Hello,
I have system with attached SATA array which contains 24 disks. I wanted to run
software RAID 5, but 24 disks means, that I would need /dev/sda to /dev/sdx
devices with major 8 and last minor 384. Minor seems to be limited to 8 bits.
Is there any chance to run software array using all 24 disks?
My test was with mknod v. 5.2.1 and kernel 2.6.14.3
--
Luk?? Hejtm?nek
On Mon, Nov 28, 2005 at 11:25:58PM +0100, Lukas Hejtmanek wrote:
> Hello,
>
> I have system with attached SATA array which contains 24 disks. I wanted to run
> software RAID 5, but 24 disks means, that I would need /dev/sda to /dev/sdx
> devices with major 8 and last minor 384. Minor seems to be limited to 8 bits.
> Is there any chance to run software array using all 24 disks?
>
> My test was with mknod v. 5.2.1 and kernel 2.6.14.3
Major 8 is not the only scsi major. Look at devices.txt in the kernel
Documentation dir. MAKEDEV also usually knows how to make more scsi
devices. For example major 65. Just use MAKEDEV /dev/sdx and see what
it creates.
Len Sorensen
Lennart Sorensen wrote:
> On Mon, Nov 28, 2005 at 11:25:58PM +0100, Lukas Hejtmanek wrote:
>
>>Hello,
>>
>>I have system with attached SATA array which contains 24 disks. I wanted to run
>>software RAID 5, but 24 disks means, that I would need /dev/sda to /dev/sdx
>>devices with major 8 and last minor 384. Minor seems to be limited to 8 bits.
>>Is there any chance to run software array using all 24 disks?
>>
>>My test was with mknod v. 5.2.1 and kernel 2.6.14.3
>
>
> Major 8 is not the only scsi major. Look at devices.txt in the kernel
> Documentation dir. MAKEDEV also usually knows how to make more scsi
> devices. For example major 65. Just use MAKEDEV /dev/sdx and see what
> it creates.
Does udev not know how to handle this?
--
-bill davidsen ([email protected])
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me
On Tue, Nov 29, 2005 at 02:22:35PM -0500, Bill Davidsen wrote:
> Does udev not know how to handle this?
Some people might not run udev. I certainly don't.
When udev stops occationally breaking old kernels while requiring a new
version for a new kernel, then I will consider using it. That might be
a while.
Len Sorensen
On Tue, Nov 29, 2005 at 05:01:21PM -0500, Lennart Sorensen wrote:
> On Tue, Nov 29, 2005 at 02:22:35PM -0500, Bill Davidsen wrote:
> > Does udev not know how to handle this?
>
> Some people might not run udev. I certainly don't.
>
> When udev stops occationally breaking old kernels while requiring a new
> version for a new kernel, then I will consider using it. That might be
> a while.
udev requirements tend to be overstated. I run 2.6.15-rc3 with udev-064
and everything works fine.
--
Tomasz Torcz RIP is irrevelant. Spoofing is futile.
[email protected] Your routes will be aggreggated. -- Alex Yuriev
On Wed, Nov 30, 2005 at 04:34:51PM +0100, Tomasz Torcz wrote:
> udev requirements tend to be overstated. I run 2.6.15-rc3 with udev-064
> and everything works fine.
Perhaps udev is just overrated. :) Maybe if I used usb devices a lot I
would find it useful or something.
Len Sorensen
Lennart Sorensen wrote:
> On Wed, Nov 30, 2005 at 04:34:51PM +0100, Tomasz Torcz wrote:
>
>> udev requirements tend to be overstated. I run 2.6.15-rc3 with udev-064
>>and everything works fine.
>
>
> Perhaps udev is just overrated. :) Maybe if I used usb devices a lot I
> would find it useful or something.
Actually, keeping /dev clean is a reason. I haven't added it to existing
machines with distros no uning udev, I use it on machines which come
with it.
--
bill davidsen <[email protected]>
CTO TMR Associates, Inc
Doing interesting things with small computers since 1979