lkml.org 
[lkml]   [1996]   [Jun]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: SCSI device numbering (was: Re: Ideas for v2.1
On Thu, 20 Jun 1996, Eric Youngdale wrote:
> Consider that some cards drive multiple channels. Currently some
> cards support 2,

Make that 5.

> I have seen some people argue for solutions that work for the small
> systems, but exclude the large systems. I believe that this is fundamentally
> wrong, since the problems with dynamic address shifting grow as the number
> of devices on the sysetem grow. We should be looking for solutions that
> work in the case of large systems (with say 500 or more disk drives), and
> then make sure that this works well for the small systems.
>
> The scsidev program is basically sort of a band-aid until a better
> solution arrives. I do believe that it should alleviate most of the
> problems that people are having until that time comes.

Another thing that would help would be to give very disk and partition a
unique name in the system. That way it would make no more problems
to move a drive from controller A channel B ID C to controller D channel E
ID F, it would be possible to move disks around in the system
without getting problems everytime if you forget to update the fstab.

The fstab could look like this:
root / ext2 defaults 1 1
usr /usr ext2 defaults 1 2
home1 /home/u1 ext2 defaults 1 2
home2 /home/u2 ext2 defaults 1 2
var /var ext2 defaults 1 2
tmp /tmp ext2 defaults 1 2
postgres /postgres ext2 defaults 1 2
none /proc proc defaults 1 2
swap1 none swap
swap2 none swap
swap3 none swap

Convex (now HP supercomputer divison) has implemented this and it has proven
to be very useful.

> Note that with the current system, there can be a maximum of 16 disk
> drives (since 16 * 16 partitions = 256 = max minor number). Clearly a larger
> dev_t is needed, no matter how we do it.

Yes, nomis.i-connect.net for example is one machine that is operating
almost at this limit.

> We need to somehow agree on how
> large this needs to be, and *then* we need to fix the filesystems so that
> they store a dev_t that is this large.

My vote goes for 64+bits

Mike

Michael Neuffer i-Connect.Net, a Division of iConnect Corp.
mike@i-Connect.Net 13455 SW Allen Blvd., Suite 140
503.677.2900 Beaverton, OR 97008

======================================================================
"Linux was made by foreign terrorists to take money from true US
companies like Microsoft." -- Some AOL'er.
======================================================================

\
 
 \ /
  Last update: 2005-03-22 13:37    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site