Aegis Orchestrator
Guides

Backup & Restore

Backup and restore procedures for AEGIS stateful services — PostgreSQL, OpenBao, SeaweedFS, and coordinated backup strategies.

Backup & Restore

AEGIS runs several stateful services that require regular backups. This guide covers backup and restore procedures for each service and a coordinated backup strategy.

PostgreSQL

PostgreSQL stores agent definitions, execution records, workflow state, and Keycloak data.

Backup

# Logical backup (recommended for portability)
podman exec aegis-database-postgres pg_dump -U aegis -d aegis -F c -f /tmp/aegis-backup.dump

# Copy backup from container
podman cp aegis-database-postgres:/tmp/aegis-backup.dump ./backups/aegis-$(date +%Y%m%d).dump

# All databases (including temporal, keycloak)
podman exec aegis-database-postgres pg_dumpall -U aegis > ./backups/all-databases-$(date +%Y%m%d).sql

Restore

# Restore from custom-format dump
podman exec -i aegis-database-postgres pg_restore -U aegis -d aegis -c /tmp/aegis-backup.dump

# Restore from SQL dump
cat ./backups/all-databases.sql | podman exec -i aegis-database-postgres psql -U aegis

Automated Schedule

# Add to crontab (daily at 2 AM)
0 2 * * * podman exec aegis-database-postgres pg_dump -U aegis -d aegis -F c -f /backups/aegis-$(date +\%Y\%m\%d).dump

OpenBao

OpenBao stores encrypted secrets, AppRole credentials, and KV data.

Backup

OpenBao uses file-based storage by default. Back up the data directory:

# Stop writes (optional but recommended for consistency)
# Back up the persistent volume
podman volume export aegis-openbao-data > ./backups/openbao-data-$(date +%Y%m%d).tar

Restore

# Restore the volume
podman volume import aegis-openbao-data ./backups/openbao-data-YYYYMMDD.tar

# Restart the secrets pod
make redeploy POD=secrets

After restoring OpenBao, you may need to unseal it. Keep your unseal keys in a secure, separate location from the backup.

SeaweedFS

SeaweedFS stores agent volume data (files created during execution).

Backup

# Export master metadata
podman volume export aegis-seaweedfs-master-data > ./backups/seaweedfs-master-$(date +%Y%m%d).tar

# Export volume data
podman volume export aegis-seaweedfs-volume-data > ./backups/seaweedfs-volume-$(date +%Y%m%d).tar

# Export filer data
podman volume export aegis-seaweedfs-filer-data > ./backups/seaweedfs-filer-$(date +%Y%m%d).tar

Restore

# Import volumes (stop storage pod first)
make teardown-pod POD=storage

podman volume import aegis-seaweedfs-master-data ./backups/seaweedfs-master-YYYYMMDD.tar
podman volume import aegis-seaweedfs-volume-data ./backups/seaweedfs-volume-YYYYMMDD.tar
podman volume import aegis-seaweedfs-filer-data ./backups/seaweedfs-filer-YYYYMMDD.tar

make deploy-pod POD=storage

Coordinated Backup Strategy

For a consistent backup across all services:

  1. Pause new executions — prevent new agent executions from starting
  2. Wait for in-flight executions to complete or timeout
  3. Back up PostgreSQL — captures all relational state
  4. Back up OpenBao — captures all secrets
  5. Back up SeaweedFS — captures all volume data
  6. Resume executions

Backup Script Example

#!/bin/bash
BACKUP_DIR="./backups/$(date +%Y%m%d-%H%M%S)"
mkdir -p "$BACKUP_DIR"

echo "Backing up PostgreSQL..."
podman exec aegis-database-postgres pg_dumpall -U aegis > "$BACKUP_DIR/all-databases.sql"

echo "Backing up OpenBao..."
podman volume export aegis-openbao-data > "$BACKUP_DIR/openbao-data.tar"

echo "Backing up SeaweedFS..."
podman volume export aegis-seaweedfs-master-data > "$BACKUP_DIR/seaweedfs-master.tar"
podman volume export aegis-seaweedfs-volume-data > "$BACKUP_DIR/seaweedfs-volume.tar"
podman volume export aegis-seaweedfs-filer-data > "$BACKUP_DIR/seaweedfs-filer.tar"

echo "Backup complete: $BACKUP_DIR"

Retention Policy

Backup TypeFrequencyRetention
PostgreSQLDaily30 days
OpenBaoDaily30 days
SeaweedFSWeekly90 days
Full coordinatedWeekly90 days

Verification

After restoring from backup, verify service health:

# Check all services
make validate

# Verify PostgreSQL data
podman exec aegis-database-postgres psql -U aegis -c "SELECT count(*) FROM agents;"

# Verify OpenBao
curl -s http://localhost:8200/v1/sys/health | jq .sealed

See Also

Multi-Tenancy Configuration

Configure and isolate multiple tenants using Keycloak realms and OpenBao namespaces.

Disaster Recovery

Recovery runbooks for AEGIS platform failure scenarios — pod failures, database corruption, secrets engine recovery, and full environment rebuild.

On this page

Backup & RestorePostgreSQLBackupRestoreAutomated ScheduleOpenBaoBackupRestoreSeaweedFSBackupRestoreCoordinated Backup StrategyBackup Script ExampleRetention PolicyVerificationSee Also