symtom: errors at list startpage
cPickle.UnpicklingError?: invalid load key, ' '.
path/to/mailman/<listname>/config.db
replace it with config.db.last or an earlier backup of config.db
symptom: errors at admindb
path/to/mailman/<listname>/request.db
replace it with an empty request.db from another or a new list
see http://mail.python.org/pipermail/mailman-developers/2001-December/010269.html for a py-script to check single request.db files
symptom: qrunner says
Traceback (most recent call last): File "/home/pacs/mih15/mailman/bin/qrunner", line 270, in ? main() File "/home/pacs/mih15/mailman/bin/qrunner", line 260, in main qrunner.run() File "/home/pacs/mih15/mailman/Mailman?/Queue?/Runner?.py", line 59, in run filecnt = self._oneloop() File "/home/pacs/mih15/mailman/Mailman?/Queue?/Runner?.py", line 88, in _oneloop msg, msgdata = self._switchboard.dequeue(filebase) File "/home/pacs/mih15/mailman/Mailman?/Queue?/Switchboard?.py", line 144, in dequeue data = self._ext_read(dbfile) File "/home/pacs/mih15/mailman/Mailman?/Queue?/Switchboard?.py", line 246, in _ext_read dict = marshal.load(fp) ValueError?: bad marshal data
??
solution: run all qrunners seperatly to get an idea at which part the problems occurs.
here the error occured with ./qrunner -o -r VirginRunner? and ./qrunner -o -r BounceRunner?
after looking at one of the *very* few postings on this issue at http://mail.python.org/pipermail/mailman-developers/2003-August/015525.html a closer look at the files at ../mailman/qfiles/virgin and /bounce made it clearer: some *.db and *.pck were not readable, thus corrupt.
delete these files, and re-run qrunner
in this special case i had to delete all files at /shunt and /bounce ( ./unshunt didn't help)