[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

RE: Partitioning problem with windows vs. linux



>
>
> Date: Thu, 1 Sep 2005 01:13:09 -0400 From: Tony Nelson
> <tonynelson georgeanelson com> At 9:13 PM -0400 8/31/05, Michael
> Wiktowy wrote:
>
>>>I am not formating the partition in any way. I am just adding a new
>>>empty one using fdisk and no matter what "partition type" I choose, w2k
>>>hates it ... even if I choose type 07 which is NTFS. When I first
>>>encountered the problem while installing Fedora, I had manually made
>>>several partitions and had them formatted ext3. Fedora would boot and
>>>run fine but then w2k would not. I have stripped away little by little
>>>to see where the problem is located and it appears that it is having
>>>issues when just an empty partition is added. This is really unusual
>>>since w2k shouldn't care how the rest of the drive is partitioned so
>>>long as it was the first partition.
>>    
>>
> ...
>
>Well, if you're just partitioning, then the old data is still at the same
>place it was, so conceivably Windows might be confused by it.  I don't
>think that's actually the problem, tho.
>
>Have you rooted around in Microsoft's Knowledge Base for anything about
>such problems (Win2K, large disks, new partitions, won't boot)?
>  
>

I have stumbled upon a few MSKB articles in my Googling about this
problem but nothing that was terribly helpful.
There is no doubt that there is nothing but junk inside the newly
created second partition but, AKAIK, w2k shouldn't be looking inside
there at al and it would interpret a valid ext3fs as junk too.

I took a look at the partition information with a hex editor last night
with reference to this website:
http://www.ata-atapi.com/hiwtab.htm
The partition table looked clean in that there was no extra junk
floating around and the partitions were clearly defined. Adding of the
second partition information did not affect the first partition info at all.
However, the fields annoted as: "Starting/Ending CHS in INT 13 call
format" seemed to max out such that the first partition started at 01 01
00 as it should and ended at FE FF FF. The second partition both started
and ended at FE FF FF. Now I suspect that Linux just ignores these
values and uses the LBA ones later on in the record (which were sane)
but w2k might be having a bit of trauma over this. According to
everything I have read, it shouldn't have any problems with big
partitions like this though.

Still looking for ways to get w2k over its psychological barrier ...

/Mike


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]