1. What version of LoadRunner (LR) or PerformanceCenter (PC) are you using? (specify which LR or PC)
2. What is the protocol you are recording?
3. If HTML - are you using HTML-Advanced with URLs?
4. URL mode with Concurrent groups, or
5. URL mode without Concurrent groups?
6. Which LoadRunner/PerformanceCenter feature or service packs are you using?
7. What kind of LoadRunner license do you have - Subscription, Perpetual, Virtual User days, or Evaluation? Subscription, full license
8. If subscription, is subscription current or expired?
9. What platform(s) and Operating Systems? Include version and service packs.
WinXP SP2, IE7
10. If you have filed a service request with HP/Mercury, what have they told you at this point with respect to your issue?
I am currently load testing a website that was previously load tested. They got another server and wanted to make sure that the configuration was correct on this server. The previous servers I tested before are fine but this new one seems to be an issue and I was hoping someone could shed a bit of light on what the issue might be. The website is simple and is just a Web Server. The connections to the server are fine and correlate fine with the users. The CPU, memory, HD and NIC cards are fine on the server as well. This server is not Load Balanced either. I also used the ip to double check that it wasn't some DNS issue here. I also checked the LG to ensure that it wasn't a network issue on its side.
From the Analysis I did the problem seems to be on the server side. The TRS times are very high and when I look at the Page Download Time Breakdown graph the First buffer Time is where all the delays are happening. When I look at the Time to First Buffer Breakdown graph all the time is on the server side. The other things I noticed is that the throughput and hits per second cap out with minimal users(45)and that is when the TRS times go up. The Web Development team has tried to mimic the configuration of the other servers. Caching, logging and other things have already been looked at.
I have googled a bit about First Buffer Time but do not have much knowledge on this area and couldn't find a good answer. Any ideas on what could be the problem on this IIS server? I know that each user is probably just waiting for the previous user to complete its First Buffer action and that is why the TRS times go up and throughput maxes out early. Any help would be appreciated.