plugin/README.md for installation and usage details.

Option 2: Standalone Binary

Stream Jenkins build logs to OpenSearch using a standalone binary:

pipeline {
    agent any
    environment {
        DEVLOGS_OPENSEARCH_URL = credentials('devlogs-opensearch-url')
        DEVLOGS_BINARY_URL = credentials('devlogs-binary-url')
    }
    stages {
        stage('Build') {
            steps {
                sh 'curl -sL $DEVLOGS_BINARY_URL -o /tmp/devlogs && chmod +x /tmp/devlogs'
                sh '/tmp/devlogs jenkins attach --background'
                sh 'make build'
            }
        }
    }
    post {
        always { sh '/tmp/devlogs jenkins stop || true' }
    }
}

Build the binary with ./build-standalone.sh and host it somewhere accessible. See HOWTO-JENKINS.md for setup details.

Configuration

Environment Variables

Collector Configuration:

  • DEVLOGS_URL - Collector base URL (where apps send logs)
  • DEVLOGS_FORWARD_URL - Forward mode: proxy to this upstream URL

OpenSearch Admin Connection:

  • DEVLOGS_OPENSEARCH_HOST, DEVLOGS_OPENSEARCH_PORT, DEVLOGS_OPENSEARCH_USER, DEVLOGS_OPENSEARCH_PASS
  • DEVLOGS_OPENSEARCH_URL - URL shortcut (e.g., https://user:pass@host:9200/index)
  • DEVLOGS_OPENSEARCH_VERIFY_CERTS, DEVLOGS_OPENSEARCH_CA_CERT - SSL/TLS settings

Index & Retention:

  • DEVLOGS_INDEX - Target index name
  • DEVLOGS_RETENTION_DEBUG, DEVLOGS_RETENTION_INFO, DEVLOGS_RETENTION_WARNING - Retention policy (e.g., 24h, 7d)

Collector Limits (Future Provisions):

  • DEVLOGS_COLLECTOR_RATE_LIMIT - Max requests/second (0 = unlimited)
  • DEVLOGS_COLLECTOR_MAX_PAYLOAD_SIZE - Max payload bytes (0 = unlimited)

See .env.example for a complete configuration template.

CLI Options

Use --url to specify connection details without a .env file:

devlogs --url 'https://admin:pass@host:9200/myindex' tail

Use --env to load from a specific .env file:

devlogs --env /path/to/.env diagnose

URL Builder

Use devlogs mkurl to interactively create a properly URL-encoded connection string:

devlogs mkurl
# Outputs the URL in three formats:
# 1. Bare URL (for --url flag)
# 2. Single DEVLOGS_OPENSEARCH_URL variable
# 3. Individual .env variables

This is especially useful for passwords containing special characters like !, @, #, which must be URL-encoded.

See HOWTO-CLI.md for complete CLI reference.

Production Deployment

Devlogs is a development tool. The examples above show how to conditionally enable it using an environment check. You can also make it an optional dependency:

# pyproject.toml
[project.optional-dependencies]
dev = ["devlogs>=2.0.0"]

Install with pip install ".[dev]" in development, pip install . in production.

Project Structure

  • src/devlogs/ - Python library, CLI, MCP server, and web UI
  • browser/ - Browser/npm package for frontend logging
  • jenkins-plugin/ - Native Jenkins plugin for log streaming
  • devlogs - Shell wrapper for local development
  • tests/ - Pytest-based tests
  • dist/ - Built packages and standalone binary

Publishing

# Release to all platforms (PyPI, npm, GitHub)
./publish/release.sh

# Bump version and release
./publish/release.sh --bump-patch

# Preview release
./publish/release.sh --dry-run

See publish/RELEASING.md for detailed publishing instructions.

Build Info Helper

Tag every log entry with a stable build identifier without requiring git at runtime:

from devlogs.build_info import resolve_build_info

bi = resolve_build_info(write_if_missing=True)
# bi.build_id = "main-20260124T153045Z"
# bi.branch = "main"
# bi.source = "file" | "env" | "generated"

# Use with DevlogsHandler
handler = DevlogsHandler(
    application="my-app",
    component="api",
    version=bi.build_id,  # Include build info in handler
)
logging.info("Started")

The build info is resolved from (in priority order):

  1. Environment variables (DEVLOGS_BUILD_ID, DEVLOGS_BRANCH)
  2. Build info file (.build.json)
  3. Generated values (branch-timestamp format)

See docs/build-info.md for CI integration examples and full API reference.

See Also

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

devlogs-2.2.5.tar.gz (121.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

devlogs-2.2.5-py3-none-any.whl (89.3 kB view details)

Uploaded Python 3

File details

Details for the file devlogs-2.2.5.tar.gz.

File metadata

  • Download URL: devlogs-2.2.5.tar.gz
  • Upload date:
  • Size: 121.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for devlogs-2.2.5.tar.gz
Algorithm Hash digest
SHA256 e5cd335e346e7a3690504d4de9d1abe11cf71933f822839baa839dfd0459f5d8
MD5 182d02690bb5d825a9dffb53b6508ddb
BLAKE2b-256 27e716b4f3de12d2f79ac6c6f6e0e15714d54bc442adf4a5811ef40b817f5122

See more details on using hashes here.

File details

Details for the file devlogs-2.2.5-py3-none-any.whl.

File metadata

  • Download URL: devlogs-2.2.5-py3-none-any.whl
  • Upload date:
  • Size: 89.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for devlogs-2.2.5-py3-none-any.whl
Algorithm Hash digest
SHA256 556d90163a21012c874fda3ba8b542e6b1c0f904d7b41d3afb4c14f38f05a15c
MD5 c9aaed290668f17eb6d9a274bc824ad4
BLAKE2b-256 767a76da2350ed2498daa44357b2713ee6c78603ef62109a6a7b0de44e6a4289

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page