From: "..." <yesss AT escape DOT com>
Message-Id: <199904022138.QAA08047@escape.com>
Subject: Re: FDISK problems
To: opendos AT delorie DOT com
Date: Fri, 2 Apr 1999 16:38:35 -0500 (EST)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Reply-To: opendos AT delorie DOT com

Having MS DOS 6.0 FDISK'd and formatted my 1gb formerly 
Caldera dos beta 7.03-partitioned drive, I find myself 
with a new "challenge."
 
This disk has not one bad physical cluster from one end 
to the other. Nothing kinky ever has been done to the 
disk--no FAT-32, no Warp, no PartitionMagic, no file 
compression, etc. Only Caldera FDISK, which strangled 
MS DOS, thus of course preventing win95 installation, 
apparently as the consequence of the bugs Charles Dye 
carefully analyzed in a report he submitted to Caldera 
and posted to the Delorie OpenDos mailing list.
 
My partitioning, old Caldera and new MS:
 
type  status   Caldera type        Caldera     MS   total alloc units
----  b - - -  ------------------- -mb-------- --mb Caldera--------MS
cpq 
diag  n 1                            3.9   3.9    4                  
----- - - - -  ------------------- ----------- ---- -----------------
dos   a 2 1 C: 16B (BigDOS >32 mb) 120   120    120  30,709    61,359
dos   n 3 2 D: 16  (BigDOS >32 mb) 523.3  64.9   65  16,603    33,175
          3 E: 16  (<=32mb)               31.4   32  16,072    16,072
          4 F: 16  (<=32mb)               31.4   32  16,072    16,072
(!)       5 G: 16B (BigDOS >32 mb)       366.1  364  46,843    46,591
          6 H: 16  (<=32mb)               29.5   32  15,066    16,072
----  - - - -  ----- ------------------------- ---- -----------------
      n        free space          385.8 385.8                       
 
Logical drives, including C: and D:, all seem to be behaving 
predictably except G:. G:--which I've reserved from the 
gitgo for win95--is in a lost cluster/wrong file length 
vicious cycle I can't resolve. I've MS-reformatted and 
relabeled G: twice (it doesn't retain the label). I haven't 
a clue what to do about it (but--no offense intended, guys--
but, please, no PartitionMagic suggestions).
 
I doubt that the symptoms matter very much, but 
I'll describe what happens: After chkdsk /f apparently 
(not really) clears 2 lost clusters in 2 chains on an empty 
G:, Norton Disk Doctor reports a lost cluster chain. 
If NDD is told to del the cluster, NDD says it has 
but it hasn't, and keeps generating the same error msg 
till it's told to assign the cluster to a file. Then it 
gives the drive a clean bill of health and chkdsk detects 
an invalid cluster in G:\FILE0000._DD, then G:\FILE0000._DD 
has an incorrect length. After chkdsk /f purports to fix 
G:\FILE0000._DD, NDD says \file0000._dd has size allocation 
errors and the drive has lost chains. NDD claims to fix 
the file length and assigns the chain to UNKNOWN (a new 
one on me), but in fact generates the same msgs over and 
over. Deleting G:\FILE0000._DD takes us back to the 
beginning of the vicious cycle.
 
Is this an MS fdisk /mbr moment? Heeeeelp! Please? 	--Fisher
 
======================================= adpFisher <yesss AT escape DOT com> nyc