Re: max. partition size on 5.x

From: Brooks Davis <brooks_at_one-eyed-alien.net>
Date: Mon, 13 Sep 2004 14:29:27 -0700
On Mon, Sep 13, 2004 at 11:02:13PM +0200, Claus Guttesen wrote:
> Hi.
> 
> Not shure whether this is the right place to ask, but
> I haven't found other sources of information.
> 
> I deployed a beta3-nfs-server which works fine. Two
> IDE-to-SCSI-cabinets are connected to the server using
> an aic7899-controller. The first cabinet has approx. 2
> TB (12 disks at 200 GB, as expected). When I connected
> the second cabinet using 250 GB disks, I exptected
> approx. 2.5 TB, but only saw 764 GB.
> 
> Is there any limit that is imposed by 5.x in terms of
> max. partition size? This is running on a Compaq DL
> 380 dual PIII (G2) with 3 GB RAM, microstorage
> cabinet.
> 
> The raid-system became visible by the usual
> 'camcontrol rescan all' and I used 'fdisk -I /dev/da2'
> and 'disklabel -w /dev/da2s1'.

The limit on the size of a disk is unreachable, but we have a lack of
good partitioning tools at the moment. bsdlabel (the program formerly
known as disklabel) is limited to 2TB.  Sunlabel appears to be limited
to 16TB in 2TB partitions and fdisk is limited to 2TB due to bugs in
the program and 4TB due to the actual format.  If you don't need to
boot off the slice, you may wish to try gpt(8) which should support
any disk money can buy, but which we can't currently boot off of GPT
partitions.  I'm in the process of getting some documentation on this in
to the handbook.

-- Brooks

-- 
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529  9BF0 5D8E 8BE9 F238 1AD4

Received on Mon Sep 13 2004 - 19:28:44 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:11 UTC