Answered question

Network-01 Error Spikes in Test

When running a test we are experiencing Network-01 error spikes at certain intervals in the test. This is happening whether we are running 15 users for a short period of time or 35 users for an hour. We are using 1 agent and the localhost to handle the load, so I don't think this is load related.

Aria C.
Aria C.

Aria C.

Level
0
34 / 100
points
Brandon S.
Brandon S.

Brandon S.

Level
0
52 / 100
points

After speaking with the team who services the application in question, it was determined that the high number of logins that we were performing in such a short time was recognized as an attack pattern, and would then give us errors. After reorganizing our test to prevent this from happening, the Network 01 errors then went away.

Did you find this useful ?

No (0)

Yes (0)

0% of users found this answer useful

Other answers

Nouredine A.
Nouredine A.

Nouredine A.

Level
4
5000 / 5000
points
Team

Do you get those errors on the two load generators at the same time when it occurs?

What are the details of this NL-Network-01 errors? Do you have any antivirus installed on your load generator machines?

Also you should check if there's no firewall between your LG's and your application. Sometimes some rules on firewall can block requests based on the request/s received from the same IP.

Brandon S.
Brandon S.

Brandon S.

Level
0
52 / 100
points

Hello Nouredine,

These errors are currently occurring on both machines, and they appear to be java end of file exceptions. Our typical solution to this is to increase the number of parallel connections for the test, but this unfortunately doesn't seem to solve these errors. We 've reached out to the team who services this application to ask if it could potentially be the result of having too many requests of the same IP, and they said they are unsure as of now, but will continue to look into it for us.

Another interesting note about these errors is that they occur roughly at the same time of the test (12.5 minutes in), regardless of the load that we have. Furthermore, the test is completing the same task over and over again, so it isn't new data or steps being introduced that causes them.

I hope this helps clarify any questions.