Personally, I like LVM (Logical Volume Manager). However, this must be set
up in advance. It allows multiple physical disks to be "joined" into a
logical volume group which can then be split up into logical volumes. These
logical volumes can be larger than a physical volume. And you can generally
expand a filesystem "on the fly" if there is sufficient space left in the
volume group. Look at http://tldp.org/HOWTO/LVM-HOWTO/index.html

However, in your case, I would likely split the /usr filesystem. I tend to
split into /usr vs. /usr/local. I may then split /usr/local down even more
if I get a large application.

--
John McKown
Senior Technical Specialist
UICI Insurance Center
Applications & Solutions Team
+1.817.255.3225


> -----Original Message-----
> From: David Booher [mailto:[EMAIL PROTECTED]
> Sent: Friday, March 14, 2003 3:35 PM
> To: [EMAIL PROTECTED]
> Subject: Multiple mount points - 3390 DASD Limitations
>
>
> Hello listers,
>
> I've been pondering this question for some time now.  I have
> older RAMAC II
> DASD arrays that emulate 3390-3's.  Each pack is roughly
> 2.36Gb.  Let's say
> a mountpoint, like /usr, starts to get close to 100%
> utilization.  Is the
> appropriate thing to do is create separate mountpoints, like /usr/lib,
> /usr/share and devote each of them to their own 3390 pack?
>
> I am looking at a new SHARK array and that will help me solve
> this problem
> by emualting larger disks, but is there anything I can do in
> the interim?
>
> Thanks,
> Dave
>

Reply via email to