Support Forum

This happens when you are tryng to connect to a server, right? If so then it's probably because Python 3.10 is more strict about what it will accept than earlier Python versions. Try the SABnzbd 3.5 beta if you can. It has some changes to make it work with more servers.

I think this regex will do what you want:

ubuntu@ubuntu:~$ sudo hdparm -Tt /dev/sda /dev/sda: Timing cached reads: 1658 MB in 2.00 seconds = 829.74 MB/sec Timing buffered disk reads: 172 MB in 3.35 seconds = 51.27 MB/sec hdparm reads the disk sequentially outside the file system and it is read speed, not write speed. The result can be quit...

Have you checked that the nzbs aren't actually empty?

Is there an error message? What does it say when you try Test server on the server config page?

Which usenet server are you using (specific address and port)? It could be a Supernews problem. Try getting the free 25 GB from xsusenet.com and add it as a secondary server. I would also run a memcheck program, if there is such a thing for Mac.

Add some debug output to the scripts. Print the directory the scripts are using as base and each command they are executing with the variables expanded. I don't know PS so I can't help you with the exact commands.

I have added printing of the error in a pull request on Github. You can download binaries for Windows here: https://github.com/sabnzbd/sabnzbd/actions/runs/1512830592 (Requires Github account). I would appreciate if you can test it and let us know what it says. There should be an info line which say...

My current work-around is keeping the notification page open at all times with ALWAYS selected, since closing the page seems to be problematic. If it's working then it's because you saved it correctly and didn't edit afterwards, not because you keep the config page open. It's the config that is sav...

I fixed it here: https://github.com/sabnzbd/sabnzbd/pull/1981

It should be saved correctly, though, so if you save it as always and don't save again on that page then you should get notifications. If not then I think something else is wrong.

Do you think that additional info will provide things like "... because in use by Proces XYZ" (virusscanner use case), or "no rights to write file", or "disk not available" ... ? Because that would indeed be useful. Errno 13 can be caused by the high level error codes ...

A possibly more fine grained error code can be read using status = hex(ctypes.windll.ntdll.RtlGetLastNtStatus() + 2**32) (requires import ctypes). The error codes are listed here: https://docs.microsoft.com/en-us/openspecs/windows_protocols/ms-erref/596a1078-e883-4972-9bbc-49e60bebca55 Maybe this sh...

It's possible that his LTSC version handles this differently than the current W10.

Come to think of it, redownloading the articles is probably the intended behaviour because SAB doesn't know which particular articles are missing from a file. If a file is not complete then it has to download all of it when you retry.

Sorry, I can't see anything in that log file that looks particularly wrong. I'm not sure if it contains the part where you reset the nzb. The orange state means that a lot of articles are missing. It should give up eventually, though.