链接：Elasticsearch Service Sink Connector for Confluent Platform | Confluent Documentation
链接：Apache Kafka
一、搭建测试环境
下载Elasticsearch Service Sink Connector
https://file.zjwlyy.cn/confluentinc-kafka-connect-elasticsearch-15.0.0.zip
为了方便，使用docker搭建kafka和elasticsearch。
docker run -d --name elasticsearch -e "discovery.type=single-node" -e ES_JAVA_OPTS="-Xms512m -Xmx512m" -p 9200:9200 -p 9300:9300 docker.elastic.co/elasticsearch/elasticsearch:7.17.1
docker run --user root -d --name kafka -p 9092:9092 -p 8083:8083 apache/kafka:3.9.1
confluentinc-kafka-connect-elasticsearch-15.0.0.zip文件复制到kafka容器里
docker cp confluentinc-kafka-connect-elasticsearch-15.0.0.zip kafka:/opt/connectors
进入kafka的容器
docker exec -it kafka /bin/bash
修改配置文件
vi /opt/kafka/config/connect-standalone.properties
plugin.path=/opt/connectors #修改为zip解压路径
解压zip
unzip confluentinc-kafka-connect-elasticsearch-15.0.0.zip
修改配置文件
vi /opt/connectors/confluentinc-kafka-connect-elasticsearch-15.0.0/etc/quickstart-elasticsearch.properties
# 基础配置
name=t-elasticsearch-sink
connector.class=io.confluent.connect.elasticsearch.ElasticsearchSinkConnector
tasks.max=3 # 根据分区数调整
topics=t-elasticsearch-sink
key.ignore=true
key.converter=org.apache.kafka.connect.storage.StringConverter
value.converter=org.apache.kafka.connect.json.JsonConverter
value.converter.schemas.enable=false
# ES连接配置
connection.url=http://192.168.1.1:9200 # 多节点负载均衡
type.name=_doc
index.name=t-elasticsearch-sink
#index.auto.create=true # 自动创建索引（或手动预创建）
schema.ignore=true
# 容错与错误处理
errors.tolerance=all
#errors.deadletterqueue.topic.name=dlq_t4_elasticsearch # 必须配置DLQ
#errors.deadletterqueue.context.headers.enable=true # 保留错误上下文
behavior.on.null.values=IGNORE # 跳过空值消息
# 性能优化
batch.size=2000 # 批量写入提升吞吐
max.in.flight.requests=5 # 并发请求数
max.retries=10 # 失败重试次数
retry.backoff.ms=5000 # 重试间隔
read.timeout.ms=10000 # 读超时
connection.timeout.ms=10000 # 连接超时
flush.timeout.ms=30000 # 刷新超时[2](@ref)启动Connector
#cd /opt/kafka/bin
#./connect-standalone.sh -daemon ../config/connect-standalone.properties /opt/connectors/confluentinc-kafka-connect-elasticsearch-15.0.0/etc/quickstart-elasticsearch.properties
二、查看Connector状态
curl -XGET http://localhost:8083/connectors/t-elasticsearch-sink/status #查看状态
curl -XGET http://localhost:8083/connectors/t-elasticsearch-sink/config #查看配置
curl -X DELETE http://localhost:8083/connectors/t-elasticsearch-sink/offsets #清理偏移量
curl -X DELETE http://localhost:8083/connectors/t-elasticsearch-sink #删除此connectors
三、测试写入
./kafka-topics.sh --bootstrap-server 127.0.0.1:9092 --list #查看topics
./kafka-topics.sh --delete --topic t-elasticsearch-sink #删除topic
./kafka-console-producer.sh --bootstrap-server 127.0.0.1:9092 --topic t-elasticsearch-sink #逐行写入消息
四、查看ES索引
curl http://127.0.0.1:9200/_cat/indices?v
