Am 2018-09-09 um 15:37 schrieb Hervé BOUTEMY:
looking at CI during vote looks like a good idea because the current CI builds
a lot of things, but I fear we don't use the results as much as we should
personally, every time I get a failing build, I find it very difficult to
understand what is failing: the branch display is not useful since you can't
know which step is failing [1], neither is Test Result [2] since it does not
display which step failed
And the output log, with mixed logs is awful [3]: barely unusable
And the stage logs, if I can understand which stage is failing, is so tiny in
the center of my wide screen that I can't imagine I will use it
Our Jenkinsfile configuration is great for mass managment, but is not IMHO
usable for investigating on issues
Is it just me not knowing something on the way to use the current display?
I do agree with you. I was having a hard time to understand what the
real cause is/was with our tests. The Jenkins output is suboptimal.
In this case, everything was rooted in bugs in SCM actually, Jenkins
simply revealed them, but they were hard to understand and to reproduce.
Moreover, testing on Jekins is ridiculous if you see something like this
'git' is not recognized as an internal or external command,
operable program or batch file.
So the slave does not have Git installed. This all real world tests useless.
The tests revealed issues SCM-905, -906, -907, -908, -909, -910, and
-911. But finding and fixing all those took me at least half a day.
Michael
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
