Randomly failing tests about shoreman HOT 6 CLOSED

If it is random, it could be something with the trap system.
You know because of processes finishing before others.

I am thinking this because I had to do a script which had to run many processes recently and because it is pretty much what Shoreman does, I looked at your code a lot.
But I remember having to fix problems regarding processes not always finishing when I wanted.
I will have a look at what I have done and see if something is different.

from shoreman.

Here is the piece of code I was talking about:

onexit() {
  quit-spotify
  kill $pid1 &>/dev/null
  kill $pid2 &>/dev/null
  kill $pid3 &>/dev/null
  sleep 1
}
trap onexit SIGINT SIGTERM EXIT
wait

I really don't know if the problem could come from this but I guess it is worth trying.
There are no big differences with your version but like I said, I remember having problems, reading forums, until this version worked well.
I think I had problems with the thing not behaving the same depending on when I call the parent normally or as a background job and stopping it with a "kill %1".

What do you think?

from shoreman.

This seems to have been fixed since #7 was closed, hooray!

from shoreman.

It still failing :( . I don't know actual reason for this particular ticket(I could open new one), but that's because of output checking, every test has: ... shoreman .. | head -n1 so first line is checked for desired output. But the case that there no guarantee that test program which produces output (echo ...) will produce output before shoreman's output line 00:04:28 example.1 | 'echo "Hello"' started with pid 14163. @hecticjeff what about using smth. like echo "$output" | grep -q "Hello" in tests to match output? I could pull request such changes.

from shoreman.

@kucaahbe Hmm that's very weird. Using grep in the tests sounds like a reasonable approach to me, go for it!

from shoreman.

opened pull request #13

from shoreman.