Open hayatbehlim opened 8 years ago
@hayatbehlim thank you for opening this issue.
I will try to find some time and investigate why this is happening.
You are welcome @anthonycorbacho , If you need any implementation side help I'm ready for that.
@hayatbehlim we only parse the first line using with grok, whenever we can.
What reason is this?I have the same problem @hayatbehlim
Did you try to use that: https://blog.codecentric.de/en/2017/09/jvm-fire-using-flame-graphs-analyse-performance/ to identify the problem ?
Parsing multi-line logs use high cpu near to 100%. There is no way to find out which grok filter taking more cpu cycle and How we can optimize it. This issue looks similar to logstash grok issue 1.https://gist.github.com/jordansissel/4646783