A7N8X deluxe sata RAID + ide RAID

[Spacey Spade]

Is it possible to have two RAIDs concurrently on the Asus A7N8X
deluxe: One IDE, one SATA? What I understand from the manual (though
it doesn't say so specifically), RAID is only on SATA... but:

What about one RAID on a PCI IDE RAID card, plus another on the mobo's
Serial ATA?

Spacey

 

[Gernot Saborowski]

Is it possible to have two RAIDs concurrently on the Asus A7N8X
deluxe: One IDE, one SATA? What I understand from the manual (though
it doesn't say so specifically), RAID is only on SATA... but:

What about one RAID on a PCI IDE RAID card, plus another on the mobo's
Serial ATA?

Should work!

-Gernot

 

[Ben Pope]

Is it possible to have two RAIDs concurrently on the Asus A7N8X
deluxe: One IDE, one SATA? What I understand from the manual (though
it doesn't say so specifically), RAID is only on SATA... but:

What about one RAID on a PCI IDE RAID card, plus another on the mobo's
Serial ATA?

That'll be fine, but I suspect that you would need to boot from the IDE RAID
partition.

This is 'cos external ATA controllers take preference over the onboard ATA
controller, AFAIK. It shouldn't be a major problem though, but worth
bearing in mind.

Ben

 

[William Barnes]

That'll be fine, but I suspect that you would need to boot from the IDE RAID
partition.
This is 'cos external ATA controllers take preference over the onboard ATA
controller, AFAIK. It shouldn't be a major problem though, but worth
bearing in mind.

Any way of changing this? I'd like to boot from SATA (single drive, not a
RAID), with an IDE RAID controller installed (RocketRAID 454) for my data
drives. Unfortunately, I've just discovered I can't seem to get it to boot
from SATA with the RocketRAID installed... :-(

 

[Ben Pope]

Any way of changing this? I'd like to boot from SATA (single drive,
not a RAID), with an IDE RAID controller installed (RocketRAID 454)
for my data drives. Unfortunately, I've just discovered I can't seem
to get it to boot from SATA with the RocketRAID installed... :-(

I don't think so. I'm not sure how the BIOS decides what a SCSI controller
is and and which one it's going to use.

I've had a play with modbin and can't see anything obvious. What you
require is not impossible, but may not be user configurable in any way.

There's nothing stopping you from sticking a bootloader on the IDE drive and
using that to boot from your SATA one.

Ben

 

[Folkert Rienstra]

Because the SATA is external (PCI). The internal is ATA.
Only(?) Intel has native SATA with Raid support.

Nope, never has been so.

Any way of changing this? I'd like to boot from SATA (single drive, not a
RAID), with an IDE RAID controller installed (RocketRAID 454) for my data
drives.

Unfortunately, I've just discovered I can't seem to get it to boot
from SATA with the RocketRAID installed... :-(

But you can with it removed?

 

[Ben Pope]

Because the SATA is external (PCI). The internal is ATA.
Only(?) Intel has native SATA with Raid support.

Yeah, the Intel ICH5 (I think thats what it is called). nForce2 does not
support SATA or RAID. nForce3 supports both, AFAIK.

Nope, never has been so.

I think you'll find that an external ATA controller on the A7N8X will be
attempted to be booted from instead of the onboard SATA controller if you
select SCSI in the boot order.

But you can with it removed?

Yes, he can.

Ben

 

[Folkert Rienstra]

Yeah, the Intel ICH5 (I think thats what it is called). nForce2 does not
support SATA or RAID. nForce3 supports both, AFAIK.


I think you'll find that an external

Presumably you mean "add-in"?

ATA controller on the A7N8X will be attempted to be booted from instead

of the onboard SATA controller

Which is external of the MoBo chipset.

if you select SCSI in the boot order.

That is entirely possible but that doesn't bite with what I commented on,
i.e. taking precedence over the MoBo chipset native ATA controller.

If the MoBo bios choses to scan the add-in controller first and only, then
that is a flaw in either the MoBo bios or the bios from that particular
controller. It should scan all controllers, not only the preferred one.
There can be a problem though when a drive attached to the add-in
controller has the potential to be bootable but isn't (primary partition).
It may be worth a try to have only an extended partition on that drive
in that case.

Yes, he can.

So the drive is bootable to begin with. Just checking ;-).

 

[Ben Pope]

Presumably you mean "add-in"?

If you prefer.

Which is external of the MoBo chipset.
Yep.


That is entirely possible but that doesn't bite with what I commented
on,
i.e. taking precedence over the MoBo chipset native ATA controller.

Thats not what I said... my comments refer to the particular motherboard in
question which does not have native SATA support. Hence "onboard" and
"external" with respect to the motherboard.

If the MoBo bios choses to scan the add-in controller first and only,
then
that is a flaw in either the MoBo bios or the bios from that
particular
controller. It should scan all controllers, not only the preferred
one.
There can be a problem though when a drive attached to the add-in
controller has the potential to be bootable but isn't (primary
partition).

You mean active?

It may be worth a try to have only an extended partition on that drive
in that case.

Only active partitions are booted (only primary partitions can be active,
but you can have up to 4 of them per drive)

Ben

 

[Folkert Rienstra]

Could you please setup your newsclient so that is doesn't wreck the quoting?

If you prefer.


Thats not what I said...

But that is what it read.

my comments refer to the particular motherboard in question which does
not have native SATA support. Hence "onboard" and "external" with
respect to the motherboard.

The problem with that is that from a logic(al) standpoint, your 'onboard'
(add-on) and 'external' (add-in) are the same, they both sit on the PCI bus.

You mean active?

Yes and No. Yes, 'not active' should be enough to let the bios continue
searching and No, a poorly written bios may see the primary as a boot
candidate and break off the search even though it shouldn't.

Only active partitions are booted (only primary partitions can be active,

Yup, that is how it is supposed to be in the perfect world.

but you can have up to 4 of them per drive)

Not with MS, AFAICT.

 

[Ben Pope]

Could you please setup your newsclient so that is doesn't wreck the
quoting?

Whats wrong with it?

I'm using OE-Quotefix so it might look like anything for all I know.

I just undid the changes it makes... don't see any major problems, if you
expand on the problem I'll see what I can do.

Can you snip and not top post please?



Ben

 

[Ben Pope]

Presumably you mean "add-in"?

Here is a test to see if quoting is still "wrecked"

How is that? Hopefully that sorted it, thanks for drawing it to my
attention.

Ben

 

[Ben Pope]

Could you please setup your newsclient so that is doesn't wreck the
quoting?

Didn't see the bits below...

But that is what it read.

"external ATA controllers take preference over the onboard ATA controller,"

I was commenting on the onboard SATA controller, context should have made it
clear but thats my bad...

The problem with that is that from a logic(al) standpoint, your 'onboard'
(add-on) and 'external' (add-in) are the same, they both sit on the PCI
bus.

Indeed. Logically they both are. But I was talking physically, hence terms
such as "onboard" and "external" I know they both reside on a PCI bus, not
the same one in this case.

Yes and No. Yes, 'not active' should be enough to let the bios continue
searching and No, a poorly written bios may see the primary as a boot
candidate and break off the search even though it shouldn't.

You come across any of these "poorly written" BIOSes? Got an example?

Yup, that is how it is supposed to be in the perfect world.

Got an example of an imperfect world?

Not with MS, AFAICT.

It works fine, MS Fdisk can't deal with it, but the OSs will boot.

Ben

 

[Folkert Rienstra]

Whats wrong with it?

Single word lines caused by auto-linebreaks?

I'm using OE-Quotefix so it might look like anything for all I know.

Aha, I was already wondering howcome the single word lines caused by
the autoline breaks were having the correct number of quote characters.
I didn't even check what you were using as it 'must' have been something
else than OE. Hadn't considered that it might have been an inconsistently
setup OE-QuoteFix.

I just undid the changes it makes... don't see any major problems, if you
expand on the problem I'll see what I can do.

Can you snip and not top post please?

I'll toppost when it is convenient and I won't snip when that's inconvenient.
I'll not toppost when it is inconvenient and I'll snip when that's convenient.
Did I leave out anything or will this suffice for you?

 

[Folkert Rienstra]

Didn't see the bits below...

I wouldn't have seen them either if I had snipped them so mindlessly.

"external ATA controllers take preference over the onboard ATA controller,"

I was commenting on the onboard SATA controller, context should have made it
clear but thats my bad...

i.e. externally from the chipset, I should have added.

Indeed. Logically they both are. But I was talking physically, hence terms
such as "onboard" and "external" I know they both reside on a PCI bus,

not the same one in this case.

Not that it matters, but I don't think so, unless it has a PCI to PCI bridge.

You come across any of these "poorly written" BIOSes? Got an example?

Not really, but when this behaviour surfaces, do you have a better explanation?

Got an example of an imperfect world?

May as well be the MoBo at hand.

It works fine, MS Fdisk can't deal with it, but the OSs will boot.

You are correct.
The check in the MBR is on multiple active partitions,
not just primaries, as I wrongly assumed.

 

[Ben Pope]

Not that it matters, but I don't think so, unless it has a PCI to PCI
bridge.

Multiple ones. There are 3 PCI buses and one AGP on the deluxe. The SATA
controller is on a seperate PCI bus to the nVidia one, but will probably be
the same bus as the add on promise card. Again, not that it really matters.

There seems to be a PCI bus for the southbridge, the slots (including the
SATA controller), the 3Com NIC and then the AGP slot. The non deluxe does
not have the 3Com NIC and presumably has one less PCI bus.

Not really, but when this behaviour surfaces, do you have a better
explanation?

No... never come across the behaviour though.

May as well be the MoBo at hand.

The A7N8X? Doesn't happen on mine.

You are correct.
The check in the MBR is on multiple active partitions,
not just primaries, as I wrongly assumed.

Ben

 

[Folkert Rienstra]

Multiple ones.

That's not likely.

There are 3 PCI buses and one AGP on the deluxe.

Presumably you know this by running a PCI bus sniffer?
Not all buses are actually PCI buses. Native devices are not on the
PCI bus but because they are organized the same way as PCI, show
up as such.

The SATA controller is on a seperate PCI bus to the nVidia one,

Presumably that is meant to read:
"The SATA controller is on a seperate PCI
bus to the nVidia native ATA controler" ?

Yes, the native ATA controler is not on the PCI bus but similar
to the Intel ICH shows up as if it is on a seperate PCI bus.

but will probably be the same bus as the add on promise card.

Yes, the actual PCI bus.

Again, not that it really matters.

There seems to be a PCI bus for the southbridge,

The MCP2 native devices.

the slots (including the SATA controller),

The actual PCI bus.

the 3Com NIC

Would expect that on the physical/actual PCI BUS but maybe
it sits behind a bridge with that many devices and 5 PCI slots.

and then the AGP slot.

AGP is a form of PCI too.

The non deluxe does not have the 3Com NIC and presumably has one
less PCI bus.


No... never come across the behaviour though.


The A7N8X? Doesn't happen on mine.

But do you have the same bios(es) as OP?

 

[Folkert Rienstra]

"Ben Pope" <[email protected]> wrote in message news:[email protected]...

[snip]

There seems to be a PCI bus for the southbridge,

The MCP2 native devices.

Or even the HyperTransport bus.

The actual PCI bus.


Would expect that on the physical/actual PCI BUS but maybe
it sits behind a bridge with that many devices and 5 PCI slots.

On second thought, this board probably has the MCP-T that
has the native 3Com compatible NIC.
Like the native ATA controller it appears as if on a PCI bus.

AGP is a form of PCI too.

[snip]

 

[William Barnes]

There's nothing stopping you from sticking a bootloader on the IDE drive and
using that to boot from your SATA one.

Hey, great! Any chance you could go into more detail as to how I would do
that? I kinda want to avoiding screwing up my system too badly when I try
to move it to a SATA boot drive... :-T

 

[William Barnes]

But you can with it removed?

Yeah. Basically my boot options are:

SATA disabled: choice between onboard IDE or PATA PCI RAID card

SATA enabled, RAID card installed: choice between onboard IDE or PATA RAID
PCI card; the system won't detect a boot partition on my SATA drive; near as
I can tell, it never even bothers checking - it only scans my IDE RAID and
discovers - hey! - there's no OS here! Think I'll stop working!

SATA enabled, no RAID card: choice between onboard IDE or SATA

Basically, I want SATA (boot) + PATA IDE RAID (for data) + onboard IDE (for
optical drives); unfortunately, with my RAID card installed, I can't get my
system to boot from SATA. Someone suggested I use a bootloader, but I've
never used one to try to get my system to boot from a different drive
before - and right now, I'm a little wary of mucking around with my system
too much...

Still, I'd really like to hear some ideas on how to resolve this issue,
please.

[And yes, I know it's been over two weeks since I first asked about this.
What can I say? Thanksgiving is hell... ]