[ale] Software RAID Problems

[Jason Vinson]

Hi Danny,

hdparm did report dma on, so i used d0 to turn it off on both drives, 
but it immediately hangs upon the mkraid if i do this, so i'm thinking 
that's not a good idea  ;)

I am going to show my greenness here, but how do i run bad blocks on 
the drives?  Are you recommending using dd to write zero's to the 
platters?

Jason


On Mar 1, 2004, at 7:24 AM, Danny Cox wrote:

> Hi, Jason!
>
> On Sun, 2004-02-29 at 20:54, Jason Vinson wrote:
>> I looked in the BIOS and I don't see a setting for toggling DMA.
>>
>> But here's what I see from dmesg:
>>
>> (this is my onboard ide controller)
>> ide0: BM-DMA at 0xfc00-0xfc07, BIOS settings: hda:DMA, hdb:DMA
>> ide1: BM-DMA at 0xfc08-0xfc0f, BIOS settings: hdc:DMA, hdd:DMA
>>
>> (and these are the Promise IDE channels)
>> ide2: BM-DMA at 0xdf50-0xdf57, BIOS settings: hde:pio, hdf:pio
>> ide3: BM-DMA at 0xdf58-0xdf5f, BIOS settings: hdg:pio, hdh:pio
>>
>> hde and hdg are my potential RAID devices, and they are set to pio.  
>> So
>> is pio conflicting with DMA somehow, or should this not be affected by
>> DMA at all? And isn't there a way to turn off DMA via hdparm, or will
>> that not work here?
>
> 	That's just the kernel reporting how the BIOS was set up.  There are
> probably messages just after that which show what else the driver set
> up.  Use "hdparm /dev/hde" and see if DMA is 0 or 1.  If necessary,
> sometime during bootup, add "hdparm -d1 /dev/hd{e,g}" to force DMA on.
>
> 	But your previous post showed the drives complaining about errors.  On
> second glance, I see that 'dma_intr' is being called, confirming that
> DMA is being used on hde at least.  The UncorrectableError is the main
> culprit here.  Try running badblocks on the whole drive (smoke 'em if 
> ya
> got 'em, it'll take a wee bit ;-), and see what it says.
>
> 	HTH!
>
> -- 
> kernel, n.: A part of an operating system that preserves the
> medieval traditions of sorcery and black art.
>
> Danny
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://www.ale.org/mailman/listinfo/ale