[Mailman-Users] old request causes checkdbs failure in 2.1

[Jon Carnes]

First check to see if the held files are still in your data directory:
  /usr/local/mailman/data/heldmsg-....

If they are, go ahead and delete them.  Then simply go to the lists
configure directory and delete the file request.db:
  /usr/local/mailman/lists/<listname>/request.db

Version 2.1 of Mailman is superior to version 2.0 in that it will
re-create request.db if needed.  Before deleting it though, be sure
you've removed the heldmsg-files that are in the ../data directory.

Good luck - Jon Carnes

===
On Sun, 2003-01-19 at 09:58, Karl Berry wrote:
> I updated my site (red hat linux 6.2, i686) to mailman 2.1 and python 2.2.2.
> 
> Unfortunately, I managed to create a request to the new `mailman' list
> during the upgrade which ended up in the old format, and is causing
> cron/checkdbs to fail (see below).  At least that's what I gather from
> the couple of relevant messages I could find about it.
> 
> It's just a junk request, and all I want to do is get rid of it so I
> stop getting the error.  Help?  Where are the pending_requests stored?
> How can I view/delete them?  Or is there another way to fix?
> 
> Any help appreciated.
> 
> Thanks,
> karl
> 
> 
> Here is the actual error:
> 
> bash$ ./cron/checkdbs
> Traceback (most recent call last):
>   File "./cron/checkdbs", line 136, in ?
>     main()
>   File "./cron/checkdbs", line 80, in main
>     text += '\n' + pending_requests(mlist)
>   File "./cron/checkdbs", line 102, in pending_requests
>     when, addr, fullname, passwd, digest, lang = mlist.GetRecord(id)
> ValueError: unpack tuple of wrong size
>