# Continuity & Recovery Managed documentation for Continuity & Recovery. # Backup and Recovery # Backup and Recovery | Field | Value | | --- | --- | | Distribution | BSS — OpenSOF | | Product | OpenSOF | | Release | 0.15.1 | | Deployment | `opensof.bss.dev` | | Source | `/srv/bss/releases/OpenSOF-v0.15.0-20app` | | Evidence | Continuity standard | | Source fingerprint | `cda583b42f2a0d296b268ab5b174932a378c8be6fb52ffa1af14af896e990f6e` | | Status | Generated baseline — human review required | > **Verification boundary:** This page combines platform design guidance with static evidence from the release. It does not prove that every detected interface is enabled, reachable, secure, or operational in the deployed environment. ## Backup scope | Asset | Backup requirement | Restore verification | | --- | --- | --- | | Release artifacts | Retain exact immutable packages/checksums; normally do not back up regenerated dependencies. | Checksum and extraction test. | | Configuration | Versioned protected copy excluding or separately encrypting secrets. | Syntax test and controlled comparison. | | Databases | Consistent database-native dump/snapshot with schema/version metadata. | Restore into isolated environment and run integrity/query checks. | | Files/evidence | Content plus metadata, ownership, access policy, checksums and provenance. | Sample restore, checksum and access-control test. | | Knowledge/ontology | Graph data, ontology versions, shapes, indexes and configuration. | SPARQL/constraint tests against restored graph. | | Audit/history | Protected retention according to policy; preserve ordering and integrity evidence. | Query selected events and verify integrity chain. | ## Recovery order 1. Infrastructure, DNS/TLS and time. 2. Protected configuration and secrets. 3. Databases, knowledge and artifact stores. 4. OpenBus/shared backends. 5. Edition-specific applications and reverse proxy. 6. Sensors/edge agents and external integrations. 7. End-to-end acceptance tests and user access. ## Recovery objective Define and test RPO/RTO by data class. A backup that has never been restored is an optimistic file collection, not a recovery capability. --- ## Maintainer Notes Add human-reviewed deployment notes, corrections, decisions, screenshots, and links here. Content outside the generated block is preserved on future runs. # Troubleshooting # Troubleshooting | Field | Value | | --- | --- | | Distribution | BSS — OpenSOF | | Product | OpenSOF | | Release | 0.15.1 | | Deployment | `opensof.bss.dev` | | Source | `/srv/bss/releases/OpenSOF-v0.15.0-20app` | | Evidence | Operational diagnostic standard | | Source fingerprint | `cda583b42f2a0d296b268ab5b174932a378c8be6fb52ffa1af14af896e990f6e` | | Status | Generated baseline — human review required | > **Verification boundary:** This page combines platform design guidance with static evidence from the release. It does not prove that every detected interface is enabled, reachable, secure, or operational in the deployed environment. ## Diagnostic sequence 1. State the exact user-visible symptom, start time, scope and last known good state. 2. Check DNS/routing/time before assuming an application defect. 3. Check process/container state and recent logs. 4. Test the backend locally on its bound address/port. 5. Test proxy/TLS path and WebSocket upgrade separately. 6. Test dependencies: database, OpenBus, OpenKnowledge, OpenFiles and external services. 7. Reproduce one primary workflow with correlation identifiers and capture evidence. 8. Apply the smallest reversible change, then retest all layers. ## Symptom matrix | Symptom | Likely layer | First checks | | --- | --- | --- | | Connection refused | Process/socket | `systemctl status`, `ss -lntup`, local curl | | Empty response/reset | App/upstream crash or protocol mismatch | Application logs, internal HTTP response, proxy protocol. | | 502/504 | Reverse proxy/upstream | `nginx -t`, error log, `proxy_pass`, backend latency | | Page loads but live data fails | WebSocket/CORS/auth | Browser network console, upgrade headers, token/role, route base. | | Wrong distribution/branding | Nginx/service instance | Hostname routing, service unit, current release symlink, environment. | | Tracks/tasks/products missing | Dependency/data flow | Producer logs, OpenBus/API, authorization, semantic type, time window. | | Intermittent correlation error | Time/PNT/data quality | Clock offset, source timestamps, coordinate frame and calibration. | ## Evidence rule Before changing configuration, preserve the relevant logs, status output, request/response, release version and timestamps. Troubleshooting without evidence tends to become configuration roulette—with worse odds and fewer free drinks. --- ## Maintainer Notes Add human-reviewed deployment notes, corrections, decisions, screenshots, and links here. Content outside the generated block is preserved on future runs.