Newsgroups: news.software.b
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: ../bad/files
Message-ID: <1990Aug8.163651.4360@zoo.toronto.edu>
Organization: U of Toronto Zoology
References: <1990Aug6.040756.9115@hawkmoon.MN.ORG>
Date: Wed, 8 Aug 90 16:36:51 GMT

In article <1990Aug6.040756.9115@hawkmoon.MN.ORG> det@hawkmoon.MN.ORG (Derek E. Terveer) writes:
>Am I correct that *any* kind of error encountered during the processing of
>incoming news will cause the deposition of the incoming batch file into
>../bad?  ...

Pretty much so.  The exit status returned by relaynews is complex and
not very useful, so newsrun just gives up and puts stuff in "bad" any
time something goes wrong.  What it ought to do is abort processing
entirely, leaving the batch where it is, when a disk fills or something
like that, while shifting stuff into bad when something about that
specific batch caused trouble.  I've been harassing Geoff about this
for some time, and there is hope for more useful exit status in future,
which would permit doing this.

>...i can see that the short article would cause the deposition, but what can
>i do about the history file being too large (disk filled)?  Can i simply move
>the batches in bad back in to the in.coming directory and rerun newsrun?

In general, doing that is always acceptable.  Articles that have already
been processed cleanly will simply be rejected as duplicates, which is
quick and harmless.

>Do "bad" newsgroups cause files to be deposited in bad?...

In general, no.  This is considered a lesser level of problem (there is
always trash on the net) and relaynews does not report it with an exit
status.
-- 
The 486 is to a modern CPU as a Jules  | Henry Spencer at U of Toronto Zoology
Verne reprint is to a modern SF novel. |  henry@zoo.toronto.edu   utzoo!henry
