Logstash \u662f Elastic Stack (ELK\/Elastic Stack) \u7684\u6838\u5fc3\u6210\u5458\u4e4b\u4e00，\u626e\u6f14\u7740\u6570\u636e\u5904\u7406\u7ba1\u9053\u7684\u89d2\u8272\u3002\u4f60\u53ef\u4ee5\u628a\u5b83\u60f3\u8c61\u6210\u4e00\u4e2a\u529f\u80fd\u5f3a\u5927\u7684\u201c\u6570\u636e\u52a0\u5de5\u5382\u201d\u3002<\/p>\n
\u5b83\u7684\u4e3b\u8981\u5de5\u4f5c\u6d41\u7a0b\u662f：<\/p>\n
\u8fd9\u4e2a\u00a0Input -> Filter -> Output\u00a0\u7684\u6d41\u7a0b\u88ab\u79f0\u4e3a\u4e00\u4e2a\u7ba1\u9053 (Pipeline)\u3002<\/p>\n
Logstash \u7684\u8fd0\u884c\u975e\u5e38\u7b80\u5355，\u6838\u5fc3\u547d\u4ee4\u662f：<\/p>\n
# -f \u53c2\u6570\u6307\u5b9a\u914d\u7f6e\u6587\u4ef6
\nbin\/logstash -f <your_config_file.conf><\/p>\n
\u5728\u5f00\u59cb\u7f16\u5199\u590d\u6742\u914d\u7f6e\u524d，\u5148\u6765\u4e00\u4e2a\u6700\u7b80\u5355\u7684\u201cHello World\u201d\u793a\u4f8b\u3002<\/p>\n
\u521b\u5efa\u4e00\u4e2a\u540d\u4e3a\u00a0first-pipeline.conf\u00a0\u7684\u6587\u4ef6：<\/p>\n
# first-pipeline.conf<\/p>\n
# \u8f93\u5165：\u4ece\u6807\u51c6\u8f93\u5165（\u4f60\u7684\u952e\u76d8）\u63a5\u6536\u6570\u636e
\ninput {
\n stdin {}
\n}<\/p>\n
# \u8f93\u51fa：\u5c06\u6570\u636e\u683c\u5f0f\u5316\u540e\u8f93\u51fa\u5230\u6807\u51c6\u8f93\u51fa（\u4f60\u7684\u5c4f\u5e55）
\noutput {
\n stdout {
\n codec => rubydebug # rubydebug \u662f\u4e00\u79cd\u683c\u5f0f，\u8ba9\u8f93\u51fa\u66f4\u6613\u8bfb
\n }
\n}<\/p>\n
\u8fd0\u884c\u5b83：<\/p>\n
bin\/logstash -f first-pipeline.conf<\/p>\n
\u542f\u52a8\u540e，\u5728\u63a7\u5236\u53f0\u8f93\u5165 \u201chello logstash\u201d，\u7136\u540e\u6309\u56de\u8f66\u3002\u4f60\u4f1a\u770b\u5230\u7c7b\u4f3c\u4e0b\u9762\u7684\u8f93\u51fa：<\/p>\n
{
\n "message" => "hello logstash",
\n "@version" => "1",
\n "host" => "your-hostname",
\n "@timestamp" => 2023-10-27T08:30:00.123Z
\n}<\/p>\n
\u8fd9\u8bc1\u660e\u4f60\u7684 Logstash \u7ba1\u9053\u5df2\u7ecf\u6210\u529f\u8fd0\u884c\u3002message\u00a0\u662f\u4f60\u8f93\u5165\u7684\u5185\u5bb9，@version,\u00a0host,\u00a0@timestamp\u00a0\u662f Logstash \u81ea\u52a8\u4e3a\u6bcf\u4e2a\u4e8b\u4ef6\u6dfb\u52a0\u7684\u5143\u6570\u636e\u3002<\/p>\n
Logstash \u914d\u7f6e\u6587\u4ef6\u7531\u4e09\u4e2a\u4e3b\u8981\u90e8\u5206\u7ec4\u6210：input,\u00a0filter,\u00a0output\u3002<\/p>\n
# \u8f93\u5165\u6bb5
\ninput {
\n # … \u4e00\u4e2a\u6216\u591a\u4e2a\u8f93\u5165\u63d2\u4ef6\u914d\u7f6e …
\n}<\/p>\n
# \u8fc7\u6ee4\u6bb5（\u53ef\u9009）
\nfilter {
\n # … \u4e00\u4e2a\u6216\u591a\u4e2a\u8fc7\u6ee4\u63d2\u4ef6\u914d\u7f6e …
\n}<\/p>\n
# \u8f93\u51fa\u6bb5
\noutput {
\n # … \u4e00\u4e2a\u6216\u591a\u4e2a\u8f93\u51fa\u63d2\u4ef6\u914d\u7f6e …
\n}<\/p>\n
\u4f5c\u7528：\u5b9a\u4e49\u6570\u636e\u4ece\u54ea\u91cc\u6765\u3002<\/p>\n
\u5e38\u7528\u63d2\u4ef6\u793a\u4f8b：<\/p>\n
file: \u4ece\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6570\u636e，\u975e\u5e38\u9002\u5408\u8bfb\u53d6\u65e5\u5fd7\u6587\u4ef6\u3002\u5b83\u4f1a\u8bb0\u5f55\u8bfb\u53d6\u4f4d\u7f6e，\u4e0d\u4f1a\u91cd\u590d\u8bfb\u53d6\u3002<\/p>\n
input {
\n file {
\n path => "\/var\/log\/nginx\/access.log" # \u8981\u8bfb\u53d6\u7684\u6587\u4ef6\u8def\u5f84
\n start_position => "beginning" # \u4ece\u6587\u4ef6\u5f00\u5934\u5f00\u59cb\u8bfb\u53d6
\n sincedb_path => "\/dev\/null" # (\u53ef\u9009) \u5728\u5f00\u53d1\u6d4b\u8bd5\u65f6，\u7981\u6b62\u8bb0\u5f55\u8bfb\u53d6\u4f4d\u7f6e，\u65b9\u4fbf\u91cd\u590d\u6d4b\u8bd5
\n }
\n}\n <\/li>\n
beats: \u63a5\u6536\u6765\u81ea Filebeat\u3001Metricbeat \u7b49 Beats \u5bb6\u65cf\u6210\u5458\u53d1\u9001\u7684\u6570\u636e\u3002\u8fd9\u662f\u751f\u4ea7\u73af\u5883\u4e2d\u975e\u5e38\u63a8\u8350\u7684\u7ec4\u5408\u3002<\/p>\n
input {
\n beats {
\n port => 5044 # \u76d1\u542c\u7684\u7aef\u53e3
\n }
\n}\n <\/li>\n
tcp\u00a0\/\u00a0udp: \u76d1\u542c\u4e00\u4e2a TCP \u6216 UDP \u7aef\u53e3\u6765\u63a5\u6536\u6570\u636e\u3002<\/p>\n
input {
\n tcp {
\n port => 9999
\n codec => "json" # \u5047\u8bbe\u4f20\u5165\u7684\u6570\u636e\u662f JSON \u683c\u5f0f
\n }
\n}\n <\/li>\n
kafka: \u4ece Kafka \u4e3b\u9898\u4e2d\u6d88\u8d39\u6570\u636e\u3002<\/p>\n
input {
\n kafka {
\n bootstrap_servers => "kafka-server:9092"
\n topics => ["log-topic"]
\n }
\n}\n <\/li>\n
jdbc: \u901a\u8fc7 JDBC \u5b9a\u671f\u67e5\u8be2\u6570\u636e\u5e93\u3002<\/p>\n
input {
\n jdbc {
\n jdbc_driver_library => "\/path\/to\/mysql-connector-java.jar"
\n jdbc_driver_class => "com.mysql.jdbc.Driver"
\n jdbc_connection_string => "jdbc:mysql:\/\/db-server:3306\/mydatabase"
\n jdbc_user => "user"
\n jdbc_password => "password"
\n statement => "SELECT * from logs WHERE timestamp > :sql_last_value"
\n schedule => "*\/5 * * * *" # \u6bcf5\u5206\u949f\u6267\u884c\u4e00\u6b21
\n }
\n}\n <\/li>\n<\/ul>\n
\u4f5c\u7528：\u6570\u636e\u5904\u7406\u7684\u6838\u5fc3，\u5bf9\u4e8b\u4ef6\u8fdb\u884c\u89e3\u6790\u3001\u8f6c\u6362\u3001\u4e30\u5bcc\u3002<\/p>\n
\u5e38\u7528\u63d2\u4ef6\u793a\u4f8b：<\/p>\n
grok:\u00a0(\u6700\u91cd\u8981)\u00a0\u7528\u4e8e\u89e3\u6790\u975e\u7ed3\u6784\u5316\u6587\u672c\u6570\u636e，\u5c06\u5176\u8f6c\u6362\u4e3a\u7ed3\u6784\u5316\u6570\u636e\u3002\u5b83\u4f7f\u7528\u9884\u5b9a\u4e49\u7684\u6b63\u5219\u8868\u8fbe\u5f0f\u6a21\u5f0f\u3002<\/p>\n
\u6280\u5de7：\u4f7f\u7528\u00a0Kibana Grok Debugger\u00a0\u5de5\u5177\u53ef\u4ee5\u975e\u5e38\u65b9\u4fbf\u5730\u6d4b\u8bd5\u548c\u521b\u5efa\u4f60\u7684 Grok \u6a21\u5f0f\u3002<\/p>\n<\/li>\n<\/ul>\n<\/li>\n
mutate: \u7528\u4e8e\u5bf9\u5b57\u6bb5\u8fdb\u884c\u901a\u7528\u64cd\u4f5c，\u5982\u91cd\u547d\u540d\u3001\u5220\u9664\u3001\u66ff\u6362\u548c\u7c7b\u578b\u8f6c\u6362\u3002<\/p>\n
filter {
\n mutate {
\n # \u5c06 "response" \u5b57\u6bb5\u7684\u503c\u8f6c\u6362\u4e3a\u6574\u6570
\n convert => { "response" => "integer" }
\n # \u91cd\u547d\u540d "clientip" \u5b57\u6bb5\u4e3a "client_address"
\n rename => { "clientip" => "client_address" }
\n # \u5220\u9664\u4e0d\u9700\u8981\u7684\u5b57\u6bb5
\n remove_field => ["field_to_remove_1", "field_to_remove_2"]
\n # \u6dfb\u52a0\u65b0\u5b57\u6bb5
\n add_field => { "processed_by" => "logstash-server-1" }
\n }
\n}\n <\/li>\n
date: \u89e3\u6790\u5b57\u6bb5\u4e2d\u7684\u65f6\u95f4\u5b57\u7b26\u4e32，\u5e76\u7528\u5b83\u6765\u66ff\u6362 Logstash \u4e8b\u4ef6\u7684\u9ed8\u8ba4\u00a0@timestamp\u00a0\u5b57\u6bb5\u3002\u8fd9\u5bf9\u4e8e\u786e\u4fdd\u4e8b\u4ef6\u65f6\u95f4\u51c6\u786e\u6027\u81f3\u5173\u91cd\u8981\u3002<\/p>\n
# Nginx \u65e5\u5fd7\u4e2d\u7684\u65f6\u95f4\u683c\u5f0f: [27\/Oct\/2023:12:00:00 +0800]
\nfilter {
\n grok {
\n match => { "message" => "%{HTTPDATE:log_timestamp}" }
\n }
\n date {
\n # \u5339\u914d log_timestamp \u5b57\u6bb5\u4e2d\u7684\u65f6\u95f4
\n match => [ "log_timestamp", "dd\/MMM\/yyyy:HH:mm:ss Z" ]
\n # \u6210\u529f\u89e3\u6790\u540e，\u5220\u9664\u539f\u59cb\u7684 log_timestamp \u5b57\u6bb5
\n remove_field => [ "log_timestamp" ]
\n }
\n}\n <\/li>\n
json: \u89e3\u6790\u503c\u4e3a JSON \u5b57\u7b26\u4e32\u7684\u5b57\u6bb5\u3002<\/p>\n
# \u5982\u679c\u539f\u59cb message \u662f: { "user": "john", "action": "login" }
\nfilter {
\n json {
\n source => "message" # \u6307\u5b9a\u8981\u89e3\u6790\u7684\u5b57\u6bb5
\n }
\n}
\n# \u89e3\u6790\u540e，\u4e8b\u4ef6\u4f1a\u589e\u52a0 user \u548c action \u4e24\u4e2a\u5b57\u6bb5\n <\/li>\n
geoip: \u6839\u636e IP \u5730\u5740\u4fe1\u606f，\u6dfb\u52a0\u5730\u7406\u4f4d\u7f6e\u6570\u636e（\u5982\u56fd\u5bb6\u3001\u57ce\u5e02\u3001\u7ecf\u7eac\u5ea6）\u3002<\/p>\n
filter {
\n geoip {
\n source => "client_address" # \u6307\u5b9a\u5305\u542b IP \u5730\u5740\u7684\u5b57\u6bb5
\n }
\n}
\n# \u6210\u529f\u540e，\u4f1a\u6dfb\u52a0\u4e00\u4e2a geoip \u5b57\u6bb5，\u91cc\u9762\u5305\u542b location, country_name \u7b49\u4fe1\u606f\n <\/li>\n<\/ul>\n
\u6761\u4ef6\u5224\u65ad (Conditionals)<\/p>\n
filter\u00a0\u5757\u652f\u6301\u00a0if,\u00a0else if,\u00a0else\u00a0\u8bed\u53e5，\u53ef\u4ee5\u6839\u636e\u4e0d\u540c\u7c7b\u578b\u7684\u65e5\u5fd7\u6267\u884c\u4e0d\u540c\u7684\u5904\u7406\u903b\u8f91\u3002<\/p>\n
filter {
\n if [type] == "nginx-access" {
\n grok { … }
\n } else if [type] == "app-json" {
\n json { … }
\n } else {
\n # \u5176\u4ed6\u5904\u7406
\n }
\n}<\/p>\n
\u4f5c\u7528：\u5b9a\u4e49\u5904\u7406\u597d\u7684\u6570\u636e\u53d1\u9001\u5230\u54ea\u91cc\u3002<\/p>\n
\u5e38\u7528\u63d2\u4ef6\u793a\u4f8b：<\/p>\n
elasticsearch:\u00a0(\u6700\u5e38\u7528)\u00a0\u5c06\u6570\u636e\u53d1\u9001\u5230 Elasticsearch \u96c6\u7fa4\u3002<\/p>\n
output {
\n elasticsearch {
\n hosts => ["http:\/\/es-node1:9200", "http:\/\/es-node2:9200"] # ES \u96c6\u7fa4\u5730\u5740
\n index => "logstash-%{+YYYY.MM.dd}" # \u7d22\u5f15\u540d\u79f0，\u6309\u5929\u521b\u5efa
\n user => "elastic"
\n password => "your_password"
\n }
\n}\n <\/li>\n
stdout: \u8f93\u51fa\u5230\u6807\u51c6\u8f93\u51fa，\u4e3b\u8981\u7528\u4e8e\u8c03\u8bd5\u3002<\/p>\n
output {
\n stdout {
\n codec => rubydebug # \u683c\u5f0f\u5316\u8f93\u51fa，\u4fbf\u4e8e\u67e5\u770b
\n }
\n}\n <\/li>\n
file: \u5c06\u6570\u636e\u5199\u5165\u6587\u4ef6\u3002<\/p>\n
output {
\n file {
\n path => "\/tmp\/logstash_output.log"
\n }
\n}\n <\/li>\n
kafka: \u5c06\u6570\u636e\u53d1\u9001\u5230 Kafka\u3002<\/p>\n
output {
\n kafka {
\n bootstrap_servers => "kafka-server:9092"
\n topic_id => "processed_logs"
\n }
\n}\n <\/li>\n<\/ul>\n
\u76ee\u6807：<\/p>\n
\u521b\u5efa\u914d\u7f6e\u6587\u4ef6\u00a0nginx-pipeline.conf:<\/p>\n
# nginx-pipeline.conf<\/p>\n
input {
\n file {
\n path => "\/path\/to\/your\/nginx\/access.log"
\n start_position => "beginning"
\n # \u5728\u751f\u4ea7\u73af\u5883\u4e2d，\u5e94\u53bb\u6389\u4e0b\u9762\u8fd9\u884c\u6216\u6307\u5b9a\u4e00\u4e2a\u771f\u5b9e\u8def\u5f84
\n sincedb_path => "\/dev\/null"
\n }
\n}<\/p>\n
filter {
\n # 1. \u4f7f\u7528 grok \u89e3\u6790\u975e\u7ed3\u6784\u5316\u65e5\u5fd7
\n grok {
\n # COMBINEDAPACHELOG \u662f\u4e00\u4e2a\u5185\u7f6e\u7684\u6a21\u5f0f，\u9002\u7528\u4e8e\u5e38\u89c1\u7684 Nginx\/Apache \u65e5\u5fd7\u683c\u5f0f
\n # \u793a\u4f8b\u65e5\u5fd7: 192.168.1.1 – – [27\/Oct\/2023:14:10:30 +0800] "GET \/api\/v1\/users HTTP\/1.1" 200 512 "-" "Mozilla\/5.0…"
\n match => { "message" => "%{COMBINEDAPACHELOG}" }
\n }<\/p>\n
# 2. \u5982\u679c grok \u89e3\u6790\u6210\u529f，\u5219\u7ee7\u7eed\u5904\u7406
\n if "_grokparsefailure" not in [tags] {
\n # 2.1 \u89e3\u6790\u65f6\u95f4\u6233，\u5e76\u8bbe\u7f6e\u4e3a\u4e8b\u4ef6\u7684\u4e3b\u65f6\u95f4\u6233
\n date {
\n match => [ "timestamp", "dd\/MMM\/yyyy:HH:mm:ss Z" ]
\n }<\/p>\n
# 2.2 \u4f7f\u7528 geoip \u63d2\u4ef6，\u901a\u8fc7 clientip \u5b57\u6bb5\u83b7\u53d6\u5730\u7406\u4f4d\u7f6e\u4fe1\u606f
\n geoip {
\n source => "clientip"
\n }<\/p>\n
# 2.3 \u8f6c\u6362\u5b57\u6bb5\u7c7b\u578b
\n mutate {
\n convert => { "bytes" => "integer" }
\n convert => { "response" => "integer" }
\n }<\/p>\n
# 2.4 (\u53ef\u9009) \u89e3\u6790\u6210\u529f\u540e，\u53ef\u4ee5\u5220\u9664\u539f\u59cb\u7684 message \u5b57\u6bb5\u4ee5\u8282\u7701\u7a7a\u95f4
\n mutate {
\n remove_field => [ "message", "timestamp" ]
\n }
\n }
\n}<\/p>\n
output {
\n # \u8f93\u51fa\u5230 Elasticsearch
\n elasticsearch {
\n hosts => ["http:\/\/localhost:9200"]
\n index => "nginx-access-%{+YYYY.MM.dd}"
\n }<\/p>\n
# \u540c\u65f6\u8f93\u51fa\u5230\u63a7\u5236\u53f0，\u65b9\u4fbf\u8c03\u8bd5
\n stdout {
\n codec => rubydebug
\n }
\n}<\/p>\n
\u8fd0\u884c\u4e0e\u6d4b\u8bd5：<\/p>\n
\u68c0\u67e5\u914d\u7f6e：\u5728\u8fd0\u884c\u524d，\u5148\u68c0\u67e5\u914d\u7f6e\u6587\u4ef6\u8bed\u6cd5\u662f\u5426\u6b63\u786e\u3002<\/p>\n
bin\/logstash -f nginx-pipeline.conf –config.test_and_exit<\/p>\n
\u5982\u679c\u770b\u5230\u00a0Configuration OK，\u8bf4\u660e\u914d\u7f6e\u6ca1\u95ee\u9898\u3002<\/p>\n<\/li>\n
\u8fd0\u884c\u7ba1\u9053：<\/p>\n
bin\/logstash -f nginx-pipeline.conf\n <\/li>\n
\u751f\u6210\u65e5\u5fd7\u00a0(\u53ef\u9009)：\u5411\u4f60\u7684 Nginx \u8bbf\u95ee\u65e5\u5fd7\u6587\u4ef6\u4e2d\u8ffd\u52a0\u4e00\u884c\u65b0\u65e5\u5fd7，\u89c2\u5bdf Logstash \u63a7\u5236\u53f0\u7684\u8f93\u51fa\u3002 \u5904\u7406\u540e\u7684\u8f93\u51fa\u5e94\u8be5\u7c7b\u4f3c\u8fd9\u6837 (rubydebug \u683c\u5f0f)：<\/p>\n
{
\n "agent" => "\\\\"Mozilla\/5.0…\\\\"",
\n "clientip" => "192.168.1.1",
\n "verb" => "GET",
\n "request" => "\/api\/v1\/users",
\n "httpversion" => "1.1",
\n "response" => 200,
\n "bytes" => 512,
\n "geoip" => {
\n "ip" => "192.168.1.1",
\n "country_name" => "China",
\n "city_name" => "Beijing",
\n "location" => { "lon" => 116.3883, "lat" => 39.9289 }
\n },
\n "@timestamp" => 2023-10-27T06:10:30.000Z, \/\/ \u6ce8\u610f\u65f6\u95f4\u5df2\u7ecf\u8f6c\u4e3a UTC
\n …
\n}<\/p>\n
\u540c\u65f6，\u8fd9\u4e9b\u7ed3\u6784\u5316\u6570\u636e\u4e5f\u4f1a\u88ab\u53d1\u9001\u5230 Elasticsearch \u4e2d\u540d\u4e3a\u00a0nginx-access-YYYY.MM.DD\u00a0\u7684\u7d22\u5f15\u91cc\u3002<\/p>\n<\/li>\n
1. Logstash \u662f\u4ec0\u4e48？
\nLogstash \u662f Elastic Stack (ELK\/Elastic Stack) \u7684\u6838\u5fc3\u6210\u5458\u4e4b\u4e00，\u626e\u6f14\u7740\u6570\u636e\u5904\u7406\u7ba1\u9053\u7684\u89d2\u8272\u3002\u4f60\u53ef\u4ee5\u628a\u5b83\u60f3\u8c61\u6210\u4e00\u4e2a\u529f\u80fd\u5f3a\u5927\u7684\u201c\u6570\u636e\u52a0\u5de5\u5382\u201d\u3002
\n\u5b83\u7684\u4e3b\u8981\u5de5\u4f5c\u6d41\u7a0b\u662f：
\n\u6536\u96c6 (Input)：\u4ece\u5404\u79cd\u6765\u6e90（\u5982\u65e5\u5fd7\u6587\u4ef6\u3001\u6570\u636e\u5e93\u3001\u6d88\u606f\u961f\u5217\u3001\u7f51\u7edc\u7aef\u53e3\u7b49）\u6536\u96c6\u6570\u636e\u3002\u5904\u7406 (Filter)：\u5bf9\u6536\u96c6\u5230\u7684\u6570\u636e\u8fdb\u884c\u89e3\u6790\u3001\u8f6c\u6362\u3001<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[99],"topic":[],"class_list":["post-75111","post","type-post","status-publish","format-standard","hentry","category-server","tag-java"],"yoast_head":"\n