Hello,
We are collecting some server metric using cmkbeat agent.
The output is available in following format:
perf_data: heap=8308.406693;15045.98125;15837.875;;15837.875 nonheap=582.264664;0;0;;0
perf_data: generic_number=4;80;90;;
perf_data: sessions=6338;10000;20000;;
The goal is to have following fields extracted from "perf_data" field:
metrics.metric_name: value (e.g. metrics.heap: 8308.406693)
metrics.metric_name.warn: warn_value (e.g. metrics.heap.warn: 15045.98125)
metrics.metric_name.crit: crit_value (e.g. metrics.heap.crit: 15837.875)
metrics.metric_name.min: min_value
metrics.metric_name.max: max_value
Case some value is missing, don't create field.
The main issue is that there are more than 40 different metric types, so my question is if there is any way how to extract field names and values from field "perf_data" dynamically (simply not to create 40 different grok patterns...)?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.