正常采集过程中不定时发生日志重采 #1275
Answered
by
linrunqi08
apothecary-ap
asked this question in
Help
正常采集过程中不定时发生日志重采
#1275
-
在集群环境使用deamonset启动ilogtail,日志量相同,采集配置相同的集群中,发现单独一台ilogtail容器内存比其它内存高,日志采集重复 ilogtial配置了多个采集配置,ilogtail_config.json为通用配置,以下只列了发生重采的采集配置 ilogtail_config.json{ "cpu_usage_limit": 1, "mem_usage_limit": 2048, "polling_ignore_file_modify_timeout": 7200, "polling_file_first_watch_timeout": 600, "check_point_dump_interval": 5 } ilogtail采集配置enable: true inputs: - Type: file_log LogPath: /logs/job-manager/serviceLog/ FilePattern: '*.opsMonitorLog.*.log' EnableLogPositionMeta: true processors: - Type: processor_add_fields Fields: logtype: generallog domain: aiops servicename: aiops-demo agenthost: localhost ilogtail: ilogtail env: normal flushers: - Type: flusher_kafka_v2 Compression: gzip Brokers: - XX.XX.XXX.XXX:XXXX Convert: ProtocolFieldsRename: contents: fields Topic: demo 重采发现的日志生成方式:每分钟打印20条数据,按小时进行文件切分 文件命名方式:xxx.日期-小时.log(jobmanager.opsMonitorLog.2023-12-19-06.log) 以下为部分debug日志: 发生重采的时候都有unregister |
Beta Was this translation helpful? Give feedback.
Answered by
linrunqi08
Dec 25, 2023
Replies: 1 comment 8 replies
-
内存如果满了,会导致logtail进程被oom kill,内存中的checkpoint无法记录下来,从而导致重复 |
Beta Was this translation helpful? Give feedback.
8 replies
Answer selected by
Takuka0311
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
内存如果满了,会导致logtail进程被oom kill,内存中的checkpoint无法记录下来,从而导致重复