Selecting RAID controller for my needs

[nanofrog]

Christ, I'm getting the same fault again as with the Mac Pro.

I'm suspecting something if wrong with the driver of the card. This is outrageous.
When I boot again, I see that the array is actually disabled. I have to build it again. I must look for an updated driver.

It appears v.1.3 is the only one available for Windows32/64 (all versions except Win7, which has a BETA version). They don't seem to list older versions.

But I don't think it's the driver. I assume this is a fresh install to the array done on the MP. If it was done on the PC, then it's not going to work (different board components, and causing conflicts).

 

[gugucom]

With 1.3 driver and AHCI driver loaded it crashes on the Shuttle as well. After expanding the files I assume Windows installer tries to boot and there is it where the array gets shot. Can it be that one must not set the array as boot device before?

 

[gugucom]

It appears v.1.3 is the only one available for Windows32/64 (all versions except Win7, which has a BETA version). They don't seem to list older versions.

I was dumb enough to use the driver from the CD that came with the card. I had actually downloaded the newest drivers but somehow missed them. But I still get the problem on the Shuttle. It is not MP related. Something else is wrong when it happens on this BIOS machine as well.

 

[nanofrog]

I was dumb enough to use the driver from the CD that came with the card. I had actually downloaded the newest drivers but somehow missed them. But I still get the problem on the Shuttle. It is not MP related. Something else is wrong when it happens on this BIOS machine as well.

So both the older driver on the CD and the newest from the site are causing the same problem?

You could try going without the AHCI drivers, and only use the RR2642 drivers. It will isolate the drivers for sure, just in case it's an interaction issue.

 

[gugucom]

I have observed that I must partition the free space of the array. And then I assume that Windows will actually format it prior to installing. That is where the array usually gets distroyed. If I format manually from a Windows disk management utility it happens just the same way.

 

[gugucom]

So both the older driver on the CD and the newest from the site are causing the same problem?

You could try going without the AHCI drivers, and only use the RR2642 drivers. It will isolate the drivers for sure, just in case it's an interaction issue.

This is actually my last ditch effort which I am just doing now.


But I think there is something else fishy. Why would the array break down when Windows tries to format it?

 

[gugucom]

If I get problems again I will load a card Bios from the internet site. I'm not sure which I have used.

 

[gugucom]

Crashed again even with just the RR2642 driver.

 

[nanofrog]

Crashed again even with just the RR2642 driver.

Can you determine the card's BIOS revision?

The newest on the site is v. 1.1.

I'll go back and take a look again at the manual, and see if there's something useful in it.

The installation description is a little confusing.

Try this:
1. Delete existing array
2. Create new array
3. Initialize

It looked to me as though they're describing an initialization prior to the array's creation, which is backwards.

 

[gugucom]

There is a parametering facility which I do not understand. Things like
Reallocate EBDA enable
Int13 support enable
BBS validation disable
Single BCV entry Enable
Stop on error Enable
Private stack size 1700
maximum array count 8

 

[gugucom]

Can you determine the card's BIOS revision?

The newest on the site is v. 1.1.

I'll go back and take a look again at the manual, and see if there's something useful in it.

The installation description is a little confusing.

Try this:
1. Delete existing array
2. Create new array
3. Initialize

It looked to me as though they're describing an initialization prior to the array's creation, which is backwards.

I have been using Bios 1.1

I ave never initialized an array after creation. I thought only drives must be initialized.

 

[nanofrog]

There is a parametering facility which I do not understand. Things like
Reallocate EBDA enable
Int13 support enable
BBS validation disable
Single BCV entry Enable
Stop on error Enable
Private stack size 1700
maximum array count 8

OK, this is part of the problem.

Set Reallocate EBDA = DISABLE (Extended BIOS Data Area; allows you to boot from the array)

Leave INT13 = ENABLE (actual boot trigger)

 

[gugucom]

OK, this is part of the problem.

Set Reallocate EBDA = DISABLE (Extended BIOS Data Area; allows you to boot from the array)

Leave INT13 = ENABLE (actual boot trigger)

Ok , I will change that

 

[nanofrog]

Ok , I will change that

Give that a go, and see what happens.

In case you're interested, here's BCV (Business Continuance Volume). Leave it alone for now though.

 

[gugucom]

I'm not sure I did this right because it asked me if I wanted to save the config. I had no idea what file extension to use so I skipped it. Hope it applied just the right config to the flash.

 

[nanofrog]

I'm not sure I did this right because it asked me if I wanted to save the config. I had no idea what file extension to use so I skipped it. Hope it applied just the right config to the flash.

As I'm not sure of exactly what you used to input those values, see if this helps you at all.

From what I can tell, it was a window/tab, with a LOAD button on it, which would save the values to the firmware.

Not quite sure what you used.

 

[gugucom]

How can they deliver a product that is configured to fail?

 

[gugucom]

As I'm not sure of exactly what you used to input those values, see if this helps you at all.

From what I can tell, it was a window/tab, with a LOAD button on it, which would save the values to the firmware.

Not quite sure what you used.

There was a pop up window which asked me to save the config to disk. I declined because I did not know how to name the file.


I have the next crash!!! With the new BIOS!

 

[gugucom]

I'm getting the funny feeling that my disks are not working with that controller.

 

[nanofrog]

I'm getting the funny feeling that my disks are not working with that controller.

For the moment, you need to check those settings.

As per the drives, did you send an email to Highpoint?
I'd mentioned it before, as I thought it odd the only Intel SSD listed was an X25-E model, rather than a X25-M. It's possible they had problems with it, and have kept it off the list. The thing is, software RAID (whether the system or via a Fake RAID card), is usually fine with consumer drives, as the recovery is handled by the system, and not the card.

Proper hardware RAID controllers OTOH, the recovery is handled by the card, and is done differently. Which is why enterprise drives are typically used (stability).

The odds should be in favor of your drives working, but do send off an email to make sure. Just in case they won't, you should have the ability to send them back if you can't retask them to another use/system.

 

[gugucom]

I have now aktivated the partition from within XP and parttioned it without formatting. Perhaps that will stop the Installer from crashing the array.

If that doesn't work I will give up on the Supertalents and use the Intel drives.

 

[nanofrog]

I have now aktivated the partition from within XP and parttioned it without formatting. Perhaps that will stop the Installer from crashing the array.

If that doesn't work I will give up on the Supertalents and use the Intel drives.

I thought you were using the Intel's for Windows.

The HCL only covers a few SSD's (pg. 4). SuperTalent's not on it. A couple of OCZ's, one Intel, one KingFast, and one MemoRight. Not much at all. Granted test drives are always limited, but that's tiny.

 

[gugucom]

I thought you were using the Intel's for Windows.

The HCL only covers a few SSD's (pg. 4). SuperTalent's not on it. A couple of OCZ's, one Intel, one KingFast, and one MemoRight. Not much at all. Granted test drives are always limited, but that's tiny.

Actually Supertalent is identical mit OCZ. Same controller and chips. I realized that I would have needed a special driver for this if it was to work at all. It says so in the list. I wanted the bigger 80GB Intels for OS X because I have a much bigger system. But in the end I gave up. I did not find a spezial driver that would work. So I cloned the OS X system on the Supertalents and tried the Intels with the Shuttle. It worked very akward and the system did not load completely.

I was able to fit the RR2642 and the Intels back into the Mac Pro, where I Wincloned my Vista64 image on the MBR partitioned Intel array. That got me some real chaos because I forced a GPT image on a MBR array. Winclone did great but it installed the system in the big 2nd unpartitioned space of 149 GB. The partition and drive letter got installed in what is usually the invisible EFI driver partition of 200 MB. It was a big miracle that the Mac Pro got it after 10 minutes of searching all together and actually started the system althouh I had to fit Vista HDD to get a boot point in EFI. The result was a strange bastard that had the start partition and the dynamic Windows file on the HDD system.

I took an image of that system and now I try a rudimentary install on the Mac which hopefully will give a stukture better prepared to take a Winclone image. So right now it is all a mixed bag of nuts but I am happy that I can boot with the array at all.

 

[nanofrog]

Actually Supertalent is identical mit OCZ. Same controller and chips. I realized that I would have needed a special driver for this if it was to work at all. It says so in the list. I wanted the bigger 80GB Intels for OS X because I have a much bigger system. But in the end I gave up. I did not find a spezial driver that would work. So I cloned the OS X system on the Supertalents and tried the Intels with the Shuttle. It worked very akward and the system did not load completely.

I wasn't familiar enough with the specific model numbers to know for sure if it was these drives or not (on the presumption some of their drives may be a little different).

I was able to fit the RR2642 and the Intels back into the Mac Pro, where I Wincloned my Vista64 image on the MBR partitioned Intel array. That got me some real chaos because I forced a GPT image on a MBR array. Winclone did great but it installed the system in the big 2nd unpartitioned space of 149 GB. The partition and drive letter got installed in what is usually the invisible EFI driver partition of 200 MB. It was a big miracle that the Mac Pro got it after 10 minutes of searching all together and actually started the system althouh I had to fit Vista HDD to get a boot point in EFI. The result was a strange bastard that had the start partition and the dynamic Windows file on the HDD system.

Don't force settings like that, as in my experience, it's almost always a mistake. That's one of the reasons I've become extremely concerned with 3rd party tools around RAID implementations, as the controll is out of the OS's hands. This even applies to FakeRAID cards, as the drivers are different from the OS (even the tiniest difference can be the cause of a broken array).

I'm actually surprised it booted at all, but it's a good start. You should be able to get it running, but I don't know if you'll actually be able to fix it without another reinstall (my bet would be reinstall, and it can avoid other errors such as driver conflicts).

I took an image of that system and now I try a rudimentary install on the Mac which hopefully will give a stukture better prepared to take a Winclone image. So right now it is all a mixed bag of nuts but I am happy that I can boot with the array at all.

I'm hopeful you'll get it all sorted, but the means may have over complicated things, and cost you a fair amount of additional time. (3rd party tools & "forced" selections). I can't be sure without being there every step (seeing the screens myself during it all), but I've had problems with such things in the past, and am cautious when using them around RAID. It's blown a few arrays on me in the past...

It reminds me of the damage caused by Registry Cleaners, particularly on 64bit versions. Instability at best, but a total disaster is more common.

 

[gugucom]

I have played around quite a bit now and one thing is really a PITA: The RR2642 array isn't shown as a boot option when the option key is pressed. I don't know how to fix this.

I have installed Windows with nothing at all in the mac in termes of HDD. only the array and an Install DVD. After a time EFI selects the DVD and is prompted for the RR242 Bios. I thought by installing Windows there I would get a record in EFI to give me an option. No way I get anathing. It is annoying. Will I have to try rEFIt? I hate installing additional stuff that shouldn't be there in the first place.