spaf@gatech.edu (Gene Spafford) (08/17/87)
The question is why are you running expire with the "-r" option twice a
week? The rebuild destroys information in the history file designed to
quash old "phantom" postings. If you want to make sure that your
expire examines everything in the spool directory, use the "-h" option
instead. That does an expire while ignoring the history file, thus
examining every file but not losing the other history information.
The "-r" option should only be used after some catasrophic event which
completely trashes your history file or spool directory.
In more detail: when an article expires, its entry in the history file
is not immediately removed. Instead, it stays around for a few more weeks
(5 is the default, I believe). If the article returns in that time period
due to a loop, or due to an extremely short expiration interval at your site,
the duplicate is rejected since the article ID is still in the history file.
When you do an "expire -r" it trashes all that information and there is no
way to recover it since those articles are long gone from your system.
Thus, you totally defeat that mechanism.
--
Gene Spafford
Software Engineering Research Center (SERC), Georgia Tech, Atlanta GA 30332
Internet: spaf@gatech.gatech.edu
uucp: ...!{decvax,hplabs,ihnp4,linus,rutgers,seismo}!gatech!spaf