Could you please advise on Overall page response time from following sample report.
I knew there is open defect when using grouping with Resources. https://github.com/gatling/gatling/issues/2090
Based on following report for “What Housing is eligible for” is there any best approach to project overall page response time.
Currently min time shown as 14 though there is one resource with min load time 109 ms & 95pct , 99pct values are 81, 197. not sure if these are correct
Kindly advise if any other better approach.
Thanks in advance.
Total OK KO %KO min max Mean Std Dev 95th pct 99th pct Req/S
One of the Page response time shown below has 40ms as min load time, where as one of its resource has load time 161 ms.
Not sure if this need minor improvement.
Regarding 1, that’s expected, as “register-a-croft” is a group, and Global information only accounts for requests.
Regarding 2, it looks like a bug. I’d probably need a reproducer.
approximation of “page load time” will only be seen on the details page “Cumulative response time” chart, nowhere else.
group “response time” measures individual requests within the group and will not provide a measurement of the “page load time” in this context.
Regarding 2 (group “response time” measures shown in table), are you saying this is not a bug ? Please could you advise.
“page load time” shown on the details page “Cumulative response time” chart, displays 95%, 99%, min, max values at each and every sec/interval. is there any approach to find page load time following overall test run for eg at end of 30 mins load test execution, finding overall page load time with min, max, 95%, 99% values.
I have tested Stéphane ’ s fix in a controlled environment and it looks good enough for what I would reasonably expect in the scope of 2.0
Ie. Although all of the resources should be sent, they will be completed as fast as the network will allow so the various real browser delays like blocking javascript and css and paint pipeline are not captured (if you want to the DSL is there to hand craft approxinations of these delays) resulting in a lower response time.
In terms of possible defect below
You are querying whether the min page response time can be lower than any of its resources. A reason check on the correctness of the data.
In the table below there are some failed page requests. Unfortunately not exactly but roughly the same number of request 414 's exist as the failed page requests. it is possible that the zero min page time happened in a different group measurement than the non zero 414 request. Ie. No bug if that is the case.
Looking forward to see updated documentation for Gatling reports especially Group Cumulated Response Time Percentiles over Time.
Currently this report showing Cumulated Response time over time (Page Load Time) , Active Users, along with with min, max, 95, 99pct values over time.
Would you please brief on 95pct or 99pct in this context, what does this indicate. is it about 95 pct resources response time, with X number of active users at selected time.