Support Forum

Without logs and actual error messages, it's impossible to tell what's going on. That said, permissions error are usually not a bug in sabnzbd itself, as the program simply passes on the message it in turn received from the operating system when trying to write/access files and directories.

Now I am away from home and want to connect me with a Wireguard VPN connection to the web UI. I had modified my local ranges to 192.168.0.0/24, 10.8.0.0/32 before and recreated the Sabnzbd container. The second IP range is for my Wireguard network. For starters, that 10.8.0.0/ 32 seems unlikely in ...

As it says on the categories config page: only relative paths are based on the complete dir. So if you want to store things somewhere else, simply fill out the full path.

This issue turns out to be the result of an issue in a patch applied to the ppa package, that caused the wrong function to be patched since 4.1.0alpha1. Will be fixed in a new upload very soon; please update and confirm the problem is gone.

The IP address not showing up unless you set debug logging is a patch applied to the packaged version on debian and ubuntu; the effect is the same as a privacy-conscious user clearing out the selftest host in config->special. That should not make a difference though, as sab shouldn't need that ip ad...

Did you define any local_ranges in config->special? If so, try clearing out that setting..

Note that sabnzbd's sorting takes its information entirely from job information, i.e. the job name and (if available) metadata embedded in the nzb. It doesn't have access to any other information sources nor any prior knowledge of the content of a download. So if you want the year or the episode nam...

So it's all setup correctly, and the exact same check the init.d script does also passes. Maybe systemd is using a cached version of the default file, in spite of the daemon-reload? Does the issue persist after a reboot?

Do you see my mistake? I genuenly don't

Not yet

What result do you get for this command:

You probably want to use a venv for this to avoid polluting the system's python install. Grab and extract the sabctools source from github, install build and/or test dependencies ("python3 -m pip install -U -r requirements.txt -r tests/requirements.txt"), then run "python3 setup.py bu...

The package ships an init.d, which will also work with systemd's compatibility mode as long as you do not add a native systemd service with the same name. That init.d script is enabled by default, but will only actually start the application after you edit /etc/default/sabnzbdplus to set the user, a...

Trial and error to narrow down the cause. You could try compiling without creating a wheel, and see if the alignment of the .so is sane at that point.

I didn't mean to point out manylinux as the cause of the issue, but rather the breakage in patchelf it was affected by.

Does qnap have a public bug tracker for their ubuntu derivative?

OneCD wrote: ↑June 20th, 2023, 2:54 pmIt looks like the page sizes don't match.

Does this suggest a config issue with 'gcc'?

Indeed, pagesize 64kB and alignment at 4kB won't work.

Something on the toolchain causing this, not necessarily gcc though; could be the linker or something like https://github.com/pypa/manylinux/issues/735

What's the page size and the load alignment on that system? Try these two commands (modify the filename and path of the sabctools .so file so it points to the one you built): $ getconf PAGESIZE 4096 $ readelf -l /usr/lib/python3/dist-packages/sabctools/sabctools.cpython-310-x86_64-linux-gnu.so Elf f...