Badfish fails with ModuleNotFoundError: No module named 'src'

[jmencak]

Your System Details
Fedora 43

• Python Version:
  3.14.2

• Operating System:
  Linux

Describe the bug

$ python-3.14 -m venv venv
$ . bin/activate
(vend) $ pip install badfish

(venv) $ badfish
Traceback (most recent call last):
  File "/tmp/d/venv/bin/badfish", line 5, in <module>
    from badfish.main import main
  File "/tmp/d/venv/lib64/python3.14/site-packages/badfish/main.py", line 16, in <module>
    from src.badfish.helpers import get_now
ModuleNotFoundError: No module named 'src'

(venv) $ pip list|grep badfi
badfish          1.0.6

Changing "from src." to "from" in main.py and http_client.py fixes the issue, but gives me:

(venv) $ ./servers.sh
Traceback (most recent call last):
  File "/tmp/d/venv/bin/badfish", line 8, in <module>
    sys.exit(main())
             ~~~~^^
  File "/tmp/d/venv/lib64/python3.14/site-packages/badfish/main.py", line 2677, in main
    loop = asyncio.get_event_loop()
  File "/usr/lib64/python3.14/asyncio/events.py", line 715, in get_event_loop
    raise RuntimeError('There is no current event loop in thread %r.'
                       % threading.current_thread().name)
RuntimeError: There is no current event loop in thread 'MainThread'.

Similar issues seen on Fedora 43 when using the copr badfish package or self-packaged rpm.

Currently, what works for me on Fedora 43 is badfish 1.0.5 with python 3.9 and "from src." removals as shown above.

[sadsfae]

@jmencak sorry for late reply, this got shuffled while we were on a relative freeze for the main branch as we're working through Fedora packaging submission. This was fixed some time ago in the development branch but if you have an older version that works, cool. We will be merging all dev changes up soon, with some luck you'll be able to dnf install badfish too in a few weeks.

[stephane-chazelas]

@jmencak sorry for late reply, this got shuffled while we were on a relative freeze for the main branch as we're working through Fedora packaging submission. This was fixed some time ago in the development branch but if you have an older version that works, cool. We will be merging all dev changes up soon, with some luck you'll be able to dnf install badfish too in a few weeks.

For the record, git diff v1.0.6 development as of now (35094eb) shows no code difference. Is the "development" happening elsewhere?

[jmencak]

@jmencak sorry for late reply, this got shuffled while we were on a relative freeze for the main branch as we're working through Fedora packaging submission. This was fixed some time ago in the development branch but if you have an older version that works, cool. We will be merging all dev changes up soon, with some luck you'll be able to dnf install badfish too in a few weeks.

For the record, git diff v1.0.6 development as of now (35094eb) shows no code difference. Is the "development" happening elsewhere?

$ git diff v1.0.6..origin/development |wc -l
939

[grafuls]

@jmencak sorry for late reply, this got shuffled while we were on a relative freeze for the main branch as we're working through Fedora packaging submission. This was fixed some time ago in the development branch but if you have an older version that works, cool. We will be merging all dev changes up soon, with some luck you'll be able to dnf install badfish too in a few weeks.

For the record, git diff v1.0.6 development as of now (35094eb) shows no code difference. Is the "development" happening elsewhere?

you might be missing to pull the latest code changes from development

$ git pull origin development

[grafuls]

The failure was likely caused by the badfish console script having a shebang pointing to a different Python interpreter (outside the venv). As a result, running badfish used a Python environment where the badfish package was not installed, causing ModuleNotFoundError.

Try removing the stale script and reinstall the package inside the venv so the script is regenerated with the correct shebang.
Make sure you use python -m :

pip uninstall badfish
rm ~/.local/bin/badfish
python -m pip install -e .

After reinstalling, the badfish script points to the venv’s Python and works correctly.

[stephane-chazelas]

@jmencak sorry for late reply, this got shuffled while we were on a relative freeze for the main branch as we're working through Fedora packaging submission. This was fixed some time ago in the development branch but if you have an older version that works, cool. We will be merging all dev changes up soon, with some luck you'll be able to dnf install badfish too in a few weeks.

For the record, git diff v1.0.6 development as of now (35094eb) shows no code difference. Is the "development" happening elsewhere?

$ git diff v1.0.6..origin/development |wc -l
939

I see a few new commits now that have been pushed (and committed) after I posted my comment (at which point 35094eb was the HEAD of origin/development, and there was no code change):

$ git diff v1.0.6 35094eb9d68003b43f9f7c0eb7560ef9ce9c6138 | wc -l
68

None of which is on python code.

The reason I asked is that I have some other issue (non-related, though I'm also affected by this one) and wanted to make sure it could be reproduced with the latest development version before raising it and the "This was fixed some time ago in the development branch" made me doubt it was the case since I could reproduce this on development head and there was hardly any difference between that and 1.0.6.

AFAICT this issue is fixed (on development) by:

commit 898b50c8e933a35fa68fa1eeb363a6e5c0abece4
Author:     Gonzalo Rafuls <grafuls@gmail.com>
AuthorDate: Thu Jan 29 11:51:31 2026 +0100
Commit:     Gonzalo Rafuls <grafuls@gmail.com>
CommitDate: Thu Jan 29 12:01:58 2026 +0100

    rpm: removed async-lru code and added dependency, fixed imports

Which is included in v1.0.7.

[sadsfae]

Hey @stephane-chazelas and @jmencak are you sorted now?

The reason I asked is that I have some other issue (non-related, though I'm also affected by this one) and wanted to make sure it could be reproduced with the latest development version before raising it

Please feel free to open a new issue even if you're not sure it's a bug, we'll try our best replicate it. We are in a weird holding pattern somewhat as we are moving from COPR to a proper Fedora package to be included in the distribution on the RPM side, this has slowed down quay and container updates.

[jmencak]

I self-built badfish v1.0.7 17e96f on Fedora 43. Now I'm getting

Traceback (most recent call last):
  File "/usr/bin/badfish", line 8, in <module>
    sys.exit(main())
             ~~~~^^
  File "/usr/lib/python3.14/site-packages/badfish/main.py", line 2677, in main
    loop = asyncio.get_event_loop()
  File "/usr/lib64/python3.14/asyncio/events.py", line 715, in get_event_loop
    raise RuntimeError('There is no current event loop in thread %r.'
                       % threading.current_thread().name)
RuntimeError: There is no current event loop in thread 'MainThread'.

when running

badfish -H <host> -u <user> -p <pass> --power-state

@jmencak are you sorted now?

I'm sorted in the sense I'm using a self-patched old (1.0.6) badfish version installed via pip. Using the self-built v1.0.7 doesn't work for me, at least not out-of-the-box.

[sadsfae]

I self-built badfish v1.0.7 17e96f on Fedora 43. Now I'm getting

Traceback (most recent call last):
  File "/usr/bin/badfish", line 8, in <module>
    sys.exit(main())
             ~~~~^^
  File "/usr/lib/python3.14/site-packages/badfish/main.py", line 2677, in main
    loop = asyncio.get_event_loop()
  File "/usr/lib64/python3.14/asyncio/events.py", line 715, in get_event_loop
    raise RuntimeError('There is no current event loop in thread %r.'
                       % threading.current_thread().name)
RuntimeError: There is no current event loop in thread 'MainThread'.

when running

badfish -H <host> -u <user> -p <pass> --power-state

Fedora 43 uses Python3.14 and it's likely the event loop behavior has changed, it used to be that it was created for you if one didn't exist but now I think this is no longer the case and will traceback if called with no loop running.

I will propose some additional logic to development branch to fix this if it's the issue.

@jmencak are you sorted now?

I'm sorted in the sense I'm using a self-patched old (1.0.6) badfish version installed via pip. Using the self-built v1.0.7 doesn't work for me, at least not out-of-the-box.

[sadsfae]

@jmencak can you please try building against the latest development branch let's see if this fixes the error for you.

[jmencak]

@jmencak can you please try building against the latest development branch let's see if this fixes the error for you.

Thank you, #495 fixed the issue for me on F43.