grephistory?

George Lindholm George.Lindholm at ubc.ca
Wed Aug 25 15:47:36 UTC 1999


meow wrote:
> 
> In nnmh:inn-workers, George Lindholm <lindholm at ucs.ubc.ca> wrote:
> 
> > On Wed, 25 Aug 1999, Katsuhiro Kondou wrote:
> 
> >> The token is different.
> 
> > I realize that. Why are they different? They are supposed to be the same
> > since they from the same source.
> 
> If you sort your history file, you'll likely find lots of duplicate
> hashes.  Tagged-hash indexing is b0rken.  Articles come in, get
> properly written to the history text and filed away, but they don't
> show up in a dbz query.  So, the article gets offered again, and is
> accepted again, but this time time indexing happens to work, and there
> you are.

That pretty wells describes my situation. Fortunately I'm still running
in slave mode (in preparation to cutting over) so I don't have any
duplicate hashes (yet). 

If tagged-hash is so broken (ie. useless) why is it still an option?

   George
-- 
George.Lindholm at ubc.ca       ITServices, UBC
Programmer/Analyst

phone:    604.822.4375       fax:      604.822.5116


More information about the inn-workers mailing list