How to count each log value separately?
("*error*","*info*","*warn*")
Hi @surens ,
ok, please try this:
index = aws_ubs_n log IN ("*error*","*info*","*warn*")
| eval kind=case(like(log,"%error%"),"error",like(log,"%info%"),"info",like(log,"%warn%"),"warn")
| stats count AS Log_count BY kind
Ciao.
Giuseppe
Thanks It's working fine
Hi @surens,
did you tried:
index = aws_ubs_n log IN ("*error*","*info*","*warn*")
| stats count BY log
For more infos see at https://docs.splunk.com/Documentation/Splunk/9.0.1/SearchReference/Stats
Don't use the search command after the main search, it's a best practice to put search conditions as left as possible to have faster searches.
Ciao.
Giuseppe
I want know each log count
like
error log 23456
info 873267
how ?
Hi @surens ,
ok, please try this:
index = aws_ubs_n log IN ("*error*","*info*","*warn*")
| eval kind=case(like(log,"%error%"),"error",like(log,"%info%"),"info",like(log,"%warn%"),"warn")
| stats count AS Log_count BY kind
Ciao.
Giuseppe
Hi @surens,
in a multiselect input, you could use a search like this:
index = aws_ubs_n log IN ("*error*","*info*","*warn*")
| eval kind=case(like(log,"%error%"),"error",like(log,"%info%"),"info",like(log,"%warn%"),"warn")
| dedup kind
| sort kind
| table kind
and use the kind value both for label and value.
Or use fixed values.
Ciao.
Giuseppe
P.S.: Karma Points are appreciated 😉