Thanks for providing the sample data & regexes. That's very helpful.
TL;DR try something like %{MMDD:mmdd}.+%{THREAD:keyword} with these patterns
MMDD [0-1][0-9]/[0-3][0-9]
THREAD _\w+
note: I think the THREAD regex had an extra character and should should be THREAD _\w+ instead of THREAD _W\w+
The reason they worked individually but not together is because %{MMDD:mmdd} %{THREAD:keyword} translates to "find MMDD regex, followed by a single space character , followed by the THREAD regex"
Look at the results if we change the input to something like that (date SPACE id)
I was under the impression that each space-separated grok filter was a separate little regex expression, and that the entire line was separately scanned for each.
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.