Re: Problems creating RAID 5 using mdadm

2006-05-10 Thread Robson
Good way of resolve this problem is to compile drivers into kernel. Then the SATA controllers are initiated too eariler than the array building. More about SATA and arrays using mdadm, but in polish you can find at http://bu.bee.pl/articles.php?id=1 Regards, Robson Clive Menzies wrote: On (1

Re: Problems creating RAID 5 using mdadm

2006-05-10 Thread Clive Menzies
On (10/05/06 16:10), Ruggiero, Stephan wrote: > Hi, > > I figured it out: > > The problem was that during the boot process the SATA controllers are > initiated after the array building. The IDE controllers are initiated > earlier, that's why the RAID1 with the IDE drives worked... > > So I did a

Re: Re: Re: Problems creating RAID 5 using mdadm

2006-05-10 Thread Ruggiero, Stephan
Hi, I figured it out: The problem was that during the boot process the SATA controllers are initiated after the array building. The IDE controllers are initiated earlier, that's why the RAID1 with the IDE drives worked... So I did a "lsmod | grep sata" and filled in the sata module names to the

Re: Re: Problems creating RAID 5 using mdadm

2006-05-09 Thread Ruggiero, Stephan
Hi Lee, Thank you, I will try that - but the biggest problem for me now is not the failure of fsck but that I cannot use (read/write) the array... maybe I missed something during the creation? I tried several times but no luck... Best regards, Stephan -- To UNSUBSCRIBE, email to [EMAIL PROTEC

Re: Problems creating RAID 5 using mdadm

2006-05-09 Thread lee
On Tue, May 09, 2006 at 07:25:22PM +0200, Stephan Ruggiero wrote: > Then I created the file system: > mkfs.ext3 -b 4096 -R stride=16 /dev/md1 > __ > fsck.ext3: Invalid argument while trying to open /dev/md1 > /dev/md1: > The superblock could not be read or does not describe