[Mailman-Users] start script started reporting errors

[Mark Sapiro]

Myron Kowalski wrote:
> 
> The qfiles/virigin has a thousand messages that aren't going away. And  
> when I
> start mailman by hand I get the following errors that repeat 4 or 5  
> times.
>  From these errors, can anybody guess as to what may have happened?
> 
> Traceback (most recent call last):
>    File "/users/mailman/bin/qrunner", line 270, in ?
>      main()
>    File "/users/mailman/bin/qrunner", line 230, in main
>      qrunner.run()
>    File "/users/mailman/Mailman/Queue/Runner.py", line 65, in run
>      filecnt = self._oneloop()
>    File "/users/mailman/Mailman/Queue/Runner.py", line 93, in _oneloop
>      msg, msgdata = self._switchboard.dequeue(filebase)
>    File "/users/mailman/Mailman/Queue/Switchboard.py", line 156, in  
> dequeue
>      data = self._ext_read(dbfile)
>    File "/users/mailman/Mailman/Queue/Switchboard.py", line 262, in  
> _ext_read
>      dict = marshal.load(fp)
> ValueError: bad marshal data
> Traceback (most recent call last):
>    File "/users/mailman/bin/qrunner", line 270, in ?
>      main()
> and so on....


This appears to be Mailman 2.1.4 or earlier.

The issue appears to be a corrupt .db file in qfiles/virgin/.

Each queue entry is a pair of files with the same name except one has a
.db extension and the other has .pck or possibly .msg extension.

The rest of the file name is a floating point time (ttttt.ttt) followed
by '+' followed by 40 hex digits.

The file causing the problem is the .db with the earliest time. Move
that file aside and stop and start Mailman.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan