Network Errors in Response Header

[avin_mehya 0]

We have a testing scenario which consists of few profiles. During execution we receive few network errors in the response header either

 

Network error: 10054 - "An existing connection was forcibly closed by the remote host." in the response header or

 

Network error: 10060 - "A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond."

 

As mentioned in this forum(http://www.loadtestingtool.com/forum/topic/45-failure-socket-error-10054/) we have tried playing around with the think time for all requests still we are getting these errors.(Set to as high as 15 seconds or Specified in Request)

 

We have also checked with our network team to validate the load balancer configuration or drop of packets but that also is fine.

 

Also this issue is not consistently observed, sometimes for 1000 users there wont be any network errors and on few runs there are around 30-40 network errors of 10054 code. We have checked logs on the IIS server and we do not get any failures on that end.

 

We have also changed few other parameters i.e. work queue length, worker processes,connection setting on load balancer etc. but still we keep on getting these errors

 

We have monitored CPU Utilization, Memory, I/O etc but there is nothing alarming even on that front.

 

What could be the issue with the same. Is there a way to debug these errors as they give out very minimal details.

 

Note: We are using WAPT Pro 3.5 with 2 load agents

[sergei 0]

First of all in this case you need to control network bandwidth. These errors may occur because load agents haven't enough network bandwidth or server itself hasn't enough bandwidth to serve 1000 users. Try to use more agents.