Hi everybody

I installed RedHat Advance Server 3 Update 3 with kernel 2.4.21-40.ELsmp installed on a HP Lxr8500 Server...

Additionally a connect to it a Disk Array VA 7400 through FiberChannel and i can create LUNs dinamically.

When I reboot the server Linux shows me the device file of this new disk: For example: /dev/sda, /dev/sdb, ..... but i need to reboot... and what happened if i want to create 60 LUNs??? so this is the question:

how can i detect those disks (Luns) without rebooting the server????

Is there an utility that detect and make the device files on the fly in Redhat ?? or i need to install an special free software for this???

Please Help me...

Hello everyone!!!.. I just discovered and I'm going to share with you to your knowledge....

My fiberchannel target is on the scsi2... so the first thing i did was see /proc/scsi/scsi this is the output...

Attached devices:
Host: scsi2 Channel: 00 Id: 00 Lun: 00
Vendor: HP Model: A6189A Rev: HP19
Type: Direct-Access ANSI SCSI revision: 03

Host: scsi4 Channel: 00 Id: 00 Lun: 00
Vendor: MegaRAID Model: LD0 RAID1 34731R Rev: H
Type: Direct-Access ANSI SCSI revision: 02


- As you see I have one Lun created and is 10GB of size.. So I create a new Luns of 2GB in the VirtualArray...

- After create the Lun.. I run this command:

echo "scsi remove-single-device 2 0 0 1" > /proc/scsi/scsi

This instruction tells the Operationg System to add an SCSI dinamically but instead of adding an Scsi drive i add a LUN... only what i need is Controller (2), Host ID (0), SCSI ID (0) and LUN ID (1)..

The Lun ID is 1 because is the new Lun that I'm going to add...

- I run dmesg to see the results. This is the output:

scsi singledevice 2 0 0 1
blk: queue f4220818, I/O limit 4294967295Mb (mask 0xffffffffffffffff)
Vendor: HP Model: A6189A Rev: HP19
Type: Direct-Access ANSI SCSI revision: 03
blk: queue f47dd018, I/O limit 4294967295Mb (mask 0xffffffffffffffff)
Attached scsi disk sdd at scsi2, channel 0, id 0, lun 1
SCSI device sdd: 4194304 512-byte hdwr sectors (2147 MB)
sdd: unknown partition table

As you see the new device file is /dev/sdd, As you observe is a 2GB disk.. and run cat/proc/scsi/scsi and see it

Attached devices:
Host: scsi2 Channel: 00 Id: 00 Lun: 00
Vendor: HP Model: A6189A Rev: HP19
Type: Direct-Access ANSI SCSI revision: 03
Host: scsi2 Channel: 00 Id: 00 Lun: 01
Vendor: HP Model: A6189A Rev: HP19
Type: Direct-Access ANSI SCSI revision: 03
Host: scsi4 Channel: 00 Id: 00 Lun: 00
Vendor: MegaRAID Model: LD0 RAID1 34731R Rev: H
Type: Direct-Access ANSI SCSI revision: 02


- And to finish you can do anything with that disk... and create Many disks without rebooting..

Greetings...

Lenux78

sorry.. instead of echo "scsi remove-single-device 2 0 0 1" > /proc/scsi/scsi is

echo "scsi add-single-device 2 0 0 1" > /proc/scsi/scsi

Typically we run this command:

echo "scsi scan-new-devices" > /proc/scsi/scsi

Hi tells!!! thanks for the suggestions... I tried to do it but for some reasons that I don't know i didn't work.. it doesn't insert any new device...

In fact, i was searching about echo "scan-new-devices" and found this link http://kbase.redhat.com/faq/FAQ_46_4011.shtm.

why do you think is that???

Greetings....
Lenux78

You should download scsiadd and scsidev. Both those are great programs for recognizing and display information (ie. serial numbers, etc) of LUNS.

-twantrd

Hi twantrd!!!..

I download scsidev and probe it.. but when i do an scsidev -v to see want he brings.. it throws the following message

#scsidev -v

scsidev 2.36 2005-08-16
/proc/scsi/scsi extensions not found. Fall back to scanning.
Building list for sg (char dev major 015)
Found /dev/scsi/sgh0-20c0i0l0 (Type 00) on LSI Logic MegaRAID H01.08 254 commands 15 targs 4 chans 7 luns
Findscsidisk: 0
INQUIRY failed for /dev/scsi/sdh0-20c0i0l0 (0-0/008:00000)!
scsidev: Strange: Could not get info from sdh0-20c0i0l0
Found /dev/scsi/sgh3-0c0i0l0 (Type 00) on Emulex LightPulse LP8000 1 Gigabit PCI Fibre Channel Adapter on PCI bus 01 device 20 irq 50
Findscsidisk: 1
INQUIRY failed for /dev/scsi/sdh3-0c0i0l0 (0-0/008:00010)!
scsidev: Strange: Could not get info from sdh3-0c0i0l0
scsidev: What's going on? Dev sdh3-0c0i0l0 is different from sgh3-0c0i0l0


that sounds like the command cannot get the dev type, and when i do: ll /dev/scsi/* this is the output:

brw-rw---- 1 root disk 8, 0 Jul 21 09:16 /dev/scsi/sdh0-20c0i0l0
brw-rw---- 1 root disk 8, 1 Jul 21 09:16 /dev/scsi/sdh0-20c0i0l0p1
brw-rw---- 1 root disk 8, 2 Jul 21 09:16 /dev/scsi/sdh0-20c0i0l0p2
crw-rw---- 1 root disk 21, 0 Jul 21 09:16 /dev/scsi/sgh0-20c0i0l0
crw-rw---- 1 root disk 21, 1 Jul 21 08:19 /dev/scsi/sgh3-0c0i0l0

As you see my operating system is in /dev/scsi/sdh0-20c0i0l0 that's ok... but my Lun is on /dev/scsi/sgh3-0c0i0l0 but doesn't create the block device and i can't use it...

maybe you can help me to use this tool... because i didn't use it before but it seems that i could use it and resolve many problems...

Greetings..


Lenux

Try using scsidev -s. It will try to print information of the LUNS if available.

To create the block devices, use scsidev -l. Sometimes, I need to delete all the /dev/scsi entries first, scsidev -f, and then scsidev -l will be issued afterwards.

-twantrd

The first option below worked for me
From the release notes:
@redhat.com /docs/manuals/enterprise/RHEL-4-Manual/release-notes/as-s390/RELEASE-NOTES-U3-s390-en.html


Support for fabric rediscovery through sysfs is now available in Red Hat Enterprise Linux 4 Update 3. For the Qlogic (qla2xxx) and Emulex (lpfc) Fibre Channel HBA drivers, run the following commands to perform redisocvery and rescan for new storage:


echo "1" > /sys/class/fc_host/host2/issue_lip
echo "- - -" > /sys/class/scsi_host/host2/scan

Where XYZ is the scsi host number of your HBA.

#

In some Fibre Channel configurations, an operating system will automatically detect the addition of a new target device while the system is running. In other configurations, it is necessary to issue the following command to detect a new target device:


echo 1 > /sys/class/fc_host/hostn/issue_lip

Where hostn corresponds to the appropriate adapter.


When a new Logical Unit is created on a particular target, a command such as the following is required to detect and configure it:


echo "b t l" > /sys/class/scsi_host/hostn/scan

Where b is the bus, t is the target, and l is the LUN to be scanned on hostn.

Wild cards can also be used, such as the following example:


echo "- - -" > /sys/class/scsi_host/host2/scan

Note that the device name (such as /dev/sdb) that is assigned when a device is added dynamically while the system is running may be different from the device name assigned the next time the system boots.

hi,
why i submit comand: echo "scsi scan-new-devices" > /proc/scsi/scsi, raised error "bash: echo: write error: Invalid argument"
Have other the way is possible about problem?
Thanks.

well, as far as I have read, that means you do not have that value in /proc/scsi/scsi, so that value is invalid.

Interesting links: http://bugzilla.kernel.org/show_bug.cgi?id=9685
and http://groups.google.com/group/linux...6c778bc8fab03b

Have you tried scsiadd and scsidev as suggested earlier?

And do you use RHEL AS 3 or some other version?

hi,
I found the other way to solve: here
Thanks so much!