Environment separation

Deployment environments must use separate database state, credentials and explicitly selected application versions. This page describes the separation model; it is not a directory of internal hosts or a statement of current production status.

Respire target	Assigned URL
Website	https://rsrs.rs
User dashboard	https://dash.rsrs.rs
Admin dashboard	https://admin.rsrs.rs
API	https://api.rsrs.rs

These domains are configured separately from older products. Local API development uses http://127.0.0.1:8787; local identity/runtime state is rooted at ~/.rsrs with runtime port 15169. Import previous local accounts explicitly; migration retains the original directories.

Migration writes a .rsrs-migrated.json receipt in the source directory. A failure can occur after the destination has been created, including while writing that receipt. Check the source and destination receipts and migration result before retrying; do not delete either directory without verifying its contents.

flowchart TD
  Change["Reviewed change"] --> Test["Isolated test environment"]
  Test --> Staging["Staging validation"]
  Staging --> Approval["Deployment decision"]
  Approval --> Production["Production environment"]

Environment	Data	Purpose
Development	Synthetic or isolated data	Iteration and integration
Staging	Approved isolated copy	Upgrade and recovery rehearsal
Production	Live data	Explicitly controlled deployment

Resource	Separation requirement
Postgres	Separate database or volume and credentials
Secrets	Scoped to the intended environment
API and website	Route explicitly to their respective services
Images	Select immutable or explicitly versioned artifacts
Backups	Store independently; test restoration away from live data

Publishing a Git branch is separate from deploying a service. Inspect the actual repository workflow triggers before making deployment assumptions.

See self-hosting for deployment prerequisites.