Enterprise Active-Active

Manage Active-Active (CRDB) databases for geo-distributed deployments.

Overview

Active-Active databases replicate data across multiple clusters with conflict-free resolution. Each cluster can handle local reads and writes with automatic synchronization.

CRDB Commands

List CRDBs

redisctl enterprise crdb list [OPTIONS]

Examples:

# List all CRDBs
redisctl enterprise crdb list

# Table format
redisctl enterprise crdb list -o table

# Get names and status
redisctl enterprise crdb list -q "[].{name:name,guid:guid,status:status}"

Get CRDB

redisctl enterprise crdb get <GUID> [OPTIONS]

Examples:

# Get CRDB details
redisctl enterprise crdb get abc-123-def

# Get instances
redisctl enterprise crdb get abc-123-def -q 'instances'

Create CRDB

redisctl enterprise crdb create --data <JSON>

Example Payload:

{
  "name": "global-cache",
  "memory_size": 1073741824,
  "port": 12000,
  "instances": [
    {
      "cluster": {
        "url": "https://cluster1.example.com:9443",
        "credentials": {
          "username": "admin@cluster.local",
          "password": "password1"
        }
      }
    },
    {
      "cluster": {
        "url": "https://cluster2.example.com:9443",
        "credentials": {
          "username": "admin@cluster.local",
          "password": "password2"
        }
      }
    }
  ]
}

Examples:

# Create from file
redisctl enterprise crdb create --data @crdb.json

# Simple two-cluster setup
redisctl enterprise crdb create --data '{
  "name": "geo-cache",
  "memory_size": 1073741824,
  "instances": [...]
}'

Update CRDB

redisctl enterprise crdb update <GUID> --data <JSON>

Examples:

# Increase memory
redisctl enterprise crdb update abc-123 --data '{"memory_size": 2147483648}'

# Add instance
redisctl enterprise crdb update abc-123 --data '{
  "instances": [
    {"cluster": {"url": "https://cluster3.example.com:9443", ...}}
  ]
}'

Delete CRDB

redisctl enterprise crdb delete <GUID>

CRDB Tasks

Active-Active operations are asynchronous and managed through tasks.

List CRDB Tasks

redisctl enterprise crdb-task list [OPTIONS]

Examples:

# All tasks
redisctl enterprise crdb-task list

# Tasks for specific CRDB
redisctl enterprise crdb-task list --crdb-guid abc-123

# Filter by status
redisctl enterprise crdb-task list -q "[?status=='completed']"

Get CRDB Task

redisctl enterprise crdb-task get <TASK-ID>

Example:

# Check task status
redisctl enterprise crdb-task get task-123 -q '{status:status,progress:progress}'

Cancel CRDB Task

redisctl enterprise crdb-task cancel <TASK-ID>

Instance Management

Get Instance Status

redisctl enterprise crdb get <GUID> -q 'instances[].{cluster:cluster.name,status:status}'

Add Instance to CRDB

redisctl enterprise crdb update <GUID> --data '{
  "add_instances": [{
    "cluster": {
      "url": "https://new-cluster:9443",
      "credentials": {...}
    }
  }]
}'

Remove Instance from CRDB

redisctl enterprise crdb update <GUID> --data '{
  "remove_instances": ["instance-id"]
}'

Common Patterns

Create Two-Region Active-Active

#!/bin/bash
# Create CRDB across two regions

cat > crdb-config.json << 'EOF'
{
  "name": "global-sessions",
  "memory_size": 2147483648,
  "port": 12000,
  "causal_consistency": false,
  "encryption": true,
  "instances": [
    {
      "cluster": {
        "url": "https://us-east.example.com:9443",
        "credentials": {
          "username": "admin@cluster.local",
          "password": "$US_EAST_PASSWORD"
        }
      }
    },
    {
      "cluster": {
        "url": "https://eu-west.example.com:9443",
        "credentials": {
          "username": "admin@cluster.local",
          "password": "$EU_WEST_PASSWORD"
        }
      }
    }
  ]
}
EOF

envsubst < crdb-config.json | redisctl enterprise crdb create --data @-

Monitor CRDB Sync Status

# Check all instances are synced
redisctl enterprise crdb get abc-123 \
  -q 'instances[].{cluster:cluster.name,lag:sync_lag}' \
  -o table

Wait for CRDB Task

TASK_ID=$(redisctl enterprise crdb create --data @crdb.json -q 'task_id')

while true; do
  STATUS=$(redisctl enterprise crdb-task get $TASK_ID -q 'status')
  echo "Status: $STATUS"
  
  case $STATUS in
    "completed") echo "Success!"; break ;;
    "failed") echo "Failed!"; exit 1 ;;
    *) sleep 10 ;;
  esac
done

Conflict Resolution

Active-Active uses CRDTs (Conflict-free Replicated Data Types) for automatic conflict resolution:

Troubleshooting

"Instance not reachable"

• Check network connectivity between clusters
• Verify firewall allows CRDB ports
• Check cluster credentials

"Sync lag increasing"

• Check network latency between clusters
• Verify cluster resources (CPU, memory)
• Check for large write volumes

"Task stuck"

• Check all instances are healthy
• Cancel and retry: redisctl enterprise crdb-task cancel
• Check cluster logs for errors

"Conflict resolution issues"

• Review data types being used
• Consider causal consistency if needed
• Check application logic for proper CRDT usage

API Reference

REST endpoints:

• GET /v1/crdbs - List CRDBs
• POST /v1/crdbs - Create CRDB
• GET /v1/crdbs/{guid} - Get CRDB
• PUT /v1/crdbs/{guid} - Update CRDB
• DELETE /v1/crdbs/{guid} - Delete CRDB
• GET /v1/crdb_tasks - List tasks
• GET /v1/crdb_tasks/{id} - Get task

For direct API access: redisctl api enterprise get /v1/crdbs