Sun Storage D1000/A1000

Latest response

I am not sure whether i can post the query here or not,

Recntly bought a used Sun Storage D1000 with for 4 Disks and scsi card Sym53c875.

I am trying to connect it to RHEL6 running on intel GGC 101 workstation,
does anything need to be done from OS end,

issue:

RHEL6 detected the disks in the OS when i try to fdisk its throughing error stating cannot read disk.
any settings need to be donw from iscsi end..

Thanks in Advance
Niranjan.

Responses

Can you post the actual output from the command(s) you ran?

I am guessing here, but I believe you are probably having issues with the disk labels on the disks. If there were currently used by a Sun Sparc system they have a Solaris Disk Label (SMI). I did not work much with Solaris on Intel, but I believe they also would use SMI or EFI. In either case, I believe you just need to relabel the disk(s) with a more RHEL friendly label.

You can put a new DOS or GPT label on them using parted.

parted -s /dev/sdb mklabel msdos 
parted /dev/sdb print

You can also create a label and partition at the same time

parted -s /dev/sdb mklabel msdos mkpart primary ext3 0 100% set 1 lvm on

Solaris uses a significantly different format for disk layout whereas slice 2 is the entire disk.
Also - I'm not sure how this would work with the A1000 (those things are as old as my entire career potentially ;-) I thought with the A1000 the array commands were sent through the SCSI devices, using the RAID Manager software. I imagine once the LUNs have been created, you can do whatever you want with them.

Thank you very much for replying..

I will try and post the update..

Thanks In Advance
NIranjan

Can i post the queries on this portal...? because these are not related to any of the case..

Thanks
Niranjan

The Customer Portal is an open forum where you can ask whatever you like. ;-) We are just volunteers that enjoy helping other people. If your issue is very important, you should actually open a case - sometimes we are not able to respond very quickly.

okay, thank you for helping hand..

Did the suggestions help? Or - do you need additional assitance?

I tried but no luck.. pasted the command output, /dev/sdb/ /dev/sdc /dev/sdc disks are from my Sun D1000 storage array, but not able to format the disks.

[root@server1 ~]# cat /proc/scsi/scsi | grep -i ibm
Vendor: IBM Model: DDYST1835SUN18G Rev: S96H
Vendor: IBM Model: DDYST1835SUN18G Rev: S96H

[root@server1 ~]# ls -l /sys/block/*/device
lrwxrwxrwx. 1 root root 0 Sep 28 02:24 /sys/block/sda/device -> ../../../5:0:0:0
lrwxrwxrwx. 1 root root 0 Sep 28 02:24 /sys/block/sdb/device -> ../../../7:0:0:0
lrwxrwxrwx. 1 root root 0 Sep 28 02:24 /sys/block/sdc/device -> ../../../7:0:1:0
lrwxrwxrwx. 1 root root 0 Sep 28 02:24 /sys/block/sdd/device -> ../../../7:0:3:0
lrwxrwxrwx. 1 root root 0 Sep 28 02:24 /sys/block/sr0/device -> ../../../4:0:1:0

[root@server1 ~]# parted -s /dev/sdb mklabel msdos

[root@server1 ~]# parted /dev/sdb print

[root@server1 ~]# fdisk /dev/sdb

Unable to read /dev/sdb

[root@server1 ~]# fdisk -l /dev/sdb
[root@]# parted -s /dev/sdb mklabel msdos mkpart primary ext3 0 100% set 1 lvm on
[root@server1 ~]# fdisk /dev/sdb

Unable to read /dev/sdb

Sorry the previous suggestions were not helpful. At this point I would "cheat" and install gparted to manage the devices.

Also - I would try the following:

dd if=/dev/zero of=/dev/sdb bs=512 count=10240
parted -s /dev/sdb mklabel msdos mkpart primary ext3 0 100% set 1 lvm on
partprobe /dev/sdb
parted /dev/sdb print

The dd command should just wipe the beginning of the drive (wiping the partition table in the process).

You could also try:

dd if=/dev/sdb of=/dev/null bs=512 count=1024

To see if it truly is unable to read from /dev/sdb (which will output an I/O error of some sort).

Gparted provides a graphical interface to manage disks that performs many tasks "behind the scenes". You can see all the steps and commands at the bottom of the application though.

If you are still stuck, I'm at a bit of a loss. It's been a long time since I have played with SCSI disk trays, but I believe if you had the incorrect termination, you would not see the drives at all. But, you may want to confirm the termination, as well.

Thank you for your reply, i was out of city for couple of days,
I tried as you said, however no luck, termination is placed properly

[root@server1 log]# dd if=/dev/zero of=/dev/sdb bs=512 count=10240
dd: writing `/dev/sdb': No space left on device
1+0 records in
0+0 records out
0 bytes (0 B) copied, 0.000402471 s, 0.0 kB/s
[root@server1 log]# parted -s /dev/sdb mklabel msdos mkpart primary ext3 0 100% set 1 lvm on
[root@server1 log]# parted /dev/sdb print
[root@server1 log]# partprobe /dev/sdb
[root@server1 log]# parted /dev/sdb print

What happens if you just type "parted -l" without specifying any disks?

Do they still make the A1000/D1000 arrays? I last used those more than 10 years ago on Solaris 2.6 (SunOS 5.6) on SPARC.

I will check the parted -l tonight and let you know, they are not making the A1000/D1000 arrays, i bought a used Array, to get hands on experience on Veritas and RedHat cluster..

Close

Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.