# Need to generate response time graphs from apache logs, average and pct95
**URL:** https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941
**Category:** Kibana
**Created:** [June 21, 2018, 9:33pm UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941 "2018-06-21T21:33:09Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@lserranov](https://discuss.elastic.co/u/lserranov)
#### Post date: [June 21, 2018, 9:33pm UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941/1 "2018-06-21T21:33:09Z")
Hi,
As a user, some time back I was able to generate Response Time graphs (Average and pct95) from apache logs. X axis was time, In Y axis there were 2 lines, pct95 and average. So I would assume I know how to do them.
Now, as a sysadmin, I just setup ELK system and I'm trying to do simmilar graphs. So I chose to create a "Line" Visualization. In there for X-axis I chose Date Histogram, no problem. But in Y axis, I chose: Aggregation = Average, However for Field value, I cannot pickup any field from the Apache logs, they do not show in the list. All I can chose is "offset" which doesn't help.
The question here is, what is the configuration I need to do so that in the Filed choice of the Line visualizations I can see and chose say response\_time coming from apache logs?
Thanks
Luis
---
### Author:  [@lserranov](https://discuss.elastic.co/u/lserranov)
#### Post date: [June 21, 2018, 11:00pm UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941/2 "2018-06-21T23:00:03Z")
I went ahead and I checked my index. Looks like the field that I want looks like this:
Name, Type, Format, Searchable, Agreeable, Excluded
===== ==== ======= ========= ========= =======
server\_time, String, "empty",yes, "empty" ,"empty"
I wonder:
1. If this could be the cause of nothing been available in the "Field" field
2. Now I wonder how Logstash is able to digest the log ok and it is able to populate field but with wrong type, string instead INT as I chose.
Thanks!
---
### Author:  [@lserranov](https://discuss.elastic.co/u/lserranov)
#### Post date: [June 21, 2018, 11:27pm UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941/3 "2018-06-21T23:27:16Z")
In logstash, in grok, made sure to capture server\_time as i read from somewhere:
INT:server\_time:int
I recreated the index and still I see the server\_time field as I reported earlier:
Name, Type, Format, Searchable, Agreeable, Excluded
===== ==== ======= ========= ========= =======
server\_time, String, "empty",yes, "empty" ,"empty"
I wonder what do I need to do in order to force Kibana (or whoever) to take server\_time as integer? I'm assuming that this is the cause of field FIELD in visualizations not showing any value.
Thanks
---
### Author:  [@lserranov](https://discuss.elastic.co/u/lserranov)
#### Post date: [June 22, 2018, 2:55am UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941/4 "2018-06-22T02:55:38Z")
I figured that I needed to reindex the index to have the proper type for the server\_time field.
After doing that, now I see server\_time in the Field component on the Visualize menu.
I do not need extra help.
Thanks
---
### Author:  [@system](https://discuss.elastic.co/u/system)
#### Post date: [July 20, 2018, 2:55am UTC](https://discuss.elastic.co/t/need-to-generate-response-time-graphs-from-apache-logs-average-and-pct95/136941/5 "2018-07-20T02:55:41Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.