Miloš Prchlík

> @psss @happz I'll try to be more specific, likely I am missing something obvious > > Including the following code at [internal.py](https://github.com/teemtee/tmt/blob/main/tmt/steps/execute/internal.py) creates **once** the desired path and gives...

@spetrosi can you try with the following patch applied? ```diff diff --git a/tmt/log.py b/tmt/log.py index 66db71c5..18d848b9 100644 --- a/tmt/log.py +++ b/tmt/log.py @@ -287,7 +287,7 @@ class LogfileHandler(logging.FileHandler): # with older...

Is there anything special, or uncommon about your terminal settings? Because I'm unable to reproduce the problem, and since it looks like at some point, `\n` stops being enough for...

Never mind, I just reproduced the problem. FTR, it works perfectly fine with `container`, but with `virtual`, I observe the reported behavior.

So, running `stty sane` *after* the test command quits makes things right again. I suspected there must be something breaking the terminal, messing with its settings, like echo and so...

3 trivial Beakerlib tests + `stty sane` - yes, eventually, when tests are done, you get `tmt try` prompt back, and you can control tmt and input values. But the...

> Many thanks for finding the cause of this. `stty sane` - very smart, I didn't know that. > > We are running tests in testing farm via https://github.com/marketplace/actions/schedule-tests-on-testing-farm, e.g....

> I plan to create simple XSD schemas for each JUnit flavor supported by TMT and add tests validating the generated JUnit files against these schemas. > > Additionally, it...

Added `blocked` label: depends on a branch that's been pushed out to 1.36, the rebase would be a mess, and it's not critical.

/packit build