[lkml]   [1996]   [Jun]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: , Drawbacks of implementing undelete entirely in user space
On Sat, 22 Jun 1996, Todd Graham Lewis wrote:

> I would think that , such an automatic expiration process could be handled
> in a daemon, rather than in the kernel. Such an fs monitor could also
> serve as a dropping-point for such files; a certain portion of disk space
> or a separate (encrypted? gzipped? jk 8^) partition could be set aside
> for such files, providing a bulkhead against out-of-control , "deleted
> files". Undelete is always and necessarily a hit-or-miss thing, so a
> limit on undeleted files would be acceptable.

Not even that - a crontab entry containing a shell/perl script that would
remove files from .wastebasket after a 'days since modified'/number of
files in .wastebasket/size of .wastebasket would be simple to program and
very small (I'd think under 5k). Then rename rm to something else that
script (and people who really really want to delete their files) can
call, and make a new rm that moves the file to the .wastebasket and
writes its path in a data file. Total modification size, including both rms
and man pages would be under 50k.

Lost anybody there? :)
Rich Tollerton

NOTE: rich is NOT the usename you want to send to - send to rtollert
instead. It has to do with Pine not being able to use a different
username for POP servers other than the one it's being run under....
suggestions welcome.

 \ /
  Last update: 2005-03-22 13:37    [W:0.104 / U:5.264 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site