To assure compliance and full support, companies upgrade their Citrix/SAP instances on a regular basis.
More often than not, users are complaining about the Citrix connections being slow compared to the previous SAP-release.
Typically, the SAP teams of these companies start with improvements utilizing SAP Solution Manager to figure out the “where” and “what” inside the different SAP-modules.
After some time, when users are still complaining, things tend to escalate. This is the time where companies start looking for ways to troubleshoot connections end-to-end in an attempt answering questions like “Is it Citrix or SAP?” and “Suppose it is Citrix, what part of Citrix and what to do about it (i.e. what is the low hanging fruit)?”.
This case study gives you a detailed overview about our approach and the potential findings when dealing with these kind of questions.
The initial findings of these troublesome TCP sessions indicate suspicious network behavior in the zones between the Citrix client and the Netscaler as well as the XenApp server and the SAP front-end.
When looking at the variations of the performance from these zones, the additional delay on the user response times varies between 500 and 4000 ms; occasionally even more as outliers tend to be 2-3 times their average values.
This made the performance of this Citrix/SAP environment erratically and unpredictable; hence user complaints come and go.
Full case study
This case study, which is only available as a PDF, describes the setup (part 1), the findings (part 2) and the discovery of the quick wins (part 3) of one of those Health-checks; including how this answers these questions.