GA-M61SME-S2 HPA makes dynamic disk invalid

[lumstar]

Hi There,

I am moving a 2-disk spanned volume from a Server 2008 machine to a Vista machine. My problem is that my motherboard (GA-M61SME-S2, BIOS ver. f9) backs up the BIOS by writing in a small HPA it creates at the end of the first hard drive. This modifies the partition size on one of the drives in the dynamic pack, making it invalid to windows.

Is there any way of disabling HPA backup of BIOS or selecting the drive with this motherboard?

Thanks,
Stephen

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

You can Zero out the Drive that has that written onto it, and then move your array over. Then the HPA would not be written into again until you flash or update the DMI pool. Once your array is moved it will be ok if new info is written into the HPA

To Directly answer you, no on that board you cannot change the backup method it will write into the HPA and the Backup chip in a NON AHCI/RAID Setup

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

I found when you zero the drive with the Drives manufacturers Dos/Linux bootable media (ie Seatools, WDC Diagnostics, Hitachi DFT) that the Gigabyte Bios has already "cut out" it's HPA usage on the Primary drive and zeroing the drive will only zero those reported as useable by the bios. That equates to all user sectors minus around 2100+ sectors = 1MB+ (assuming 1MB bios image size) where the HPA resides.

The Only way you can realise the whole drive capacity again, is make it a non primary drive or an external drive, where you could then zero the full drives useable area.

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Ohh ya, thanks VorLon!! I use KillDisk windows and DOS versions, have you tested this program and does the above apply? Or did you mean with some other programs that may be free, or just in general other apps? As with KillDisk partitioned or not you can select the disk as a whole or partition

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Ohh ya, thanks VorLon!! I use KillDisk windows and DOS versions, have you tested this program and does the above apply? Or did you mean with some other programs that may be free, or just in general other apps? As with KillDisk partitioned or not you can select the disk as a whole or partition

As far as i know there are only a few proffesional utilities that will give you any form of access to the GB HPA.
When you consider it, that's exactly the purpose of an HPA. It's so that the hardware is protecting an area that is impervious to software manipulation, ie a virus.

I don't know about KillDisk, but I either use the low level utilities (WDC Diagnostics, Seatools, Hitachi DFT) to quickly zero the whole disk or the first and last million sectors, or Acronis Tools.

I don't know how much info KillDisk gives you, but i bet it only shows (if it indicates as such) the user sectors the bios reports as available, ie after it's retained it's HPA.

You can check this by the standard most HD manufacturers lay down for the sectors available to the user.
New Hard drives (ie ones where the reallocated sector reserve pool hasn't run out - so to speak) are advertised as such:-

320GB = 625,142,448 User Sectors
400GB = 781,422,768 User Sectors
500GB = 976,773,168 User sectors
750GB = 1,465,149,168 User Sectors
1TB = 1,953,525,168 User Sectors

The Figures have been taken from the WDC site, but are usually the same for all other makes.

What you should see with a low level program is your bios reporting these amounts when your hard drive is ANY OTHER drive on your Gigabyte System other than the PRIMARY drive.

On your Primary Drive you'll see figures similiar to this, except they will be in the region of (for 1MB bios images) 2100 sectors less.

Out of interest I did use tools such as MHDD, Victoria & the HD manufacturers tools to shrink or delete the HPA. But as the bios is "carving" out that area as it's own and using an HPA for what HPA's were designed for, any HD wiping utility, will only wipe what it is told is available by the bios.

A few of these tools sprang up when some oems (apparently) were sending out larger capacity drives to replace faulty smaller ones. Rather than saying (for instance) "here's a 160GB drive to replace your faulty 80GB drive", they would deliberately SET drive size via HD tools to the original size, denying the user the full potential of the replacement drive. In depth user tools appeared not only to see HPA's exist but to also reclaim (Set Drive Max) the drives full capacity, so the user could use all the replacement drive.

If a utility did display the drives whole advertised amount and say accessed it by different means, such as the Secure Erase feature some new HD's have such as Hitachi (where the drive invokes Self Erasing), I wonder what affect it would have on the normal running of the Motherboard. Especially if you prevent the Write of the bios image into HPA. That could cause booting problems with GB boards.

[lumstar]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

I was able to fix the dynamic disk using a hard drive tool on bootable cd to change the drive size back to its native size, effectively removing the HPA. The 2 drives are back in the original machine and the dynamic disk is working fine.

I'm wondering if there is anything I can do to move these 2 drives (compromising one volume in windows) to the computer with the Gigabyte mobo. You mentioned there is no way to disable or modify the location of the HPA. Is there any other way this could be accomplished without backing up the drive and starting all over? (these are both 1TB drives and I would have to go buy a couple more to backup and start over).

Thanks for the help so far and I look forward to any suggestions!

Stephen

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Ahh well VorLon, I do think KillDisk wipes it then, it even makes it so the disks are non bootable and not set active, I think all data is zero'd using D.o.D specifications (Or worse if you choose to spend that long. >>>




Here is some info about it if you want to look, not sure if it will offer you any knowledge about what we are discussing though>>>

Active@ Kill Disk Hard Drive Eraser. Low Level Format.

I use 3.1 and 5.0 Professional. They do show the correct Sectors, and also show the HPA Section I assume as unallocated, but 1MB on my RAID array as I used to have two drives there instead of three and one section I left out of the array to let the BIOS be copied into it and it worked as shortly after I recovered from it. So needless to say I have left that section alone >>>



Lumstar, can't you just backup the Data you need or move it to another system thru your network and then move the disks over and set them up?

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

I see KillDisk shows the full User Sector Count for a 250GB drive, but it looks like you have Disk 1 selected on the screenshot?

It seems the HPA normally resides on the Primary Drive (Disk 0 if others are connected too)

Unallocated space might have something to do with the Raid setup? Because as far as I can tell KillDisk and other programs like it can only see the size the Bios says - no more no less (and this is after the HPA chunk has been taken). I suppose it could name the HPA as "Unallocated" although it's terminology would be wrong. Programs such as MHDD, will indicate through it's interface if an HPA is present.

The thing is, whatever OS you use, the bios and post are always going to occur Before any software can be used indepenently of it. So the bios takes control from the outset.

I'm assuming that KillDisk is no different in this regard to all the other utilities (like it) I've used on a GB Motherboard.

An HPA wouldn't be very effective (nor really an "HPA") if it could be manipulated by software.

Yep KillDisk has all the usual types of secure erase Acronis has, right up to the very lengthy (but thorough) Peter Gutmann Algorithm.
Secure Deletion of Data from Magnetic and Solid-State Memory

Lumstar:-

I was able to fix the dynamic disk using a hard drive tool on bootable cd to change the drive size back to its native size, effectively removing the HPA. The 2 drives are back in the original machine and the dynamic disk is working fine.

I must admit I'm not too well informed about Dynamic disks, but on your original post you said:-

This modifies the partition size on one of the drives in the dynamic pack, making it invalid to windows.

The partition size will effectively change simply because the Physical drive size is different - but the GB HPA doesn't change any partitions as such.

I will read up on this, but I'm assuming that the Dynamic Disk pair are not happy as they are of unequal sizing? If that is the case and you can't remove the GB HPA (which I'm fairly sure you can't and if you could it would cause board issues), using the HD SET MAX type tools, could you not offset the second drive's size, ie 2100 sectors less (approx). If your board uses the 1024kb size bios flash I should be able to tell you the exact amount of sectors to offset it by. Or i could tell you how to find this info.

The disks might not be an exact match in the way that geometry's line up , ie you may have offsets at the beginning of one drive and at the end of the other, although as that portion is only 1MB in size it wouldn't neccessarily be noticed.

Edit:-

Just re-read your post (again - old age )

So it's not the sizing issue per-se that's causing the problem with your Dynamic Volumes, which I assume are spanned? But the fact that when simply moved the Volume to the GB Board the Bios HPA invalidates the complete Volume, because it essentially alters/corrupts what has already been written?

I've just had an idea.

Get an el cheapo HD and connect it to sata port 0 and install windows... The HPA will be invoked.
Then add your your Dynamic Volume pair to say port 1 & 2. They shouldn't be altered in any way. As long as there are compatabilities between server 2008 and Vista, the Dynamic Volume should stay intact.
I might be missing something, but if you could just use a spare sata basic say 40GB drive, it could prove the point.

Edit 2:-

Didn't Realise this board has only 2 sata ports! However some people might only use the IDE side of things and I'm assuming if only an IDE HD is used the HPA is created on that.
So as above, try installing just an IDE drive and put the OS on it. Then connect up the Volume on sata ports 0 & 1 and Hope! that the board doesn't reassign the HPA. You are limited for options though with this board, as it is a budget board.

Hopefully too you'll have no other hardware issues, because it can be a unreliable process moving something created on one particular hardware setup to another.

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Yes, Vorlon I do have HDD 1 Selected, that is the only NON RAID Drive I have. My RAID is Disk 0

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Yes, Vorlon I do have HDD 1 Selected, that is the only NON RAID Drive I have. My RAID is Disk 0

It must have put the HPA on one of the drives in the RAID array.

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Ya, I think there is due to I left space outside the Array and flashed several times, and then did the same with only my Storage drive and then there was extra "Unallocated" space that was used that was not before.

They say it wont work with RAID Setup, but I am sure it does as long as you leave some space outside of the array, as I said I did it purposely and soon after recovered my BIOS from the Drive and have since a time or two as well. Only once has it recovered from my backup chip since I (Assume) made it place backups into the drives.

I would have to look at my Array outside of windows to be sure which disk the one out of those three it is on, but I did it when I only had two in the array instead of three so I assume it would not be hard to find which.

See ya tonight gotta jet.

Hope your help here helps this user as you seem to know more about this issue then I do!

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

I could understand "Virtual Dual Bios" not working with a Raid Setup and disabled if one is detected - however I've not seen anything from Gigabyte to say as such (you would have thought they would have an asterix somewhere with a foot note saying this feature not available in raid mode etc, if that were the case)

In your setup if that is the case the bios hasn't moved the HPA to the next non-raid drive. So either it completely aborts the feature when Raid is detected, or it still uses part of the array.

Trouble is, many of the good low level bootable utilities, don't work correctly under AHCI and most likely Raid too. Plus even if you did temporarily select IDE mode, so all drives were detected, it would negate the scenario you wanted to check - an HPA would be created as the drive would now be in a standard config.

Seagate are actively updating their windows Diagnostic/Tools (Not to be confused with the Acronis driven Seagate DiscWizard software).

Their Seatools for windows is right upto date and will support most drives (not just seagate) and USB devices too.

What I like about this utility, is that along with generic short and long tests it can invoke the drives short and long self tests even if the drives aren't Seagate.

On the Information Tab, it does give the user sector count on both my AHCI mode drives. I wonder if it will report the same information for your Raid Drives. If it will, then if the HPA is present on One and all your drives are of the same capacity (which I suspect they are) then the drive with the HPA should have a reduced sector Count.

Try with this LSDme.... Seatools for Windows
Seagate Technology - SeaTools

Another Program that doesn't suffer from AHCI type restrictions is PcWizard, where it will display a lot of info about the drives. Not sure about Raid limitations though.
Some Programs (mainly older) won't show things like SMART data on the primary drive on an AHCI enabled system.

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Ya, that is for sure a note Made by Gigabyte, it has been for a while but it concerns the use of another program that does the Backup of the BIOS.

The Dual BIOS feature works from the chips with RAID, it is just not supposed to backup a additional one into the Drives/HPA in RAID or AHCI is all. As it is Actually "Supposed" to be part of Express Recovery, but that does not need to be installed. So that is where I get the RAID/AHCI limitation about it. But it seems that is not so much a factor, but I cannot be sure as I specifically made areas on my Array that was outside of the array so I cannot be sure


I have that installed for a while, it does not work (Fails then crashes) with RAID and my NON RAID member drive as well. This is just the drive information test. >>>



Non Member drive just fails to run, but does not crash >>>




Have you used Microscope before? I just wonder how it fairs at showing this type of info, Ill have to dig it up as I have not used it for a while

Probably some great tools on Hirens that would show it as well, Ill have to Email him and get a updated ICH10R driver before I can check

[VorLonUK]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

Is this with the latest version V1.1.1.0 Lsdme? It does get updated regularly.

The odd thing is too, that even your Seagate Single drive is identified as Seagate (hence the Gold Icon) and yet it fails to initialise the test.

I've found all the Recent (last year or so) Seagate Versions of this tool have worked with AHCI and my WDC drives, although some of the advanced tests are unavailable and for Seagate drives only.

[Lsdmeasap]

Re: GA-M61SME-S2 HPA makes dynamic disk invalid

No, that is 1.1.0.21

it may fail just because it is controlled by the RAID or maybe it just fails for both cause I need to update it? Maybe the version I have is not update enough for ICH10R? Ill grab the new one and see then edit this post

*EDIT*

Same errors with the new one, so it must not be able to read RAID Disks, or even NON RAID Disks controlled by the RAID BIOS. It will test them, just will not provide drive information