# Sprint: SPRINT_20260117_029_Runbook_coverage_expansion
# Task: RUN-004 - Backup/Restore Runbook
# Backup and Restore Operations Runbook

Status: PRODUCTION-READY (2026-01-17 UTC)

## Scope
Comprehensive backup and restore procedures for all Stella Ops components including database, evidence locker, configuration, and secrets.

---

## Backup Architecture Overview

### Backup Components

| Component | Backup Type | Default Schedule | Retention |
|-----------|-------------|------------------|-----------|
| PostgreSQL | Full + WAL | Daily full, continuous WAL | 30 days |
| Evidence Locker | Incremental | Daily | 90 days |
| Configuration | Snapshot | Daily + on change | 90 days |
| Secrets | Encrypted snapshot | Daily | 30 days |
| Attestation Keys | Encrypted export | Weekly | 1 year |

### Storage Locations

- **Primary:** `/var/lib/stellaops/backups/` (local)
- **Secondary:** S3/Azure Blob/GCS (configurable)
- **Offline:** Removable media for air-gap scenarios

---

## Pre-flight Checklist

### Environment Verification
```bash
# Check backup service status
stella backup status

# Verify backup storage
stella doctor --check check.storage.backup

# List recent backups
stella backup list --last 7d

# Test backup restore capability
stella backup test-restore --latest --dry-run
```

### Metrics to Watch
- `stella_backup_last_success_timestamp` - Last successful backup
- `stella_backup_duration_seconds` - Backup duration
- `stella_backup_size_bytes` - Backup size
- `stella_restore_test_last_success` - Last restore test

---

## Standard Procedures

### SP-001: Create Manual Backup

**When:** Before upgrades, schema changes, or major configuration changes
**Duration:** 5-30 minutes depending on data volume

1. Create full system backup:
   ```bash
   stella backup create --full --name "pre-upgrade-$(date +%Y%m%d)"
   ```

2. Or create component-specific backup:
   ```bash
   # Database only
   stella backup create --type database --name "db-pre-migration"
   
   # Evidence locker only
   stella backup create --type evidence --name "evidence-snapshot"
   
   # Configuration only
   stella backup create --type config --name "config-backup"
   ```

3. Verify backup:
   ```bash
   stella backup verify --name "pre-upgrade-$(date +%Y%m%d)"
   ```

4. Copy to offsite storage (recommended):
   ```bash
   stella backup copy --name "pre-upgrade-$(date +%Y%m%d)" --destination s3://backup-bucket/
   ```

### SP-002: Verify Backup Integrity

**Frequency:** Weekly
**Duration:** 15-60 minutes

1. List backups for verification:
   ```bash
   stella backup list --unverified
   ```

2. Verify backup integrity:
   ```bash
   # Verify specific backup
   stella backup verify --name <backup-name>
   
   # Verify all unverified
   stella backup verify --all-unverified
   ```

3. Test restore (non-destructive):
   ```bash
   stella backup test-restore --name <backup-name> --target /tmp/restore-test
   ```

4. Record verification result:
   ```bash
   stella backup log-verification --name <backup-name> --result success
   ```

### SP-003: Restore from Backup

**CAUTION: This is a destructive operation**

#### Full System Restore

1. Stop all services:
   ```bash
   stella service stop --all
   ```

2. List available backups:
   ```bash
   stella backup list --type full
   ```

3. Restore:
   ```bash
   # Dry run first
   stella backup restore --name <backup-name> --dry-run
   
   # Execute restore
   stella backup restore --name <backup-name> --confirm
   ```

4. Start services:
   ```bash
   stella service start --all
   ```

5. Verify restoration:
   ```bash
   stella doctor --all
   stella service health
   ```

#### Component-Specific Restore

1. Database restore:
   ```bash
   stella service stop --service api,release-orchestrator
   stella backup restore --type database --name <backup-name> --confirm
   stella db migrate  # Apply any pending migrations
   stella service start --service api,release-orchestrator
   ```

2. Evidence locker restore:
   ```bash
   stella backup restore --type evidence --name <backup-name> --confirm
   stella evidence verify --mode quick
   ```

3. Configuration restore:
   ```bash
   stella backup restore --type config --name <backup-name> --confirm
   stella service restart --graceful
   ```

### SP-004: Point-in-Time Recovery (Database)

1. Identify target recovery point:
   ```bash
   # List WAL archives
   stella backup wal-list --after <start-date> --before <end-date>
   ```

2. Perform PITR:
   ```bash
   stella backup restore-pitr --to-time "2026-01-17T10:30:00Z" --confirm
   ```

3. Verify data state:
   ```bash
   stella db verify-integrity
   ```

---

## Backup Schedules

### Configure Backup Schedule

```bash
# View current schedule
stella backup schedule show

# Set database backup schedule
stella backup schedule set --type database --cron "0 2 * * *"

# Set evidence backup schedule
stella backup schedule set --type evidence --cron "0 3 * * *"

# Set configuration backup schedule
stella backup schedule set --type config --cron "0 4 * * *" --on-change
```

### Retention Policy

```bash
# View retention policy
stella backup retention show

# Set retention
stella backup retention set --type database --days 30
stella backup retention set --type evidence --days 90
stella backup retention set --type config --days 90

# Apply retention (cleanup old backups)
stella backup retention apply
```

---

## Incident Procedures

### INC-001: Backup Failure

**Symptoms:**
- Alert: `StellaBackupFailed`
- Missing recent backup

**Investigation:**
```bash
# Check backup logs
stella backup logs --last 24h

# Check disk space
stella doctor --check check.storage.diskspace,check.storage.backup

# Test backup operation
stella backup test --type database
```

**Resolution:**

1. **Disk space issue:**
   ```bash
   stella backup retention apply --force
   stella backup cleanup --expired
   ```

2. **Database connectivity:**
   ```bash
   stella doctor --check check.postgres.connectivity
   ```

3. **Permission issue:**
   - Check backup directory permissions
   - Verify service account access

4. **Retry backup:**
   ```bash
   stella backup create --type <failed-type> --retry
   ```

### INC-002: Restore Failure

**Symptoms:**
- Restore command fails
- Services not starting after restore

**Investigation:**
```bash
# Check restore logs
stella backup restore-logs --last-attempt

# Verify backup integrity
stella backup verify --name <backup-name>

# Check disk space
stella doctor --check check.storage.diskspace
```

**Resolution:**

1. **Corrupted backup:**
   ```bash
   # Try previous backup
   stella backup list --type <type>
   stella backup restore --name <previous-backup> --confirm
   ```

2. **Version mismatch:**
   ```bash
   # Check backup version
   stella backup info --name <backup-name>
   
   # Restore with migration
   stella backup restore --name <backup-name> --with-migration
   ```

3. **Disk space:**
   - Free space or expand volume
   - Restore to alternate location

### INC-003: Backup Storage Full

**Symptoms:**
- Alert: `StellaBackupStorageFull`
- New backups failing

**Immediate Actions:**
```bash
# Check storage
stella backup storage stats

# Emergency cleanup
stella backup cleanup --keep-last 3

# Delete specific old backups
stella backup delete --older-than 14d --confirm
```

**Resolution:**

1. **Adjust retention:**
   ```bash
   stella backup retention set --type database --days 14
   stella backup retention apply
   ```

2. **Expand storage:**
   - Add disk space
   - Configure offsite storage

3. **Archive to cold storage:**
   ```bash
   stella backup archive --older-than 30d --destination s3://archive-bucket/
   ```

---

## Disaster Recovery Scenarios

### DR-001: Complete System Loss

1. Provision new infrastructure
2. Install Stella Ops
3. Restore from offsite backup:
   ```bash
   stella backup restore --source s3://backup-bucket/latest-full.tar.gz --confirm
   ```
4. Verify all components
5. Update DNS/load balancer

### DR-002: Database Corruption

1. Stop services
2. Restore database from latest clean backup:
   ```bash
   stella backup restore --type database --name <last-known-good>
   ```
3. Apply WAL to near-corruption point (PITR)
4. Verify data integrity
5. Resume services

### DR-003: Evidence Locker Loss

1. Restore evidence from backup:
   ```bash
   stella backup restore --type evidence --name <backup-name>
   ```
2. Rebuild index:
   ```bash
   stella evidence index rebuild
   ```
3. Verify anchor chain:
   ```bash
   stella evidence anchor verify --all
   ```

---

## Offline/Air-Gap Backup

### Creating Offline Backup

```bash
# Create encrypted offline bundle
stella backup create-offline \
  --output /media/usb/stellaops-backup-$(date +%Y%m%d).enc \
  --encrypt \
  --passphrase-file /secure/backup-key

# Verify offline backup
stella backup verify-offline --input /media/usb/stellaops-backup-*.enc
```

### Restoring from Offline Backup

```bash
# Restore from offline backup
stella backup restore-offline \
  --input /media/usb/stellaops-backup-*.enc \
  --passphrase-file /secure/backup-key \
  --confirm
```

---

## Monitoring Dashboard

Access: Grafana → Dashboards → Stella Ops → Backup Status

Key panels:
- Last backup success time
- Backup size trend
- Backup duration
- Restore test status
- Storage utilization

---

## Evidence Capture

```bash
stella backup diagnostics --output /tmp/backup-diag-$(date +%Y%m%dT%H%M%S).tar.gz
```

---

## Escalation Path

1. **L1 (On-call):** Retry failed backups, basic troubleshooting
2. **L2 (Platform team):** Restore operations, schedule adjustments
3. **L3 (Architecture):** Disaster recovery execution

---

_Last updated: 2026-01-17 (UTC)_