Hi All,
I'm trying to graph the usage of bandwidth fetched by Apache logs but
Kibana histogram doesn't graph any data. Could someone help me to use
correctly the histogram panel?
If you are using COMMONAPACHELOG or COMBINEDAPACHELOG in your logstash
config and
you've not changed default type for "bytes" it was probably imported into
elasticsearch as string instead of int
Elasticsearch cannot aggregate this data in such form.
You'd can either modify "grok-patterns" file for your variable so bytes
look like below or use mutate function in logstash to convert it to int
(?:%{NUMBER:bytes:int}|-)
regards
m
On Tuesday, 3 February 2015 16:31:12 UTC, Marcello A wrote:
Hi All,
I'm trying to graph the usage of bandwidth fetched by Apache logs but
Kibana histogram doesn't graph any data. Could someone help me to use
correctly the histogram panel?
I prefer to change the configuration file because the file patternss
grok-pattern can change with the RPM updates.
I added this part
mutate {
convert => [ "bytes", "integer" ]
}
}
Marcello
Il giorno martedì 3 febbraio 2015 18:54:02 UTC+1, Marcin izo ha scritto:
If you are using COMMONAPACHELOG or COMBINEDAPACHELOG in your logstash
config and
you've not changed default type for "bytes" it was probably imported into
elasticsearch as string instead of int
Elasticsearch cannot aggregate this data in such form.
You'd can either modify "grok-patterns" file for your variable so bytes
look like below or use mutate function in logstash to convert it to int
(?:%{NUMBER:bytes:int}|-)
regards
m
On Tuesday, 3 February 2015 16:31:12 UTC, Marcello A wrote:
Hi All,
I'm trying to graph the usage of bandwidth fetched by Apache logs but
Kibana histogram doesn't graph any data. Could someone help me to use
correctly the histogram panel?
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.