I don't succeed finding some interesting bugs with my SoapUI
Hello everybody,
I am new on this forum and I hope I will get at least one answer to an issue becoming quite frustrating.
I have been working as a software tester for 15 years with a main focus on GUI test automation and manual functional tests.
Begin February of this year I started on a new project some web service integration tests with SoapUI (light version).
I am more and more frustrated as I have the impression my tests have no real added value. I do not succeed finding any 'relevant' bugs. Only a few reported bugs are about wrong response content but all the rest is minor or trivial.
Application architecture is based on SOA suite (Oracle) and all coding stuff is done with JAVA (struts 2). Front-End is web based. There are something like 150 web services to test.
When I get a java exception, response is not what I am expecting or I succeed in creating something which shouldn't be allowed (e.g. an item with endDate < beginDate) it's often rejected by technical team because it's not possible to do the same through GUI.
When I report on missing controls it's often rejected by technical team because even if controls are missing at BE, error is catched at Front-End. For some parts of the applications, most of controls are indeed implemented in FE!
When I get something like an SQL exception it's also rejected as GUI does not allow primary key violations for example.
But what can I test then? How can I find some 'real' web service bugs? Is there something else I can try? All my web service 'happy flows' are working. And it's really becoming frustrating having a look at my functional colleagues finding bugs and bugs through GUI and complaining why serious integration tests were not done before!
Thanks a lot in advance for your support.
Kind Regards,
Michael Christian
I am new on this forum and I hope I will get at least one answer to an issue becoming quite frustrating.
I have been working as a software tester for 15 years with a main focus on GUI test automation and manual functional tests.
Begin February of this year I started on a new project some web service integration tests with SoapUI (light version).
I am more and more frustrated as I have the impression my tests have no real added value. I do not succeed finding any 'relevant' bugs. Only a few reported bugs are about wrong response content but all the rest is minor or trivial.
Application architecture is based on SOA suite (Oracle) and all coding stuff is done with JAVA (struts 2). Front-End is web based. There are something like 150 web services to test.
When I get a java exception, response is not what I am expecting or I succeed in creating something which shouldn't be allowed (e.g. an item with endDate < beginDate) it's often rejected by technical team because it's not possible to do the same through GUI.
When I report on missing controls it's often rejected by technical team because even if controls are missing at BE, error is catched at Front-End. For some parts of the applications, most of controls are indeed implemented in FE!
When I get something like an SQL exception it's also rejected as GUI does not allow primary key violations for example.
But what can I test then? How can I find some 'real' web service bugs? Is there something else I can try? All my web service 'happy flows' are working. And it's really becoming frustrating having a look at my functional colleagues finding bugs and bugs through GUI and complaining why serious integration tests were not done before!
Thanks a lot in advance for your support.
Kind Regards,
Michael Christian
