Clariion drives on a new rp3440 HP-UX

[Randy Breitfleder]

I am configuring a new HP-UX server running 11.11 with SAN attached storage. The ioscan -fnC disk displays the san disks as RAID 5 on my other 11.11 HP-UX servers the disk display as CX3-80WDR5. Both server are at the same patch level and the storage group said the disk are configured the same. Does anybody know what would cause these disk to be displayed as RAID 5 instead of CX3-80WDR5?

New HP-UX server
disk 9 0/4/1/0.1.139.255.0.0.1 sdisk CLAIMED DEVICE DGC RAID 5

Other HP-UX server
disk 38 1/0/6/1/0.1.244.0.0.1.0 sdisk CLAIMED DEVICE DGC CX3-80WDR5

Thanks, Randy

[Torsten.]

Hi Randy,

I'm not familiar with the clariions, but this is coming from the array side.

Ask your storage people to double check all the presentation settings again.

[sujit kumar singh]

This has got something to do with the Emulation of the LUN and the presentation of that to the server from the storage side.

regards
sujit

[TTr]

The difference in your two servers' hardware paths are in two places, the "139 vs. 244" and "255 vs. 0". The first, 139 and 244 are the domain IDs of each of the fiber switches that the servers are attached to. This does not affect the LUNs in any way. The second, 255 and 0, are the ports on the array side where the LUNs exist. So it depends where the LUNs are in the array, they can be coming from different EMC frames, cabinets, EMC disk controllers or different types of spindles as well.

[Steven E. Protter]

Shalom Randy,

The Clarion is an EMC NAS device. You are connected to it via a fiber/fabric network.

The difference in name is due to the configuration of the LUN, or the Clarion controller, which if I recall is a Linux system.

SEP

[Don Mallory]

Hi Randy,

Since you are connected to a CX3-80, it's clear that you are fabric attached and that the path to your siwtch is on Domain ID 1.

Your Port number is 139 on the new server and port 244 on the other, which suggests that you have zoned this new host's HBAs to see different ports on the Storage Processors (SP's) than your old host.

Before you get too far, access the EMC eLab Navigator via Powerlink (https://powerlink.emc.com) and verify that the hosts both match the existing support matrix for drivers, firmware, h/w, etc. Specifically, verify the version of FibrChanl-00 and -01 depending on which HBAs have installed. Occasionally the H/W enablement patches will show different values.

You can also verify that you are looking at the LUNS correctly via EMC's inq tool. You can get that from ftp.emc.com, it's a free tool.

Another possible reason I have seen is that if zone the host to see the array, then add the host connectivity info in Navisphere (or populate the info using NaviAgents), before you add the host to the storage group, if will show some funny LUNs (LUNZ).

It goes away once PowerPath is properly configured, the LUNs and host are in the Storage group properly and everything is logged in, and you reboot, but inq still shows odd entries.

Here's an example output:

host1 # inq -no_dots
Inquiry utility, Version V7.3-771 (Rev 0.0) (SIL Version V6.3.0.0 (Edit Level 771)
Copyright (C) by EMC Corporation, all rights reserved.
For help type inq -h.



-----------------------------------------------------------------------------
DEVICE :VEND :PROD :REV :SER NUM :CAP(kb)
-----------------------------------------------------------------------------
/dev/rdsk/c0t0d0 :TEAC :DV-28E-C :C.4D : : -----
/dev/rdsk/c2t0d0 :HP 36.4G:ST336754LC :HPC2 :3KQ048QM : 35566480
/dev/rdsk/c2t1d0 :HP 36.4G:ST336754LC :HPC2 :3KQ048KF : 35566480
/dev/rdsk/c58t0d0 :DGC :CX3-80 :HP03 :Ch2 CONT : FAILED
/dev/rdsk/c59t0d0 :DGC :CX3-80 :HP03 :Ch2 CONT : FAILED
/dev/rdsk/c60t0d0 :DGC :CX3-80 :HP03 :Ch2 CONT : FAILED
/dev/rdsk/c61t0d0 :DGC :CX3-80 :HP03 :Ch2 CONT : FAILED
/dev/rdsk/c62t0d0 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 10485760
/dev/rdsk/c62t0d1 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 26214400
/dev/rdsk/c62t0d2 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 78643200
/dev/rdsk/c62t0d3 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 26214400
/dev/rdsk/c62t0d4 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 26214400
/dev/rdsk/c62t0d5 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 10485760
/dev/rdsk/c62t0d6 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 10485760
/dev/rdsk/c62t0d7 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 52428800
/dev/rdsk/c62t1d0 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 52428800
/dev/rdsk/c62t1d1 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 52428800
/dev/rdsk/c62t1d2 :DGC :CX3-80WDR5 :HP03 :Ch2 CONT : 52428800


Note the 4 failed paths at the top.

Hope that helps.

Don

[Don Mallory]

Just a note for Scott.

The CLARiiON or CX3-xx, now called the CX4-xx and previously was the CXx00 series are SAN arrays not NAS devices. The newest ones are capable of iSCSI as well.

The EMC Celerra or NSxxx is the NAS. The xxx is now a number 120, 240, 480, 960 which denotes the max number of disks it can handle. The older models were CX3-10, 20, 40, 80 respectively. There was no CX3-160 (960 drives)

NS is "Network Storage". Ironically, they have a CLARiiON back-end, as do the EMC EDL virtual tape libraries. The Celerra has a usable back-end FC array, the EDL array is entirely captive, and is also ironically running the FalconStor VTL software.

The OS running on the CX Storage Processors is known as "FLARE", and is a Windows XP embedded OS.
The OS on a Celerra is known as "DART" and is Linux based.
The OS on the EDL engine is also Suse Linux 9U3 at this time.

Hope that helps clarify things a little.