KafkaMQ 日志采集最佳实践

概述

Kafka 是由 LinkedIn 开发、后由Apache软件基金会维护的分布式流处理平台，采用Scala和Java编写。它本质是一个高吞吐、持久化的发布-订阅消息系统，专注于处理实时数据流（如用户行为日志、点击流等）。在收集日志的场景中，Kafka 可以作为一个消息中间件，用于接收、存储和转发大量的日志，链路，指标数据。

观测云

观测云是一款专为 IT 工程师打造的全链路可观测产品，它集成了基础设施监控、应用程序性能监控和日志管理，为整个技术栈提供实时可观察性。这款产品能够帮助工程师全面了解端到端的用户体验追踪，了解应用内函数的每一次调用，以及全面监控云时代的基础设施。此外，观测云还具备快速发现系统安全风险的能力，为数字化时代提供安全保障。

本实践主要是通过观测云消费 Kafka 队列收集到的日志数据，并将数据通过 Pipeline 进行字段提取和分类，便于用户对日志数据进行可视化分析。

部署 Kafka

目前 DataKit 支持的 Kafka 版本有 [ version:0.8.2 ~ 3.2.0 ]。

下载 3.2.0 版本，解压即可使用。

wget https://archive.apache.org/dist/kafka/3.2.0/kafka_2.13-3.2.0.tgz

1、启动 Zookeeper 服务

$ bin/zookeeper-server-start.sh config/zookeeper.properties

2、启动 KafkaServer

$ bin/kafka-server-start.sh config/server.properties

3、创建 Topic

创建名为 testlog 的 Topic 。

$ bin/kafka-topics.sh --create --topic testlog --bootstrap-server localhost:9092

4、启动 Producer

$ bin/kafka-console-producer.sh --topic testlog --bootstrap-server localhost:9092

部署 DataKit

DataKit 是一个开源的、跨平台的数据收集和监控工具，由观测云开发并维护。它旨在帮助用户收集、处理和分析各种数据源，如日志、指标和事件，以便进行有效的监控和故障排查。DataKit 支持多种数据输入和输出格式，可以轻松集成到现有的监控系统中。

登录观测云控制台，在「集成」 - 「DataKit」选择对应安装方式，当前采用 Linux 主机部署 DataKit。

开启 KafkaMQ 采集器

进入 DataKit 安装目录下 (默认是 /usr/local/datakit/conf.d/ ) 的 conf.d/kafkamq 目录，复制 kafkamq.conf.sample 并命名为 kafkamq.conf 。

类似如下：

-rwxr-xr-x 1 root root 2574 Apr 30 23:52 kafkamq.conf
-rwxr-xr-x 1 root root 2579 May  1 00:40 kafkamq.conf.sample

调整 kafkamq 采集器配置如下：

• addrs = ["localhost:9092"]，该文采集器 DataKit 和 Kafka 安装到同一台操作系统中，localhost 即可。
• kafka_version = "3.2.0"，该文使用 Kafka 的版本。
• [inputs.kafkamq.custom]，删除注释符号“#”。
• [inputs.kafkamq.custom.log_topic_map]，删除注释符号“#”。
• "testlog"="log.p"，testlog 为 Topic 的名字，log.p 为观测云 Pipeline 可编程数据处理器的日志字段提取规则配置。涉及的业务日志和 log.p 的内容详细见下面的《使用 Pipeline》。
• 其他一些配置说明：
  • group_id = "datakit-group"：消费者组名称，相同组内消费者共享分区消费进度。不同消费者组可独立消费同一主题。
  • assignor = "roundrobin"：分区轮询分配给消费者，​适合组内消费者订阅相同主题列表​，实现负载均衡。

注意：开启或调整 DataKit 的配置，需重启采集器（shell 下执行 datakit service -R）。

[[inputs.kafkamq]]
  addrs = ["localhost:9092"]
  # your kafka version:0.8.2 ~ 3.2.0
  kafka_version = "3.2.0"
  group_id = "datakit-group"
  # consumer group partition assignment strategy (range, roundrobin, sticky)
  

  ## rate limit.
  #limit_sec = 100
  ## sample
  # sampling_rate = 1.0

  ## kafka tls config
  # tls_enable = true
  ## PLAINTEXT/SASL_SSL/SASL_PLAINTEXT
  # tls_security_protocol = "SASL_PLAINTEXT"
  ## PLAIN/SCRAM-SHA-256/SCRAM-SHA-512/OAUTHBEARER,default is PLAIN.
  # tls_sasl_mechanism = "PLAIN"
  # tls_sasl_plain_username = "user"
  # tls_sasl_plain_password = "pw"
  ## If tls_security_protocol is SASL_SSL, then ssl_cert must be configured.
  # ssl_cert = "/path/to/host.cert"

  ## -1:Offset Newest, -2:Offset Oldest
  offsets=-1

  ## skywalking custom
  #[inputs.kafkamq.skywalking]
  ## Required: send to datakit skywalking input.
  #  dk_endpoint="http://localhost:9529"
  #  thread = 8 
  #  topics = [
  #    "skywalking-metrics",
  #    "skywalking-profilings",
  #    "skywalking-segments",
  #    "skywalking-managements",
  #    "skywalking-meters",
  #    "skywalking-logging",
  #  ]
  #  namespace = ""

  ## Jaeger from kafka. Please make sure your Datakit Jaeger collector is open!
  #[inputs.kafkamq.jaeger]
  ## Required: ipv6 is "[::1]:9529"
  #  dk_endpoint="http://localhost:9529"
  #  thread = 8 
  #  source: agent,otel,others...
  #  source = "agent"
  #  # Required: topics
  #  topics=["jaeger-spans","jaeger-my-spans"]

  ## user custom message with PL script.
  [inputs.kafkamq.custom]
    #spilt_json_body = true
    #thread = 8
    #storage_index = "" # NOTE: only working on logging collection

    ## spilt_topic_map determines whether to enable log splitting for specific topic based on the values in the spilt_topic_map[topic].
    #[inputs.kafkamq.custom.spilt_topic_map]
    #  "log_topic"=true
    #  "log01"=false
    [inputs.kafkamq.custom.log_topic_map]
       "test_log"="log.p"
    #  "log01"="log_01.p"
    #[inputs.kafkamq.custom.metric_topic_map]
    #  "metric_topic"="metric.p"
    #  "metric01"="rum_apm.p"
    #[inputs.kafkamq.custom.rum_topic_map]
    #  "rum_topic"="rum_01.p"
    #  "rum_02"="rum_02.p"

  #[inputs.kafkamq.remote_handle]
    ## Required
    #endpoint="http://localhost:8080"
    ## Required topics
    #topics=["spans","my-spans"]
    # send_message_count = 100
    # debug = false
    # is_response_point = true
    # header_check = false

  ## Receive and consume OTEL data from kafka.
  #[inputs.kafkamq.otel]
    #dk_endpoint="http://localhost:9529"
    #trace_api="/otel/v1/traces"
    #metric_api="/otel/v1/metrics"
    #trace_topics=["trace1","trace2"]
    #metric_topics=["otel-metric","otel-metric1"]
    #thread = 8 

  ## todo: add other input-mq

编写 Pipeline

log.p 规则内容：

data = load_json(message)
protocol = data["protocol"]
response_code = data["response_code"]
set_tag(protocol,protocol)set_tag(response_code,response_code)
group_between(response_code,[200,300],"info","status")
group_between(response_code,[400,499],"warning","status")
group_between(response_code,[500,599],"error","status")
time = data["start_time"]
set_tag(time,time)
default_time(time)

效果展示

发送业务日志样例

业务日志样例文件如下：

#info
{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":204,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-05-01T00:37:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}
#error
{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":504,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-05-01T00:39:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}
#warn
{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":404,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-05-01T00:38:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}

日志发送命令

在 Producer 启动后，分别发送如下三条日志内容，三条日志一条为 info 级别（"response_code":204），另一条为 error 级别（"response_code":504)，最后一条为 warn 级别日志（"response_code":404）。

>{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":204,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-04-30T08:47:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}
>{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":504,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-04-30T08:47:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}
>{"protocol":"HTTP/1.1","upstream_local_address":"172.20.32.97:33878","response_flags":"-","istio_policy_status":null,"trace_id":"5532224c1013b9ad6da1efe88778dd64","authority":"server:1338","method":"PUT","response_code":404,"duration":83,"upstream_service_time":"83","user_agent":"Jakarta Commons-HttpClient/3.1","bytes_received":103,"downstream_local_address":"172.21.2.130:1338","start_time":"2024-04-30T08:47:11.230Z","upstream_transport_failure_reason":null,"requested_server_name":null,"bytes_sent":0,"route_name":"routes","x_forwarded_for":"10.0.0.69,10.23.0.31","upstream_cluster":"outbound|1338|svc.cluster.local","request_id":"80ac7d31-a598-4dc8-bb74-1850593f61e4","downstream_remote_address":"10.23.0.31:0","path":"/api/dimensions/items","upstream_host":"172.20.9.101:1338"}

效果

• 通过 DataKit 采集到 Kafka 的三条业务日志

• 使用 Pipeline 对日志进行字段提取的效果展示

下图 protocol、response_code 以及 time 都是使用 Pipeline 提取后的效果。

结语

观测云通过集成 KafkaMQ ，实现了 Kafka 队列日志数据的高效采集和处理，并结合观测云的 Pipeline 功能，能够实时采集业务日志并进行字段提取和分类，便于后续分析和可视化；此外，DataKit 的 KafkaMQ 采集器可扩展应用于其他数据处理场景，如还支持链路（如开源 otel，skywalking，jaeger），指标，RUM 等数据的消费，这种集成方案提升了系统的可观测性，同时反映了观测云平台的开放和包容性，加速了企业的数字化转型。