Denis Roy, Head of Information Technology, Eclipse Foundation
As the Open VSX ecosystem continues to grow, keeping the registry stable is a top priority. Behind the scenes, we are strengthening the infrastructure so that even during peak loads or major provider outages, developer workflows remain uninterrupted.
In recent posts, we shared how the Open VSX Registry is strengthening supply-chain security with pre-publish checks and introducing operational guardrails through rate limiting to scale responsibly. As adoption and usage increase, the underlying infrastructure behind those improvements becomes just as important. This post focuses on that work: improving availability, reducing single points of failure, and making recovery faster and more predictable when incidents occur.
A hybrid, fail-safe architecture
We are currently transitioning to a hybrid infrastructure model, moving core services to AWS as our primary environment, while keeping our on-premise infrastructure fully operational as a secondary site.
This is deliberate architectural diversity. AWS provides scale and flexibility. Our on-premise environment provides an independent fallback. If a cloud region experiences an outage, services can shift to infrastructure under our direct control.
The objective is simple: keep the registry online even when part of the underlying environment is not.
High-availability storage
Compute alone does not keep a registry running. The data must be available wherever the service is active.
As part of our infrastructure improvement plan, we are adding a dedicated fallback storage cluster and synchronizing extension binaries and metadata across locations. This reduces reliance on any single storage layer and prevents situations where one environment is healthy but lacks the data it needs.
If one storage layer becomes unreachable, the other is ready to step in.
Seeing issues before they become outages
Reducing downtime starts with visibility.
We are modernizing our observability stack across both cloud and on-prem environments, strengthening monitoring, centralized logging, and real-time alerting. This makes it easier to detect slowdowns, rising error rates, or unusual traffic patterns before they impact users.
Earlier detection leads to faster resolution and fewer user-visible incidents.
Faster recovery through clearer process
Technology improves reliability. Process makes it consistent.
We are formalizing incident response and recovery procedures for our multi-site architecture. Updated runbooks and rehearsed failover scenarios reduce mean time to recovery and remove uncertainty during high-pressure events.
When something does go wrong, clarity and speed make all the difference.
Why this work matters
The Open VSX Registry now supports a rapidly expanding ecosystem of developer platforms, CI systems, and AI-enabled tools. Growth brings higher expectations for uptime and reliability.
These infrastructure improvements are a long-term investment in keeping the Open VSX Registry stable, secure, and dependable as it scales.
Security builds trust. Operational guardrails support sustainability. Infrastructure upgrades ensure the service remains available when it matters most.
The Open VSX Registry is shared public infrastructure. Keeping it reliable requires continuous investment, thoughtful architecture, and disciplined operations. This work strengthens the registry so developers, publishers, and platform providers can rely on it with confidence, today and as the ecosystem continues to evolve.
It’s a team effort
This work reflects the effort of many people across the Eclipse Foundation and the broader Open VSX community. From the IT teams to Software Development, Security and beyond, including our community of users, developers, testers and integrators, all have contributed to making Open VSX a world‑class, high‑value extension registry that continues to grow through focused stewardship, open collaboration, and a commitment to empowering developers everywhere.
We also appreciate the collaboration of our cloud and infrastructure partners who continue to support the reliability and performance of the Open VSX Registry.
