Urgent help needed with "stuck" bnews spooled postings

Michael G. Beirne mikebe at i88.isc.com
Tue Mar 12 02:18:26 AEST 1991


In article <1991Mar08.200236.10786 at ferret.ocunix.on.ca> clewis at ferret.ocunix.on.ca.UUCP (Chris Lewis) writes:
>In article <2026 at public.BTR.COM> thad at btr.btr.com writes:
>>I posted an "en passant" remark about a bunch of my postings to comp.sys.3b1
>>never making it out to the world.  Last night, at dinner after a computer
>>meeting, Dave Herron (dherron at twg.com) commented in response to my question
>>the files could be found in /usr/spool/news/.rnews and that they were placed
>>there while expire is running (causing the "Your post is spooled for later
>>processing" messages when originally posted).
>
>>I checked, and, by golly, they ARE there!
>
>>Question: HOW to fire these up so they make it out to the world?
>
>I sent mail, but my mail system is a bit flakey at the moment, so I'm
>posting too:
>
>In bnews, to unstick news, run "rnews -U".  It's a good idea to place

This should be run as user news. One place I worked limited the size
of a file a user could create to 2 meg. The history file for news is
usually larger than 2 meg. Child processes inherit the limits of their
parents. Therefore if you run "rnews -U" on a system that has such a limit
you will corrupt the history file and cause even more grief for yourself as
old articles will not get expired because they are not in the history file.

To check the size of files you are allowed to create type "ulimit".
This works in both csh and ksh but not in sh.

>an "rnews -U" into your crontab to be run a couple of times per day.
>Especially an hour or so after your cron fires up expire.
>B-news is SUPPOSED to run an "rnews -U" at the end of an expire run,
>but I've seen *that* not work on some systems.
>-- 
>Chris Lewis,
>clewis at ferret.ocunix.on.ca or ...uunet!mitel!cunews!latour!ecicrl!clewis
>Psroff support: psroff-request at eci386.uucp, or call 613-832-0541 (Canada)
--
mikebe at i88.isc.com or beirne at chinet.chi.il.us



More information about the Comp.sys.3b1 mailing list