Subject: Re: mkfs problem (solved) with large disks
To: Brad Salai <bsalai@servtech.com>
From: Bob Nestor <rnestor@metronet.com>
List: port-mac68k
Date: 11/08/1996 20:16:30
>Is the problem with the old mkfs apparent when you first use it, or does 
>it lurk
>in the background waiting for the space above the upper limit to be used?
>I have a 1.6G netbsd partition that I'm almost sure I made with mkfs from 
>netbsd, not from the Mac mkfs utility about 2 months ago. So far, I've had 
>no problems with it, but I've been below about 1 G of space actually used. 
>I'm getting close now and wonder if I'm going to trash the fs someday?

You'd know almost immediately after running the old mkfs!  Any disk block 
address above the 1Gig limit would have been wrapped to a lower disk 
block address during the initializtion process.  Depending on what that 
wrapped address was it could have zapped your disk Partition Map, Disk 
Driver Partition(s), or (misplaced) MacOS Partitions.  Most likely you 
partitioned your disk placing the Root&Usr first, followed by a SWAP and 
then the MacOS partitions. This is the default placement for most disk 
formatters I've tested.  Since you probaby didn't run mkfs on the SWAP 
(it's not necessary), the position of SWAP didn't matter. As long as the 
top of the last Root&Usr Partition that you ran "mkfs" (MacOS version) on 
was under 1Gig, you're in fine shape.  Also if the Root&Usr Partition had 
crossed this boundary and by some fluke nothing important got zapped by 
"mkfs", the first time you booted into NetBSD and it ran fsck it would 
have complained about the filesystem not being properly initialized.

Hope this helps explain things,
-bob