(This post was written by Dan Belfer, from the Performance Quality Assurance Team)
VuGen 12.55 arrives with a new exciting feature called the ‘Recording Summary Report’. This feature enables our customers to quickly view of all of the data that was recorded in a session. The information is clear, colorful, and self-explanatory.
One of the main pain points customers complain about is that while recording an application, unwanted information is often included in the script. This undesired information often includes analytics of different engines, ads, third-party information etc. The new “Recording Summary Report” comes to the rescue and enables the user, in the easiest way possible, to filter out any unwanted hosts or domains. Moreover, it gives the user a clear and simple view of almost all information he may need after finishing recording or activating the script code regeneration.
When this feature is enabled, the first change you will notice is the addition of a new item in the Solution Explorer related to the script being developed:
Clicking on it will generate the report with the collected data.
The simple example shown in this blog post is a recording of a simple navigation to the English Wikipedia page using Web HTTP/HTML protocol. You will notice that several unwanted steps were added to the script even though it was not my intention when recording the script. Notice: In LoadRunner 12.55 we support the Recording Summary Report only for Web HTTP/HTML scripts.
Below, you can see the “Start Recording” dialog box for this example:
Let’s click the ‘Recording Options’ button in this dialog box (you can also activate the Recording Options from the menu):
In addition, in ‘Recording Options’, under ‘HTTP Properties’ -> ‘Advanced’, you can filter out headers that need to be eliminated:
The third filter shown here is related to the idiom ‘Resource/Non-Resource’. When VuGen records a browser, all traffic is recorded, and not all resources that are recorded can be used in the same manner. Part of them contain important information like HTML pages or application data (Json, XML), while other resources (like images) do not.
VuGen distinguishes between these types of resources according to Response Content types. Content types that can include information that might be needed for correlations, and should be analyzed in replay, are defined as Non-Resources. The default Non-Resource content types are text/html, text/xml and application/json.
You can see here how it appears in VuGen:
Now that we’ve gone through the explanation of the main three filters and how they can be activated on the generated code, let’s move on to recording.
As said before, I’m recording a page from the English Wikipedia using Internet Explorer as my browser and utilizing the Web HTTP/HTML protocol:
After recording only the basic navigation to http://en.wikipedia.org , we stop the recording – it is as simple as that.
We receive a message that the Recording Summary report is being processed:
And voila: a new, colorful, clear and comprehensive page is displayed:
What do we see?
The top section displays information about the recording itself: the number of bytes sent/received, times, kind of browser used to record, the number of requests and connections. Just above this dashboard we can see a notification on Correlation design studio – it doesn’t pop up as before.
Clicking the ‘Open Design Studio’ button opens the original dialog with the number of correlations as written above. The main area of the report is divided into three sections – Hosts and Domains information in the middle; above it, the Headers section; and on the right – Content Type resources. You can drill down into each of these three sections for details.
Here you can see the “Hosts recorded” details:
In this detail, you can see information on each host like the IP address, SSL version, SSL cipher, version of HTTP, number of requests and amount of data received. You can perform a search on the list, show hosts that were excluded or included (Note – all hosts that are recorded even after filtration will appear! The filtered hosts appear in gray with the option to re-include them in the script as needed). The Include button shows which hosts are included and are being filtered, while the rest of the hosts will be part of the excluded hosts list (click the Excluded button). Hosts are grouped by domain. (Note: Excluding a domain automatically excludes all hosts in that domain.)
Each part in the report has a link in the top right area that enables the user to ‘jump’ to the relevant point in the ‘Recording Options’ section.
The next part of the report is the headers section. I clicked the headers link to open the list of headers that were added to the script.
Here we can only search and display headers that are included—or excluded in the script. There is no option to manipulate the headers displayed in the list. But, you can always click the link at the top right of page to jump to the dialog of headers appearing in ‘Recording Options’ section.
The final part is a detailed list of the ‘Content Types’ resources that were recorded. As explained above, we can see ‘Resources’ that will not affect the script, and ‘Non-Resources’ that appear in the script. An example is the ‘text/html’ (shown below) that was involved with detection of correlation. Detailed information on each resource like content type, count, size, and more is shown:
With regards to the script itself, you can see here that third-party resources were added – although the navigation was only to http://en.wikipedia.org (in this case using the ‘Bing’ search engine from the IE browser). Be aware that recording with different browsers, will add other third-party resources to the results.
We will now filter the Bing/Microsoft resources collected in the recording. As shown below, indicators are added to the excluded domains and a ‘Regenerate’ button appears.
Clicking this button will result in two actions:
- In ‘Recording Options’, under the network section, new string for filtering is added
- The script is regenerated with the new filters
In the result we immediately see the change in the ‘Dashboard’ header of the report – the number of requests and connections is lower than before
And in the script itself, all mentions of ‘Bing’ and ‘Microsoft’ are eliminated:
Finally, when observing the ‘Recording Options’ -> ‘Mapping and Filtering’ section, you will see two new strings, referring to ‘Bing’ and ‘Microsoft’:
This feature is activated by default, and exists in the installation of LoadRunner 12.55 or VuGen Standalone 12.55 and can be turned off/on in the Tools>Options>Scripting>Recording section:
In LoadRunner 12.55, we enable users to receive important information immediately after recording—in a clear and pleasant way. Setting important recording options is a relatively tedious and error-prone process. To overcome this issue, we created the new Recording Summary Report. We assist in filtering resources, hosts and unwanted third-party domains that are added during recording and provide users with increased control on recording information for their scripts.
Are you interested in more information about LoadRunner, Performance Center or StormRunner? Visit the LoadRunner, Performance Center or StormRunner forums to find information, submit questions and collaborate with peers.
To learn more on recording summary report, visit the VuGen Help Center.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.