On Sat, Mar 01, 2008 at 02:31:30PM +1100, Hamish Moffatt wrote:
Doesn't udev assign constant sdN/hdN identifiers? It appears to look up
the drive serial number and attempt to do this.
Certainly didn't do it in the past.
Also, if you run LVM or software RAID then you get static identifiers
On Sat, Mar 01, 2008 at 11:46:54AM -0500, Damon L. Chesser wrote:
How do you determine the UUID?
blkid /dev/devicename
Swap only has a UUID if you use v1 format. v0 didn't have one.
--
Len Sorensen
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble?
Hamish
On Sat, 2008-03-01 at 14:31 +1100, Hamish Moffatt wrote:
On Fri, Feb 29, 2008 at 12:01:49PM -0500, Lennart Sorensen wrote:
UUIDs work for me. Driver modules load in parallel and nothing will
ensure consistent device names. A unique identifier for the filesystem
No it doesn't, hence
On Sat, Mar 01, 2008 at 11:00:01AM +, Steve Dobson wrote:
On Sat, 2008-03-01 at 14:31 +1100, Hamish Moffatt wrote:
Doesn't udev assign constant sdN/hdN identifiers? It appears to look up
the drive serial number and attempt to do this.
From what I've observed hdN is consistant. hda is
Lennart Sorensen wrote:
On Fri, Feb 29, 2008 at 01:54:25PM -0500, Douglas A. Tutty wrote:
Is there any advantage of using UUID (which are very long) and a
filesystem label (e.g. LABEL=root) which is shorter and more intuitive?
If you ever have to deal with a disk from another system
Damon L. Chesser wrote:
How do you determine the UUID?
If the system is in a working and known state, you can simply do
something like
$ ll /dev/disk/by-uuid/
total 0
lrwxrwxrwx 1 root 10 2008-03-01 04:31 3A5C47515C4706DB - ../../hda1
lrwxrwxrwx 1 root 10 2008-03-01 04:31 9819-26DD -
Eduardo M KALINOWSKI wrote:
Damon L. Chesser wrote:
How do you determine the UUID?
If the system is in a working and known state, you can simply do
something like
$ ll /dev/disk/by-uuid/
total 0
lrwxrwxrwx 1 root 10 2008-03-01 04:31 3A5C47515C4706DB - ../../hda1
lrwxrwxrwx 1 root 10
Damon
On Sat, 2008-03-01 at 11:46 -0500, Damon L. Chesser wrote:
Lennart Sorensen wrote:
How do you determine the UUID?
I use blkid(8).
For example on workstation with many types of disk (inc s/w RAID):
# blkid
/dev/sdb1: UUID=89cc5b7e-049f-4e41-8e13-0017180ae87a TYPE=xfs
/dev/sda5:
Hamish
On Sat, 2008-03-01 at 22:31 +1100, Hamish Moffatt wrote:
On Sat, Mar 01, 2008 at 11:00:01AM +, Steve Dobson wrote:
On Sat, 2008-03-01 at 14:31 +1100, Hamish Moffatt wrote:
Doesn't udev assign constant sdN/hdN identifiers? It appears to look up
the drive serial number and
Hi All
This isn't strictly an AMD64 question as the problem would probably also
appear on i386 systems too. However, you lot are way too
knowledgeable... :-)
I have a system with two RAID controllers (3wave) installed and a
separate SATA disk for the OS. I pulled the RAID disk and installed
On Fri, Feb 29, 2008 at 12:20:24PM +, Steve Dobson wrote:
As the kernel is assigning disk IDs dynamically is there some way of
forcing given devices to have given SCSI addresses. Our plan is to have
the 16 disks configured into 4 arrays of four disks each. There maybe
times when a disk
On Fri, Feb 29, 2008 at 04:51:26PM +, Steve Dobson wrote:
I am leaning towards the UUID method to solve this. Labels after all
are not guaranteed to remain unique, the UUID will so long as the
hardware remains constant.
However, I wasn't sure this was the best method and though I would
Steffan
On Fri, 2008-02-29 at 15:09 +0100, Steffen Grunewald wrote:
On Fri, Feb 29, 2008 at 12:20:24PM +, Steve Dobson wrote:
As the kernel is assigning disk IDs dynamically is there some way of
forcing given devices to have given SCSI addresses. Our plan is to have
the 16 disks
Len
On Fri, 2008-02-29 at 12:01 -0500, Lennart Sorensen wrote:
On Fri, Feb 29, 2008 at 04:51:26PM +, Steve Dobson wrote:
I am leaning towards the UUID method to solve this. Labels after all
are not guaranteed to remain unique, the UUID will so long as the
hardware remains constant.
On Fri, Feb 29, 2008 at 05:58:19PM +, Steve Dobson wrote:
It's good to know that UUIDs work well for someone at least.
I've tried it using the /home partition on my laptop and it worked like
a dream. Next stage is to change my workstation over test and then
re-order the disks.
Are
Len
On Fri, 2008-02-29 at 13:04 -0500, Lennart Sorensen wrote:
On Fri, Feb 29, 2008 at 05:58:19PM +, Steve Dobson wrote:
Are there any pointers to grub config changes that need to be made?
I have this:
# kopt=root=UUID=35963d32-f15e-497e-859a-ed1cb366b0f3 ro
Then update-grub will
On Fri, Feb 29, 2008 at 01:54:25PM -0500, Douglas A. Tutty wrote:
Is there any advantage of using UUID (which are very long) and a
filesystem label (e.g. LABEL=root) which is shorter and more intuitive?
If you ever have to deal with a disk from another system using labels
you can get into
On Fri, Feb 29, 2008 at 01:04:56PM -0500, Lennart Sorensen wrote:
I have this:
# kopt=root=UUID=35963d32-f15e-497e-859a-ed1cb366b0f3 ro
Then update-grub will make all the entries have that and it just worked.
Is there any advantage of using UUID (which are very long) and a
filesystem
On Fri, Feb 29, 2008 at 12:01:49PM -0500, Lennart Sorensen wrote:
UUIDs work for me. Driver modules load in parallel and nothing will
ensure consistent device names. A unique identifier for the filesystem
Doesn't udev assign constant sdN/hdN identifiers? It appears to look up
the drive serial
19 matches
Mail list logo
