Re: Newbie scsi question

2006-05-04 Thread talon
Karthik Subramanian wrote: > Hi Hans, > >> udev is not a fix to this, it is the cause. You actually can use udev >> to get stable device nodes. Most distros don't care, though. The linux >> kernel provides a stable addressing in sysfs. udev should be configured >> to use this and not just enumera

Re: Newbie scsi question

2006-05-04 Thread Karthik Subramanian
Hi Hans, udev is not a fix to this, it is the cause. You actually can use udev to get stable device nodes. Most distros don't care, though. The linux kernel provides a stable addressing in sysfs. udev should be configured to use this and not just enumerate serially. So the problem isn't there "i

Re: Newbie scsi question

2006-05-04 Thread Hans-Werner Hilse
Hi, very off-topic (Linux specific), but... On Wed, 3 May 2006 16:09:42 +0530 "Karthik Subramanian" <[EMAIL PROTECTED]> wrote: > On linux, the disks from the JBOD are recognised as /dev/sd[a-n]. The > hitch is that the device naming is not consistent across reboots; the > first time around,

Re: Newbie scsi question

2006-05-03 Thread Matthew Dillon
: :On 5/3/06, Matthew Dillon <[EMAIL PROTECTED]> wrote: :... :> It would, but probably not in the fstab file. What we really :> need is a devfs based disk ID / serial-number / volume label :> type of thing. Actually what we really need to do is rewrite :> devfs entirely. : :What

Re: Newbie scsi question

2006-05-03 Thread Chuck Tuffli
On 5/3/06, Matthew Dillon <[EMAIL PROTECTED]> wrote: ... It would, but probably not in the fstab file. What we really need is a devfs based disk ID / serial-number / volume label type of thing. Actually what we really need to do is rewrite devfs entirely. What about extending

Re: Newbie scsi question

2006-05-03 Thread Matthew Dillon
:More on the same topic! : :Here's an excerpt from the linux fstab(5) manpage: : :--- snip --- : :Instead of giving the device explicitly, one may indicate the (ext2 or :xfs) filesystem that is to be mounted by its UUID or volume label (cf. :e2label(8) or xfs_admin(8)), writing LABEL= or UUID=, :e

Re: Newbie scsi question

2006-05-03 Thread Karthik Subramanian
On 5/3/06, Chuck Tuffli <[EMAIL PROTECTED]> wrote: Unfortunately, DragonFly (and most of the other *BSDs I think) doesn't have a good answer for persistent binding. You can setup the kernel to "wire down" a particular bus/target/lun to a specific device, but I don't think this would work for SAN

Re: Newbie scsi question

2006-05-03 Thread Chuck Tuffli
On 5/3/06, Karthik Subramanian <[EMAIL PROTECTED]> wrote: ... Here's the problem: Consider a typical SAN setup - an FC switch with a couple of JBOD's and a couple of machines hooked up to the switch. On linux, the disks from the JBOD are recognised as /dev/sd[a-n]. T

Re: Newbie scsi question

2006-05-03 Thread Justin C. Sherrill
On Wed, May 3, 2006 6:39 am, Karthik Subramanian wrote: > Consider a typical SAN setup - an FC switch with a couple of JBOD's > and a couple of machines hooked up to the switch. > > On linux, the disks from the JBOD are recognised as /dev/sd[a-n]. The > hitch is that the device naming is not consi

Newbie scsi question

2006-05-03 Thread Karthik Subramanian
Hi Folks, I faced a small problem with Linux recently, saw how Solaris solved the same problem, and was wondering what DragonFly did. Disclaimer: I haven't worked with *BSD too much; so this could potentially be a question best answered by an RTFM. If this is the case, please point me to the ri