SEO
vBulletin Search Engine Optimization
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
02-21-2008, 04:53 AM
| ||||
| ||||
 Re: ATARAID Weirdness, can't "see" two arrays. T.G. Reaper schrieb: > On Thu, 12 Jun 2003 08:18:47 +0100, Alan Hughes wrote: > > > >>>The only problem is when *both* arrays are attached to the controller. >>>When both arrays are attached, Linux (Mandrake or Gentoo) only "sees" the >>>secondary array, even if it's the primary array that contains Linux. >> >>That RAID configuration are you using? The ataraid subsystem can only handle >>RAID-0 at the moment; RAID-1 is unsupported. > > > Well, that shouldn't be a problem as *both* arrays are RAID-0. Why is it > that Linux can't see two RAID-0 arrays when they are both connected to the > controller. Yet it can can see either one when it is connected by > itself? > I assume the hpt bios can see both arrays? I have one idea - both arrays were created independent of each other, meaning you created the win array, unplugged created the gentoo and then replugged the win - at least that's how I understood your procedure. Could it be that both have the same identifier of some sort as both were the first and only array at creation time. I'm no raid expert, but the first array is numbered 0 (or so) in most cases. If both are 0 or share a simmilar ID ... Well it's just an idea. Jesore      |
|
02-21-2008, 04:54 AM
| ||||
| ||||
| Re: ATARAID Weirdness, can't "see" two arrays. On Wed, 16 Jul 2003 13:19:48 +0200, Marc Remolt wrote: >>> T.G. Reaper schrieb: >>>>The only problem is when *both* arrays are attached to the controller. >>>>When both arrays are attached, Linux (Mandrake or Gentoo) only "sees" >>>>the secondary array, even if it's the primary array that contains >>>>Linux. > I assume the hpt bios can see both arrays? Yes. > I have one idea - both arrays were created independent of each other, > meaning you created the win array, unplugged created the gentoo and then > replugged the win - at least that's how I understood your procedure. No the primary array was attached to the controller during the creation of the secondary array. Also the HPT-Bios has a field for the "Array Name" and both arrays were named differently. > Could it be that both have the same identifier of some sort as both were > the first and only array at creation time. I'm no raid expert, but the > first array is numbered 0 (or so) in most cases. If both are 0 or share > a simmilar ID ... I *am* somewhat of a RAID expert, but most of my experience is with hardware and software SCSI implementations. The HPT-Bios numbers the arrays as zero and one depending upon which is designated as the "boot" array. Since we are talking about separate physical drives on different IDE channels it's hard to imagine why a label would cause completely separate physical drives not to be seen at all. This seems to be a bug/limitation in the Linux HPT driver. That's really the most logical conclusion I can come up with at this point. -- Cheers T.G. Reaper Linux Thunder 2.4.20-gentoo-r5.smp.acpi #6 SMP Fri Jul 4 2003 Dual 1.0 GHz Pentium III (Coppermine) GenuineIntel GNU/Linux *************Reaper At Debug1 . com*************************** |
Linear Mode
