CNFS offset problems

[Russ Allbery]

Jeffrey M Vinocur <jeff at litech.org> writes:

> /usr/local/news # grep '^[^#]' etc/cycbuff.conf
> cycbuff:CONTRL1:/usr/local/news/spool/cnfs/control1:10000
> cycbuff:GENERL1:/usr/local/news/spool/cnfs/general1:1500000
> cycbuff:GENERL2:/usr/local/news/spool/cnfs/general2:1500000
> metacycbuff:CONTROL:CONTRL1
> metacycbuff:GENERAL:GENERL1,GENERL2:SEQUENTIAL

> /usr/local/news # du -k spool/cnfs/*
> 56840   spool/cnfs/control1
> 1500000 spool/cnfs/general1
> 1500000 spool/cnfs/general2

Well, there goes my theory that it was a signed/unsigned overflow related
to my cycbuffs being exactly 1GB in size....  Although come to think of
it, that doesn't make sense anyway.

Hm.  Not sure where to start looking for this one.  I guess I can go try
to understand the code that detects that a cycbuff is full.

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

    Please send questions to the list rather than mailing me directly.
     <http://www.eyrie.org/~eagle/faqs/questions.html> explains why.