RAID1 - what happens if your motherboard RAID controller dies?

[Squeeze]

Arno Wagner wrote in news:[email protected]

Oh, there is space on the disk that can be used, but there is
no reason why the whle disks would need to be partitioned.
The way this is done is by hiding some sectors of the disk and
possibly remapping the rest. Partitioning is done after and does
not touch the hidden sectors.

Side note: I have ni idea what DDF is (and no need to find out,
since I have never heard the term before ;-), but you can
hide an arbitrary amount of space. Linux software RAID
uses a few kB.

There, that should teach you to not take that babblebot seriously,
ever again.

 

[Christian Franke]

...
Side note: I have ni idea what DDF is (and no need to find out,
since I have never heard the term before ;-), but you can
hide an arbitrary amount of space. Linux software RAID
uses a few kB.

Common RAID Disk Data Format (DDF) is an effort to standardize RAID in
order to provide interoperability between different vendors. It
specifies RAID levels 0-6, with several flavors of RAID 5.

AFIAK, this format is used at least by some recent Adaptec (and ICP)
controllers. There is also a DDF "white paper" from Intel, so this
format may also be used by some Intel controllers. Linux dmraid is able
to detect DDF metadata.

So there may probably be some "need to find out" before further
discussions


@OP: DDF RAID1 does no sector remapping, see section 4.2.2 of the spec.
As a consequence, a disk from a DDF RAID1 array can be accessed as a
single disk on a foreign controller. The metadata block will appear as
unpartitioned space at the end of the disk.

I have seen this layout also on other non-DDF controllers, e.g. 3ware.

Christian

 

[Arno Wagner]

Common RAID Disk Data Format (DDF) is an effort to standardize RAID in
order to provide interoperability between different vendors. It
specifies RAID levels 0-6, with several flavors of RAID 5.

Aha. Interesting.

AFIAK, this format is used at least by some recent Adaptec (and ICP)
controllers. There is also a DDF "white paper" from Intel, so this
format may also be used by some Intel controllers. Linux dmraid is able
to detect DDF metadata.

So there may probably be some "need to find out" before further
discussions

When I find the time.

@OP: DDF RAID1 does no sector remapping, see section 4.2.2 of the spec.
As a consequence, a disk from a DDF RAID1 array can be accessed as a
single disk on a foreign controller. The metadata block will appear as
unpartitioned space at the end of the disk.

I have seen this layout also on other non-DDF controllers, e.g. 3ware.

Linux software RAID also does this. (Note that dmraid is not
Linux software RAID). I believe the main reason for placing
the metadata at the start (as some older controllers do) was
intentional incompatibility.

Arno

 

[bbbl67]

Says Yousuf Khan, comp.sys.ibm.pc.hardware.storage kOOk of the year Award nominee.

Squeeze, I know you enjoy proudly showing off your generally low IQ,
but why lash out at the world due to your father ass-raping you as a
kid? Aren't there other ways of dealing with your sad predicament?
Counseling, anti-depressants, etc.?

Gee, maybe because that is what BIOS means?

Gee, mongoloid-boy, you've read something about this BIOS, and now
you're showing off to the world that you know what it means? I'm so
proud for you, everybody needs hope.

Like that doesn't use the system processor.

Now, now, Squeeze, it's time for some reading comprehension studies,
you've made so much progress otherwise. I have said Windows and BIOS
make use of the central processor.

Or the driver just uses some of the 32-bit routines supplied by the bios
and not know of any array structures, very similar to hardware assisted
RAID.

No, not a chance! The Windows drivers would never make use of routines
inside the BIOS, as the BIOS routines are written for the Real Mode of
the processor, whereas Windows operates in Protected Mode. Real Mode
routines will never work once the processor enters Protected Mode.
Windows device drivers are basically translations of the BIOS routines
from Real Mode to Protected Mode. The BIOS will pass some data
structures stored in memory off to the Windows device drivers, but
once those structures have been passed, the BIOS is ignored and
essentially shut down, while the Windows device drivers take over all
of the hardware I/O.

No difference there, whether single drive, firmware RAID
or Hardware assisted RAID.

Listen homo erectus, who can take you seriously, if you don't even
know the difference between hardware raid and firmware raid?

Go get educated first, or stop coming here, we don't need your idiotic
rants here.

Yousuf Khan

 

[Yousuf Khan]

I posted earlier about deleting Raid.. I will have no problem with
reinstalling the os then Right? The Bios will allow acces to the dvd drive.
even though I have a clean system.? I know it is a stupid question, but I
always wondered when someone built a computer how they ever get to load the
driver for the External drives to install the OS?

As others have said, the BIOS provides support for booting up off of
several different kinds of drives. Most will provide support for
IDE/SATA drives, as well CD/DVD-ROM drives, as well as USB-based drives
(memory sticks and hard disks). Some very specialized BIOSes will
provide RAID drive support too.

 

[Flasherly]

As others have said, the BIOS provides support for booting up off of
several different kinds of drives. Most will provide support for
IDE/SATA drives, as well CD/DVD-ROM drives, as well as USB-based drives
(memory sticks and hard disks). Some very specialized BIOSes will
provide RAID drive support too.

The cheap way is PCI support -- a couple SILICON Image 680 series
chipsets for $15 ea. delivered. Promise, KOUMTEC (sp?), ROSEWILL,
etc., whavetever your poison. Although I'm not much of a raiders fan,
and and after the option to variously run multiple HDs/DVDs, it's with
'most' DVDs there occurs something of a problem. Bus latency,
firmware, I'm not exactly sure the solution, per se -- but buying
either a PCI Serial/Parallel ATA raid/non-raid board these days, most
"imply" if not stipulate DVD channel support -- is a misnomer from
what I'm seeing. Many optical devices will balk at loading or
subsequently misfunction. I've a couple of LG DVD writers I've been
fighting a losing battle with a couple MBs and various controller (MB
+PCI) configs, and after going through discussions, I can say I'm not
the only one. Though NON-RAID appears a suggested optical
alternative, I wouldn't stake a DVD on either for surety. DVD -
issues- are a little more than just the occasional experience with
these new "subset and
cut-&-dry mini MBs" with a spare PCI slot or two, three at most.