docker-stacks

Automations with docker stacks

Scaffolding

This project uses several tools to ensure code quality and consistency:

• MegaLinter: Runs in GitHub Actions to perform comprehensive linting on every Pull Request.
• Pre-commit: Local git hooks to format and lint code before committing.
• Renovate: Automatically manages and updates dependencies.
• EditorConfig: Ensures consistent coding styles across different editors and IDEs.
• Yamlfmt: Specifically handles YAML file formatting.
• GitHub Actions: Workflows for CI, security scans (OSSF Scorecard, Semgrep, etc.), and automerge.

OCI Resource Manager — ARM Free Tier VPS

The oci-rm/ directory contains a modular Terraform stack for OCI Resource Manager that provisions an Ampere A1 (ARM) Always Free VPS, including VCN, subnet, internet gateway, security rules, and compute instance.

Always Free limits

Resource	Limit
Shape	VM.Standard.A1.Flex
OCPUs	4 total across all A1 instances
Memory	24 GB total across all A1 instances
Block storage	200 GB total

Defaults: 4 OCPU / 24 GB RAM / 200 GB boot volume (Max Always Free).

Features

• Modular Architecture: Clean separation of concerns with network and compute modules.
• Structured Cloud-init: Automatically updates packages and installs curl and git.
• Configurable Networking: Easily specify additional TCP/UDP ports for ingress.
• ORM Optimized: Enhanced schema.yaml with logical grouping and descriptive tooltips.

Variables

All variables have sensible defaults. Required inputs:

• Compartment — target OCI compartment
• SSH Public Key — key for instance access
• Cloud-init Script (optional) — paste scripts/init.sh to auto-install Dokploy. This will be wrapped in a structured cloud-config.

Deploy via OCI Console (recommended)

1. Zip the stack:

   cd oci-rm && zip -r ../oci-rm-stack.zip . && cd ..

2. Open Developer Services → Resource Manager → Stacks in the OCI Console.

3. Click Create Stack → Upload a .zip file → select oci-rm-stack.zip.

4. Fill in the form (variables pre-populated with defaults).

5. Click Apply.

Deploy via OCI CLI

Requires OCI CLI configured (oci setup config).

Update an existing stack's Terraform config:

STACK_ID=<stack-ocid> bash scripts/oci-rm-stack-update.sh

Apply via cron (out-of-capacity retry)

A1.Flex Free Tier capacity is limited and apply jobs may fail with 500-InternalError, Out of host capacity. Since networking resources are created first and are idempotent, re-applying the same stack retries only the instance.

scripts/oci-rm-stack-apply.sh is designed for cron:

• Skips if previous apply already succeeded (idempotent)
• Skips if a job is already in progress
• Prints logs from previous failed job before retrying
• Exits 0 on capacity errors (no cron failure spam)

# Run once
STACK_ID=<stack-ocid> bash scripts/oci-rm-stack-apply.sh

# Run at interval in terminal (every 10 minutes)
export STACK_ID=<stack-ocid>
watch -n 600 bash scripts/oci-rm-stack-apply.sh

# Cron example (every 10 minutes)
*/10 * * * * STACK_ID=<stack-ocid> /path/to/scripts/oci-rm-stack-apply.sh

Dokploy Setup on OCI

This repository includes a script to automate the setup of Dokploy on an Oracle Cloud Infrastructure (OCI) Ubuntu instance.

Initialization Script

The scripts/init.sh script performs the following actions:

1. Updates and upgrades system packages.
2. Configures iptables to allow traffic on essential ports:
   • 80 (TCP): HTTP traffic.
   • 443 (TCP/UDP): HTTPS traffic.
   • 3000 (TCP): Dokploy Dashboard.
3. Ensures iptables rules are persistent across reboots using iptables-persistent.
4. Installs Dokploy if it is not already present, or updates it if it is.

Usage

As Cloud-init User Data

When creating a new OCI instance, you can provide the contents of scripts/init.sh as the Cloud-init script (User Data) to automate the entire setup process.

Manual Execution

You can also run the script manually on an existing Ubuntu instance:

sudo ./scripts/init.sh

Managing Deployments with Dokploy

Once Dokploy is installed, you can access the dashboard at http://<your-instance-ip>:3000.

Dokploy allows you to easily manage and deploy:

• Docker Compose Stacks: Deploy complex multi-container applications by providing your docker-compose.yml directly in the Dokploy interface.
• Applications: Deploy web applications from GitHub, GitLab, or Bitbucket.
• Databases: Easily provision and manage PostgreSQL, MySQL, MongoDB, and Redis instances.