Shouldn't commands_post run after a SIGINT?

[ssbarnea]

If a SIGINT (Ctrl-C) happens during test run, commands_post is not run, so basically we have no chance of doing any cleaning.

IMHO, I think that commands_post should run instead of just existing quickly.

[gaborbernat]

If a SIGINT (Ctrl-C) happens during test run, commands_post is not run, so basically we have no chance of doing any cleaning.

This is by design, and I don't see how we can change this. We can add a flag that would enable users to opt-in for a command post to always run.

[ssbarnea]

I can understand that changing this would be really problematic, but a general opt-in to make it always run would seem ok to me. Basically what I described like if: always() from github actions (hopefully we would not endup with the same unfortunate naming).

[carlcsaposs-canonical]

Ran into this as well—it would be useful to have a commands_post or something similar that always runs if commands_pre run

Or maybe something like first Ctrl-C interrupts commands and second Ctrl-C interrupts commands_post—but I imagine that might break uses of the current behavior

[gaborbernat]

Converted to issue #3858