Reducing Artemis’ deployment downtime with blue-green deployments

2026-02-06 · Source: James' Coffee Blog · Field: Technology & Digital — Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, short

Summary

Artemis, a web reader application, implemented blue-green deployments to significantly reduce downtime during software updates. Previously, restarting the systemd process for updates caused 5-10 seconds of unavailability, displaying a maintenance message. The new approach involves running two Artemis processes: a primary ("blue") and a fallback ("green"). During an update, the new code is deployed, the blue process restarts, and the green process temporarily handles requests. Once the blue process is back online, requests revert to it, and the green process is then updated. This setup uses Nginx to route traffic between `localhost:9000` (blue) and `localhost:9001` (green), with the green fallback designed as read-only to simplify its implementation and testing, disabling account pages and forms.

Key takeaway

For DevOps Engineers managing web applications with systemd, adopting a blue-green deployment strategy can drastically cut user-facing downtime during updates. You should configure your web server, like Nginx, to route traffic between primary and fallback instances, potentially making the fallback read-only to streamline its implementation and reduce testing complexity. This approach improves user experience by avoiding full service interruptions.

Key insights

Blue-green deployment minimizes application downtime by using a fallback instance during updates.

Principles

• Isolate primary and fallback deployments.
• Route traffic dynamically between versions.

Method

Deploy new code, restart primary process while fallback handles requests, revert traffic to primary, then update fallback process. Nginx routes traffic to `localhost:9000` (blue) and `localhost:9001` (green).

In practice

• Use Nginx `upstream` for traffic routing.
• Implement fallback as read-only for simplicity.
• Use environment variables to control deployment mode.

Topics

• Blue-Green Deployment
• Nginx Configuration
• Web Application Deployment
• Downtime Reduction
• Flask Framework

Best for: Software Engineer, DevOps Engineer, MLOps Engineer

Related on AIssential

• Zero-Downtime Patching in Lakebase Part 1: Prewarming — Lakebase is addressing planned database maintenance, which causes more workload disruption than unplanned failures, by aiming for "zero-downtime patching." This initial post details "prewarming," a te…
• Nvidia wants to cut data center water use, but that’s not the same as fixing AI’s water problem — Nvidia's new cooling system addresses only a fraction of AI data centers' total water footprint by ignoring external electricity generation.
• Must- Know Deployment Strategies: From Big-Bang to Progressive Delivery — Deployment strategies primarily aim to reduce risk and control user exposure during code releases.
• How to achieve zero-downtime updates in large-scale AI agent deployments — AI agent "zero-downtime" requires functional uptime, not just system uptime, due to their non-deterministic, stateful nature.
• How we replaced Ingress-NGINX at Stack Overflow — Migrating from Ingress-NGINX to Gateway API requires thorough evaluation of conformant implementations and their performance under load.
• Why Artemis is invite-only — Invite-only access for hobby projects ensures sustainability, controlled growth, and a better user experience.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by James' Coffee Blog.