I have been working on an application where I have two workload types. One being incremental and the other a queuing workload. My problem is trying to justify why when the incremental workload is implemented, I get all sorts of errors ranging from http 404 to Winsock 10060. The logout time for a user is 30mins but i have got around this by looging out each of the users through scripting (so the timeout is negated). Anyone have any ideas as to why i'm getting such a difference in results? Any help very much appreciated
The problem could be that the incrementing workload is stressing out the app or the site. I would try testing the app with varying number of users to see if the problem is load related. IE: Start with 25 users and then try 50 users etc...
This may help narrow the field and could possibly show the bottleneck.
The queuing workload may not show the problem as each Vuser will complete its prescribed tasks and then finish the test, so there may not be enough load placed on the site to show the problems reported via the increasing workload
Dheeboy, thanks for your reply, however I discovered yesterday that the application database is not Oracle (as i was led to believe) but a clustered SQL Server box. I can get the app stable up to 95 VUsers but it then begins to crap out after that. Could the SQL Server clusters be causing the problems i am experiencing?
Thanks in advance. [img]images/icons/confused.gif[/img]
What types of errors are you getting, are they SQL native errors?
What I know about clustering, is that this process automatically allows one physical server to take over the tasks and responsibilities of another physical server that has failed. The goal behind this is to ensure that users running mission-critical applications will have little or no downtime when such a failure occurs.
So things to check out are
- Is the SQL code applicable to the other servers etc..
The winsock errors could be because the host can not be found, same applies to the 404 errors which mean the file/image etc,, can not be located
The errors i'm getting are Winsock 10060, HTTP 500, 404 and also WebEngine43 errors.... I was up till pretty late last night trying to get this thing bottomed out...... Here is what i've found...
I have managed to clear up the Winsock 10060, WebEngine43 and HTTP 500 errors, amongst which there were jave redirect call errors and also 'html not found' errors. After having a pow wow with the networks guys it appears that the server logs were basically reporting similar indicators. The load balancer which the connects go through seems to spread the login details and request to different servers sequentially???!! So to cut a long story short.... I've failed it and passed it all back to the network and dev guys.....