OK, I'm trying to improve performance by replacing some join queries with stats, but struggling on a filter. I have the below query, two source types where the common field between events is 'Correlator' . In source_one I have fields 'Correlator', 'sysplex' and 'servername'. In source_detail I have 'Correlator', 'sysplex' and multiple other fields, the one for this data is Sample_NAME. 'servername' in source_one could have multiple values and I want to filter on a match so search servername=xyz* I've tried a number of ways and I can't seem to manage to limit results to a filter on 'servername' without losing everything else, 'sysplex' which is in both sourcetypes filters just fine. Any thoughts would be appreciated. index=my_index sourcetype=source_one OR sourcetype=source_detail sysplex=ABC* | stats values(SAMPLE_NAME) AS SampleName values(SAMPLE_TIME) AS SampleTime by Correlator,SampleTime | eval _time=strptime(SampleTime,"%Y-%m-%d %H:%M:%S.%N") | timechart span=1m count by SampleName
... View more