Many good points, Dave, to which I'd add two more.
(1) It should be somewhat faster when it comes time to fsck a partition when
they're smaller. The partitioning will also restrict the damage if/when
something happens to the disk.
(2) I've found it quite useful to set aside an archive partition, which is
maintained across updates, reinstalls, etc. One of the things I put on here
is an RCS directory where I can checkin/out critical system configuration
files as well as development projects.
[EMAIL PROTECTED] said:
> > What is the "technical" reason for doing this (Splitting up a
> partition that is). Does it make it faster? Thanks for the info...
> Several reasons. First, no, it doesn't make it faster, until you
> split your partitions across drives (specifically SCSI) or buses/
> controllers.
> On a home machine, it doesn't really make a damn bit of difference,
> and perhaps can be argued its more trouble than its worth.
> On a server machine, creating multiple partitions has several
> advantages:
> - filling up your /var paritition, for example, with log data won't
> crash your box, due to your root partition not having any space.
> - easier to backup, assuming the partition is still small enough to
> fit on a tape. This way you can store one whole partition on a tape.
> - improved ability to increase security. This way you can mount your
> user's home directory read-only, or noexec, or nosuid, preventing
> unauthorized execution of programs they would otherwise have no place
> to write.
> See the Filesystem security section of the new Security-HOWTO,
> co-authored by yours truly. Hopefully its made it to sunsite by now..
> Dave
--
Rick Forrister <[EMAIL PROTECTED]>
Opera: Greek word meaning "death by music".
--Anonymous
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
