java - 如何从非结构化日志创建 cloudwatch 指标过滤器？

Question

考虑日志输出：

2018-07-09 14:59:48,938 INFO [req-id:aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa] [sqs-m-id:aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa] [bucket:my-bucket] [s3Key:folder/myFile.pdf] com.mycompany.MyJavaClass - some message with spaces
2018-07-09 14:59:48,938 INFO [req-id:bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb] [sqs-m-id:bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb] [bucket:my-bucket] [s3Key:folder/myOtherFile.pdf] com.mycompany.MyJavaClass - some other message also with spaces
2018-07-09 14:59:48,938 INFO [req-id:bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb] [sqs-m-id:bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb] [bucket:my-bucket] [s3Key:folder/myOtherFile.pdf] com.mycompany.MyJavaClass - ELAPSED_TIME_NS 1520093301
START RequestId: aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa Version: $LATEST

如何提取仅应用于ELAPSED_TIME_NS行的指标。并从中获取价值？

标签： javaamazon-web-servicesamazon-cloudwatchamazon-cloudwatchlogsamazon-cloudwatch-metrics

Answer 1

这种模式应该适合你：

[..., message=ELAPSED_TIME_NS, nanoseconds]

在这里查看更多详细信息：https ://docs.aws.amazon.com/AmazonCloudWatch/latest/logs/FilterAndPatternSyntax.html#extract-log-event-values