I think this might be of interest to some of the other developers.
Soemtimes we have a problem that the test suite itself is not 100% stable, and it's not easy to reproduce a test that fails sporadically every Nth time.
As the "execute" statement in the "control" section actually returns a verdict type, you can use the following construct in a control section:
control { for (var integer i := 0; i < 10000; i := i+1) { var verdicttype v; v := execute( TC_paging_ps_ptp_lac() ); if (v != pass) { break; } } }
This way you keep running the test suite until the given test fails for the first time. Then go do something else for some hour(s) and by the time you get back, hopefully you will have reproduced the problem.
Regards, Harald
Harald/Osmocom team gave TTCN3 real meaning and public attention. Before you came up, TTCN3 was a prisoner of worthless, dry academic papers.
On Sat, 13 Feb 2021, 18:08 Harald Welte, laforge@osmocom.org wrote:
I think this might be of interest to some of the other developers.
Soemtimes we have a problem that the test suite itself is not 100% stable, and it's not easy to reproduce a test that fails sporadically every Nth time.
As the "execute" statement in the "control" section actually returns a verdict type, you can use the following construct in a control section:
control { for (var integer i := 0; i < 10000; i := i+1) { var verdicttype v; v := execute( TC_paging_ps_ptp_lac() ); if (v != pass) { break; } } }
This way you keep running the test suite until the given test fails for the first time. Then go do something else for some hour(s) and by the time you get back, hopefully you will have reproduced the problem.
Regards, Harald --
- Harald Welte laforge@osmocom.org
============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)
On Sat, Feb 13, 2021 at 07:27:34PM +0100, Mirko Kovacevic wrote:
Harald/Osmocom team gave TTCN3 real meaning and public attention. Before you came up, TTCN3 was a prisoner of worthless, dry academic papers.
thanks for the feedback. I think the real difference only is that we develop our TTCN-3 code as open source. There were and are quite a number of users of TTCN-3 before outside of academia - but for some strange reason none of them (I know of) seems to have been doing open source development.
Regards, Harald