problem with fast computer and -newer option ?

[Mathieu Roy]

Apparently, find have a problem with fast computers and the -newer
option.


       -newer fichier
              Fichier  modifié  plus  récemment  que  le  fichier
              indiqué.   L'option  -newer  n'est   affectée   par
              l'option  -follow que si celle-ci la précède sur la
              ligne de commande.


I have a little program written in shell that use find to remove
outdated file.

I begin by creating a tmp file.
It touch/create all normal files that are up-to-date
It test files that are not -newer (! -newer) than tmp file and remove
them.

See the debug output. With the same information (so files that are
up-to-date are the same), sometimes somes files are picked (called
ORPHANS) as outdated, sometimes not (in fact they are not). 


address@hidden:~/pdbv# pdbv /var/www -d
DEBUG = 1
FORCE = 
OUTPUT = xhtml
WORKING_DIR = /var/www/pdbv
address@hidden:~/pdbv# pdbv /var/www -d -f
DEBUG = 1
FORCE = 1
OUTPUT = xhtml
WORKING_DIR = /var/www/pdbv
-rw-r--r--    1 root     root            0 Thu Sep 19 18:42:45 2002 
/var/www/pdb          v/package/tmp.test
ORPHANS: /var/www/pdbv/package/tmp.test
/var/www/pdbv/package/adduser.html

address@hidden:~/pdbv# pdbv /var/www -d -f
DEBUG = 1
FORCE = 1
OUTPUT = xhtml
WORKING_DIR = /var/www/pdbv
-rw-r--r--    1 root     root            0 Thu Sep 19 18:43:15 2002 
/var/www/pdb          v/package/tmp.test
ORPHANS: /var/www/pdbv/package/tmp.test
/var/www/pdbv/package/adduser.html

address@hidden:~/pdbv# pdbv /var/www -d -f
DEBUG = 1
FORCE = 1
OUTPUT = xhtml
WORKING_DIR = /var/www/pdbv
-rw-r--r--    1 root     root            0 Thu Sep 19 18:43:40 2002 
/var/www/pdb          v/package/tmp.test
ORPHANS: /var/www/pdbv/package/tmp.test
/var/www/pdbv/package/adduser.html
/var/www/pdbv/package/analog.html
/var/www/pdbv/package/apache-common.html


The computer is a Pentium IV 2.00GHz with 1 Giga Bytes RAM.


Someone else related a similar problem with another fast computer. I
noticed nothing on computer =< Pentium II 350 MHz.

Is it normal, Am I forgetting something?

In the meantime, doing a little `sleep` just after the creation of the
tmp test file should help, I think.

Regards,


-- 
Mathieu Roy
 
 << Profile  << http://savannah.gnu.org/users/yeupou <<
 >> Homepage >> http://yeupou.coleumes.org           >>
 << GPG Key  << http://gpg.coleumes.org              <<