specweave 0.18.1 → 0.20.1

package/plugins/specweave-kafka/templates/migration/mirrormaker2-config.properties

@@ -0,0 +1,234 @@
+# MirrorMaker 2 Configuration
+# Replicates topics between source and target Kafka clusters
+
+# ========================================
+# CLUSTER CONNECTION CONFIGURATION
+# ========================================
+
+# Source cluster (primary)
+clusters = source, target
+
+source.bootstrap.servers = source-kafka-1:9092,source-kafka-2:9092,source-kafka-3:9092
+source.security.protocol = SASL_SSL
+source.sasl.mechanism = SCRAM-SHA-512
+source.sasl.jaas.config = org.apache.kafka.common.security.scram.ScramLoginModule required \
+    username="source-user" \
+    password="source-password";
+
+# Target cluster (replica)
+target.bootstrap.servers = target-kafka-1:9092,target-kafka-2:9092,target-kafka-3:9092
+target.security.protocol = SASL_SSL
+target.sasl.mechanism = SCRAM-SHA-512
+target.sasl.jaas.config = org.apache.kafka.common.security.scram.ScramLoginModule required \
+    username="target-user" \
+    password="target-password";
+
+# ========================================
+# REPLICATION FLOWS
+# ========================================
+
+# Define replication flows (source -> target)
+source->target.enabled = true
+
+# Topic whitelist (regex)
+# Replicate all topics except internal topics
+source->target.topics = .*
+source->target.topics.blacklist = __.*,_.*
+
+# Create topics on target with same configuration
+source->target.sync.topic.configs.enabled = true
+source->target.sync.topic.configs.interval.seconds = 60
+
+# Sync ACLs from source to target
+source->target.sync.topic.acls.enabled = true
+source->target.sync.topic.acls.interval.seconds = 600
+
+# ========================================
+# OFFSET SYNC CONFIGURATION
+# ========================================
+
+# Sync consumer group offsets from source to target
+# Allows consumers to fail over to target cluster
+source->target.emit.checkpoints.enabled = true
+source->target.emit.checkpoints.interval.seconds = 60
+
+# Offset translation (source partition -> target partition)
+source->target.offset-syncs.topic.replication.factor = 3
+source->target.offset-syncs.topic.location = target
+
+# Sync interval for offset mapping
+source->target.sync.group.offsets.enabled = true
+source->target.sync.group.offsets.interval.seconds = 60
+
+# Consumer groups to replicate (regex)
+source->target.groups = .*
+source->target.groups.blacklist = console-consumer-.*,connect-.*
+
+# ========================================
+# CHECKPOINT CONFIGURATION
+# ========================================
+
+# Checkpoint storage (Kafka topic on target cluster)
+source->target.checkpoints.topic.replication.factor = 3
+checkpoints.topic.replication.factor = 3
+
+# Emit heartbeats to track replication lag
+source->target.emit.heartbeats.enabled = true
+source->target.emit.heartbeats.interval.seconds = 5
+heartbeats.topic.replication.factor = 3
+
+# ========================================
+# PERFORMANCE TUNING
+# ========================================
+
+# Tasks (parallelism)
+# Higher = faster replication, more resource usage
+tasks.max = 10
+
+# Replication throughput
+# Max bytes/sec per task (0 = unlimited)
+source->target.max.tasks = 10
+source->target.throughput.bytes.per.second = 104857600  # 100 MB/s
+
+# Consumer fetch size
+source->target.consumer.max.poll.records = 1000
+source->target.consumer.fetch.max.bytes = 52428800  # 50 MB
+
+# Producer batch size
+source->target.producer.batch.size = 32768  # 32 KB
+source->target.producer.linger.ms = 10
+source->target.producer.compression.type = lz4
+
+# ========================================
+# REPLICATION POLICY
+# ========================================
+
+# Naming convention for replicated topics
+# Options: DefaultReplicationPolicy, IdentityReplicationPolicy
+replication.policy.class = org.apache.kafka.connect.mirror.DefaultReplicationPolicy
+
+# Topic name prefix (default: source cluster alias + ".")
+# Example: "orders" -> "source.orders"
+# Set to "" for IdentityReplicationPolicy (no prefix)
+replication.policy.separator = .
+
+# ========================================
+# INTERNAL TOPICS CONFIGURATION
+# ========================================
+
+# MirrorMaker 2 internal topics
+# These store checkpoints, offsets, and metadata
+offset-syncs.topic.replication.factor = 3
+heartbeats.topic.replication.factor = 3
+
+# Retention for internal topics
+offset.storage.topic = mm2-offset-syncs.source.internal
+offset.storage.replication.factor = 3
+offset.storage.partitions = 25
+
+config.storage.topic = mm2-configs.source.internal
+config.storage.replication.factor = 3
+
+status.storage.topic = mm2-status.source.internal
+status.storage.replication.factor = 3
+status.storage.partitions = 5
+
+# ========================================
+# FAILOVER CONFIGURATION
+# ========================================
+
+# When failing over from source to target:
+# 1. Stop producers on source
+# 2. Wait for replication lag to reach 0
+# 3. Redirect consumers to target cluster
+# 4. Consumers will resume from last committed offset (synced by MirrorMaker 2)
+
+# Monitor replication lag via heartbeats topic:
+# kafka-console-consumer --bootstrap-server target:9092 \
+#   --topic heartbeats \
+#   --from-beginning
+
+# Check offset sync status:
+# kafka-console-consumer --bootstrap-server target:9092 \
+#   --topic mm2-offset-syncs.source.internal \
+#   --from-beginning
+
+# ========================================
+# BIDIRECTIONAL REPLICATION (ACTIVE-ACTIVE)
+# ========================================
+
+# For active-active setups, enable reverse flow:
+# target->source.enabled = true
+# target->source.topics = .*
+# target->source.topics.blacklist = __.*,_.*,source.*
+
+# WARNING: Bidirectional replication requires careful design
+# to avoid replication loops. Use topic prefixes to prevent loops.
+
+# ========================================
+# MONITORING
+# ========================================
+
+# MirrorMaker 2 exposes JMX metrics:
+# - kafka.connect.mirror:type=MirrorSourceConnector,target=*,topic=*,partition=*
+#   - record-count (total records replicated)
+#   - record-age-ms (replication lag in ms)
+#   - replication-latency-ms (end-to-end latency)
+#
+# - kafka.connect.mirror:type=MirrorCheckpointConnector
+#   - checkpoint-latency-ms (offset sync lag)
+
+# Prometheus JMX Exporter config:
+# - Check replication lag: record-age-ms < 5000 (5s)
+# - Alert on high lag: record-age-ms > 60000 (1 minute)
+
+# ========================================
+# USAGE EXAMPLES
+# ========================================
+
+# Start MirrorMaker 2 (standalone mode):
+# connect-mirror-maker.sh mirrormaker2-config.properties
+
+# Start MirrorMaker 2 (distributed mode):
+# 1. Start Connect workers on 3+ nodes
+# 2. Create MirrorSourceConnector via REST API
+# 3. Create MirrorCheckpointConnector via REST API
+# 4. Create MirrorHeartbeatConnector via REST API
+
+# Verify replication:
+# kafka-topics --bootstrap-server target:9092 --list | grep source.
+# kafka-consumer-groups --bootstrap-server target:9092 --list
+
+# ========================================
+# BEST PRACTICES
+# ========================================
+
+# 1. Replication Factor: Match source cluster RF (typically 3)
+# 2. Tasks: Start with 10, increase if lag is high
+# 3. Throughput Limits: Set per-task limit to prevent network saturation
+# 4. Topic Filters: Use whitelist to avoid replicating unnecessary topics
+# 5. Offset Sync: Enable for seamless consumer failover
+# 6. Monitoring: Track replication lag, alert if > 1 minute
+# 7. Testing: Test failover procedure regularly (quarterly)
+# 8. Security: Use SASL/SSL for both clusters
+# 9. Network: Ensure low latency between clusters (< 50ms)
+# 10. Disaster Recovery: Document failover/failback procedures
+
+# ========================================
+# COMMON ISSUES & SOLUTIONS
+# ========================================
+
+# Issue: High replication lag
+# Solution: Increase tasks.max, check network bandwidth
+
+# Issue: Offset sync not working
+# Solution: Verify emit.checkpoints.enabled = true, check consumer group names
+
+# Issue: Duplicate messages on failover
+# Solution: Use exactly-once semantics (transactional producers + read_committed consumers)
+
+# Issue: Topics not created on target
+# Solution: Enable sync.topic.configs.enabled, check topic ACLs
+
+# Issue: Replication loop (bidirectional)
+# Solution: Use topic prefixes, configure blacklist properly