Closed incertum closed 11 months ago
That doesn't look like an innocent break. Looks like process
should be a Popen
object but it's getting a SinspProcessStreamer
in that parameter (the joys of weakly typed languages!).
I'll look into it, but I find it really odd.
After some issues trying to run the tests on Fedora 38 (apparently we are not detecting container ids correctly with podman but I think this may have been a known issue), I gave up, ran the tests on Ubuntu 22.04 and it seems to be working properly.
@incertum, any chance you can send me the full report generated in build/report/
? Either by attaching here and/or via DM? That should have everything needed for me to dig a little deeper.
Discussed a bit with Mauro on slack, the problem appears to be running everything as root if you don't run the tests from a container and just on your box and it's looking for the wrong python and wrong installed packages location even if I run it from a fresh venv. We don't have a solution atm, other than a possible major refactor or building everything in a way so it can be run from a compatible container, e.g. using falco-builder container to build sinsp-example and drivers respectively and not use the build/ folder builds.
Describe the bug
Noticed that I cannot run the
e2e
tests anymore locally while it used to work when it was last refactors to support localhost test runs.How to reproduce it
Try running it locally on current master. Does it still work for everyone?
Additional context
@Molter73 if it still works for you, please let me know what additional information you need from me?