Is it the application?
We have looked at Is it the network? to assess network round trip and latency issues and Is it the server? to assess server response time issues. We determined that neither the network nor server were the culprit.
The Application Performance dashboard helps you determine if application transaction response times and application error conditions and the impact they can have on application server response times are the source of the problem affecting users.
Select the Application Performance dashboard. To get the dashboard, see Importing a user-contributed dashboard.
This dashboard provides detailed application transaction information such as transaction types and errors, as well as detailed response times. This information helps you determine how the application is performing by transaction and whether it is the application that is impacting users. To see application level information in Apex, you must have application transaction analysis enabled for the desired servers on the GigaStor or probe. Observer and Apex support application transaction analysis for many different enterprise applications such as DNS, SQL, HTTP, Citrix, LDAP, CIFS/SMB, SIP, and more.
Below are two tables from the Application Performance dashboard that show transaction responses and requests for an HTTP server of interest. You can see the request and response codes and description, plus the corresponding average and maximum response time for each transaction type. In Figure 57, if you look at the Request table (lower table), the most common request transaction is a GET, which is typical for HTTP. If you look at the GET request line the average response time is low, 120 ms, but the maximum is almost 10 seconds (9996 milliseconds). Clearly some user transactions are experiencing long response time delays.
 
Figure 57: Application Analysis Server-Response
 
Let’s now look at the top table for transaction responses in Figure 57. We have sorted the response codes to show various HTTP transaction errors that show up in the 400 response code range. You will notice several transaction problems. We had 12,651k 0x02 = Name Sever Internal Error that took almost 7 seconds max to process. This issue is causing a slow response time. By looking at the Application Performance dashboard, we can determine if we have application errors causing user problems as well as the response time metrics associated with various transaction types.
In this example, we identified a major application error problem from our Application Performance dashboard. We were getting many 404 File Not Found errors. We can use the ability to launch a drilldown connection from a dashboard to the packet data in the probe (this passes a context-based filter to the GigaStor), or we can just go to our probe and analyze the packet data for a particular server to find the root cause of any application error.