'Ghost' or cached NZB's loading and causing duplicate fails
Posted: Sat Mar 23, 2019 3:01 pm
hi there, wondering if anyone has encountered this issue. Periodically newsbin will combine a current NZB with previous NZB's, like it's cached the file name and combines past + current NZB content. from the NZB file as a for example:
subject="28c9fbf312340d8154b38164b465cb0a [1/27] "f54ff7fca7a644a1bdbe8be601f3a020.par2"
This is theoretically a 300mb file yet newsbin shows it as 9gb. looking at the expanded file list, there are 5 completely different files in the list (and each one has a part01, part02, part03, etc). I've tried using the filter to only show the specific file I'm looking for but filters appear to work at a file level and not part level, so the list always ends up empty. Newsbin happily tries to download the first part01, realizes it's been downloaded and it's a duplicate so it stops. Even if that's not the file I just loaded - it's something it downloaded previously and doesn't exist in the NZB file I just selected. I've tried clearing out the various newsbin directories (not the db's however) and tried loading the NZB file 3 different ways (double-click, drag & drop, file -> load nzb). No change. To confirm my theory, I manually edited the NZB file and changed all occurrences of:
28c9fbf312340d8154b38164b465cb0a to
18c9fbf312340d8154b38164b465cb0a
The file then loaded as expected, a 300mb download. There must be some kind of cache it's using and it's getting confused with the obfuscated filenames, but I'm not sure how to clear it or override it... any suggestions?
subject="28c9fbf312340d8154b38164b465cb0a [1/27] "f54ff7fca7a644a1bdbe8be601f3a020.par2"
This is theoretically a 300mb file yet newsbin shows it as 9gb. looking at the expanded file list, there are 5 completely different files in the list (and each one has a part01, part02, part03, etc). I've tried using the filter to only show the specific file I'm looking for but filters appear to work at a file level and not part level, so the list always ends up empty. Newsbin happily tries to download the first part01, realizes it's been downloaded and it's a duplicate so it stops. Even if that's not the file I just loaded - it's something it downloaded previously and doesn't exist in the NZB file I just selected. I've tried clearing out the various newsbin directories (not the db's however) and tried loading the NZB file 3 different ways (double-click, drag & drop, file -> load nzb). No change. To confirm my theory, I manually edited the NZB file and changed all occurrences of:
28c9fbf312340d8154b38164b465cb0a to
18c9fbf312340d8154b38164b465cb0a
The file then loaded as expected, a 300mb download. There must be some kind of cache it's using and it's getting confused with the obfuscated filenames, but I'm not sure how to clear it or override it... any suggestions?