Skip to main content

Service Queue Information

Overview​

APIs, apps, and game servers within the service queue are run directly on my own home lab. A brief overview is provided below.

Resource Limits​

There are no hard resource limits when availing of services via the service queue. Hosting is catered to on a first-come-first-serve basis.

Resources in each server however are shared amongst other services, based on the server node you are assigned to. Internet traffic is shared amongst all servers, with speeds of up to 80mbps. There is no ingress/egress limit.

In general, each server will have the following minimum specs

  • 8-core processor
  • 64GB DDR4
  • 1TB NVME storage

Ingress/Egress​

Internet traffic is shared amongst all servers across 4 ISPs, with speeds of up to 80mbps for each ISP connection. There is no ingress/egress limit.

Traffic is usually pinned on the first ISP for lower traffic services. If you have specific bandwidth or requirements, please have this outlined in your request.

Backups/Continuity​

Hosted services are backed up using Borg

  • Two local backup locations
  • Backup snapshots taken every 8 hours

Server nodes are also backed by a UPS-capable power station, allowing for continued runtime for up to an hour during blackouts.

Servers are unaffected by blackouts occurring between 8 AM to 3 PM thanks to PV systems.

CI/CD Pipelines​

For frontend and backend applications hosted locally, automated deployments are possible via Woodpecker. This is preferred over GitHub Actions or similar platforms.

Sample .yml files are provided in this repository for starters.