Remove Nginx/Certbot and expose app container directly via --host-port

[SaravanakumarR2018]

Motivation

• The deployment flow should serve the application directly from the container on the server (no reverse proxy), so Nginx and Certbot provisioning/traffic switching should be removed.
• Simplify operations and make the script suitable for direct-to-server deployments by exposing a host port from the container.

Description

• Removed Nginx/Certbot provisioning, config generation, snippets, timers, and site-management logic from deploy/ebs-staging-prod.sh and stopped installing Nginx/certbot packages.
• Added a --host-port CLI option and HOST_PORT default (80) and removed the blue/green host port flags and cert/email requirements; --domain is now optional and only used for EBS environment inference.
• Replaced Nginx upstream switching with a simple active-color marker update written to ACTIVE_FILE, and updated switch_traffic, rollback_switch, and verify_domain to operate without a proxy by checking http://127.0.0.1:${HOST_PORT}.
• Changed Docker run port mapping to publish the container directly with -p "${HOST_PORT}:${CONTAINER_PORT}" while preserving the stop-first, health-check, and rollback flow.

Testing

• Ran a shell syntax check with bash -n deploy/ebs-staging-prod.sh, which completed successfully.

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: f826cd2479

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Persist active color across reboots

get_symlink_color now always returns empty, so whenever /var/run/${APP_NAME}-active-color is absent (which is common after reboot because /var/run is ephemeral), the script defaults ACTIVE_COLOR to green even if app_blue is the live container. In that state HAD_ACTIVE is computed against the wrong container name, and the deploy can delete the actually-running container as the “target” before the new image is proven healthy, which removes rollback safety and can leave no app running if startup fails.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Stop legacy nginx before binding direct host port

The new direct-to-server flow starts only Docker here, but does not stop/disable nginx from prior deployments. Because the script now publishes the app directly on ${HOST_PORT} (default 80), hosts upgraded from the previous nginx-based version can fail deployment with a bind error (port is already allocated) when nginx is still listening on 80.

Useful? React with 👍 / 👎.