test(influxdb3): Improve InfluxDB 3 Core and Enterprise Docker services

[jstirnaman]

Improve Docker Compose services and initialization script for testing InfluxDB 3 Core and Enterprise endpoints.

✅ All commands from SKILL.md work correctly!

• ✅ Verified token format works with both Core and Enterprise
• ✅ Docker secrets correctly mounted at /run/secrets/
• ✅ Database operations work with tokens in both formats
• ✅ Documentation examples tested and validated

Core Workflow Testing

1. ✅ Initialize Core: ./test/scripts/init-influxdb3.sh core - Successfully started
2. ✅ Check Core status: Required authentication (as expected with --admin-token-file)
3. ✅ Get token: jq -r .token test/.influxdb3/core/.token - Returns token correctly
4. ✅ Authenticated ping: Works with Authorization: Bearer header
5. ✅ Create database: Successfully created test_db
6. ✅ List databases: Shows _internal, test_core_db, and test_db

Enterprise Workflow Testing

1. ✅ Initialize Enterprise: ./test/scripts/init-influxdb3.sh enterprise - Successfully started
2. ✅ Get admin token: Token extracted from JSON file correctly
3. ✅ Check Enterprise status: /ping returns version info with authentication
4. ✅ Create worktree database: Created docs_v2_pr6750_db (worktree name converted correctly)
5. ✅ List databases: Shows _internal, docs_v2_pr6750_db, and test_enterprise_db

compose.yaml changes:

• Add influxdb3-core service with per-worktree data isolation
• Add influxdb3-core-noauth service for endpoint testing without auth
• Add influxdb3-enterprise service with trial license configuration
• Fix env_file handling (use optional path, remove variable override)
• Add detailed usage comments for each service

New files:

• test/scripts/init-influxdb3.sh: Helper script to initialize services, create directories, and validate configuration

• Enterprise configuration uses shared data at ~/influxdata-docs/.influxdb3/ to persist license and data across worktrees.

• .claude/skills/influxdb3-test-setup/SKILL.md. It covers:

  1. Architecture overview - Shows the directory structure for shared vs per-worktree data
  2. Quick reference - Common commands table
  3. Setup workflows - Core only, Enterprise only, or both
  4. Worktree-specific databases - How to create isolated databases named after the worktree
  5. Test environment configuration - Setting up .env.test files for code block tests
  6. Troubleshooting - Common issues (license, auth, ports)
  7. Service comparison - Core vs Enterprise differences

  The skill references the init script but keeps database creation as a separate workflow step, allowing flexibility for different testing scenarios.

[jstirnaman]

@copilot Update .github/instructions or skills for Copilot to reference and use the new .claude/skills/influxdb3-test-setup skill.

[Copilot]

Pull request overview

This PR improves Docker Compose services for testing InfluxDB 3 Core and Enterprise by adding proper data isolation, authentication handling, and initialization tooling. It introduces per-worktree Core instances and shared Enterprise instances with automatic token generation.

Changes:

• Adds initialization script for automated service setup with token generation
• Refactors Docker Compose services to use file-based token authentication instead of secrets
• Adds comprehensive documentation for test setup workflows and troubleshooting

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 7 comments.

File	Description
test/scripts/init-influxdb3.sh	New initialization script for automating service setup, directory creation, and token generation
compose.yaml	Updates service configurations for influxdb3-core and influxdb3-enterprise with new authentication model, adds influxdb3-core-noauth service
.claude/skills/influxdb3-test-setup/SKILL.md	New comprehensive documentation covering architecture, workflows, and troubleshooting

Comments suppressed due to low confidence (2)

compose.yaml:1

• The setup instructions mention creating an admin token JSON file but don't specify the required format or content. Users won't know what to put in this file. Consider adding a comment with the expected format or referencing where to find it.

# This is a Docker Compose file for the InfluxData documentation site.

test/scripts/init-influxdb3.sh:1

• The comment references creating an admin-token.json file in setup step 2, but line 109 states 'Admin token is auto-generated on first run.' This is contradictory. Either the file must be pre-created or it's auto-generated - clarify which is correct.

#!/bin/bash

[Copilot]

@jstirnaman I've opened a new pull request, #6751, to work on those changes. Once the pull request is ready, I'll request review from you.

[github-actions]

PR Preview Action v1.4.8
🚀 Deployed preview to https://influxdata.github.io/docs-v2/pr-preview/pr-6750/
on branch gh-pages at 2026-01-20 22:51 UTC

[github-actions]

PR Preview

Status	Details
Preview	View preview
Pages	4 page(s) deployed
Build time	65s
Last updated	2026-01-20 22:52:05 UTC

Pages included in this preview

• /influxdb3/enterprise/admin/tokens/admin/preconfigured/
• /influxdb3/core/admin/tokens/admin/preconfigured/
• /influxdb3/enterprise/get-started/setup/
• /influxdb3/core/get-started/setup/

---

_{Preview auto-deploys on push. Will be cleaned up when PR closes.}