Can't delete failed jobs or re-add orphaned jobs

[jlsturgeon]

Version: 0.7.18
OS: Synology DSM 5.0 (Linux)
Install-type: SynoCommunity repository
Skin (if applicable): Plush
Firewall Software: None
Are you using IPV6? no
Is the issue reproducible? yes

The problems began yesterday with three downloads that were showing as 100% but would never finish and actually didn't have anything in the "incomplete" folder when I checked. Today I was also unable to purge failed jobs from the history; when I click the button "Purge Failed NZBs and delete files" nothing happens. When I check the "incomplete" folder, the folders for those failed jobs are still present and they are still listed in history. I tried restarting SAB, which resulted in the jobs that had stalled at 100% disappearing from the queue and being listed under "orphaned jobs" on the status pages. I still could not purge failed jobs from the history. I tried to add the orphaned jobs but when I click the "add" button, nothing happens apart from the page refreshing sometimes. The jobs are still listed as orphaned and they are not returned to the queue. I decided at last to try rebuilding the queue. When I did that, something like 600-700 new jobs appeared that apparently SickBeard had snatched at some point but had not been added to the queue. However, an additional 20+ jobs disappeared from the queue and are now listed under "orphaned jobs." I still cannot purge failed jobs or add these orphaned jobs to the queue again. I did have SAB set to ignore duplicate nzbs, so thinking this might be the issue, I set "detect duplicate downloads" to OFF, restarted SAB and tried again. No joy. I then suspected a file permissions problem so I double checked that the entire "Downloads" directory and all its subdirectories (i.e. "complete," "incomplete" etc.) were set to give enough access. They are set so that owner, group and others all have read, write, and execute permissions. I verified that this is the case for one of the jobs that would not purge from history. If I use the "delete" button next to the individual failed job and pick, "purge failed nzb and delete files," it deletes the job from the history list but does not in fact delete the files.

I've not been receiving error messages. When I try to re-add an orphaned job, the log file says that it was added (/volume1/DiskStation/downloads/incomplete/The.Godfather.Part.II.1974.720p.BluRay.x264-SiNNERS.cp(tt0071562).1/__ADMIN__/__verified__ missing
2014-08-12 14:08:17,821::DEBUG::[nzbqueue:167] Repair job The.Godfather.Part.II.1974.720p.BluRay.x264-SiNNERS.cp(tt0071562).1 by reparsing stored NZB
2014-08-12 14:08:17,822::INFO::[__init__:626] Adding The.Godfather.Part.II.1974.720p.BluRay.x264-SiNNERS.cp(tt0071562).nzb.gz) but it isn't in the queue.

I think this is what the log shows when I try to purge failed jobs:
2014-08-12 14:08:36,741::DEBUG::[_cplogging:55] [12/Aug/2014:14:08:36] HTTP Traceback (most recent call last):
File "/volume1/@appstore/sabnzbd/share/SABnzbd/cherrypy/_cprequest.py", line 618, in respond
cherrypy.response.body = self.handler()
File "/volume1/@appstore/sabnzbd/share/SABnzbd/cherrypy/_cpdispatch.py", line 25, in __call__
return self.callable(*self.args, **self.kwargs)
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/interface.py", line 413, in tapi
return api_handler(kwargs)
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/api.py", line 101, in api_handler
response = _api_table.get(mode, _api_undefined)(name, output, kwargs)
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/api.py", line 457, in _api_history
del_job_files(history_db.get_failed_paths(search))
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/api.py", line 1538, in del_job_files
remove_all(path, recursive=True)
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/misc.py", line 1288, in remove_all
files = globber(path, pattern)
File "/volume1/@appstore/sabnzbd/share/SABnzbd/sabnzbd/misc.py", line 78, in globber
return glob.glob(os.path.join(path, pattern).replace('[', '[[]'))
File "/usr/local/python/lib/python2.7/glob.py", line 27, in glob
return list(iglob(pathname))
File "/usr/local/python/lib/python2.7/glob.py", line 60, in iglob
yield os.path.join(dirname, name)
File "/usr/local/sabnzbd/env/lib/python2.7/posixpath.py", line 80, in join
path += '/' + b
UnicodeDecodeError: 'ascii' codec can't decode byte 0xbd in position 2: ordinal not in range(128)

Any ideas as to what is going on here?

[shypike]

Looks like an odd character in the title.
Do you have low level (ssh) access to the NAS?
if so, try to rename the "incomplete" job to something else and then restart SABnzbd.

[jlsturgeon]

I do think there was a failed job that had an invalid character in it, despite my setting SAB to replace invalid characters. I manually deleted the folder for it before receiving your response. I now have no more failed jobs in the history to try to purge to check if I'm having further problems there. I still however cannot re-add orphaned jobs to the queue. I'm still getting log entries that LOOK like the orphaned job is being added, but it never re-appears in the queue. The log is still showing like this:
/volume1/DiskStation/downloads/incomplete/The.Wolf.of.Wall.Street.2013.BluRay.720p.x264.DTS-HDW.cp(tt0993846)/__ADMIN__/__verified__ missing
2014-08-12 15:51:21,911::DEBUG::[nzbqueue:167] Repair job The.Wolf.of.Wall.Street.2013.BluRay.720p.x264.DTS-HDW.cp(tt0993846) by reparsing stored NZB
2014-08-12 15:51:21,911::INFO::[__init__:626] Adding The.Wolf.of.Wall.Street.2013.BluRay.720p.x264.DTS-HDW.cp(tt0993846).nzb.gz

I tried using the interface to manually add a .nzb and then browsing to the "incomplete" subdirectory and selecting the .nzb.gz inside the job's folder there. That worked to add the job to the queue again, but it created a new folder for it and the other orphaned folder is still in the "incomplete" subdirectory. That's a tolerable workaround since I didn't have any significant amount of articles downloaded for those orphaned jobs, but it is not pretty.