Rebuilding history after a powerdown

Russ Allbery rra at stanford.edu
Tue Jan 9 10:02:29 UTC 2001


StediS <stedis at radiolink.net> writes:

> Yesterday  we had a powerdown, history was corrupted so I tried:

> makehistory -O -b

> watching news.notice I've seen a lot of "bad article handle" errors

> tradspool: can't determine class: [...] : Bad article handle

This happens whenever you have articles in your spool that were
crossposted to newsgroups stored in tradspool and to newsgroups stored in
some other storage method and that, when running makehistory, end up
getting put into a different storage method than tradspool.  In other
words, it usually means that storage.conf has changed.

Those articles end getting a class of zero, and I'm not sure what that
then does to their retrievability.

> but configs are ok.

> Anyway it finished but just with a partial success.

> HEAD nntp commands with Message-ID as argument always return 430.

> I'm rebuilding history for the second time, but I'd like to know if
> there is a specific better strategy to deal with these situations. A
> safe way to recover only once.

Have you changed your storage.conf?  If so, in what ways?

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>



More information about the inn-workers mailing list