CX3-20 Requested Bypass Disks

can you slide one drive out and then back in ..does its status change ?

No, but something more interesting is happening.  I removed the cables to the LCC's, and powered up the DAE alone. The ID is 1, which is correct, and all the drives blink amber once long and 3 short, and then eventually go solid amber.

I re-ID's the LCC's to 2, and the process repeats with the same results.

Yeah I was thinking the same thing. The chance of 15 drives going bad in shipping is pretty rare.

Guess i'll reach back out to the seller.

interesting, part number is valid but i wonder if somebody re-formatted those drives and now FLARE cannot recognize them.

Hi Mnemonic,

I understand you added a new DAE to your running clariion. In that case my query is 'was there any error/ warning while adding the DAE/ disks (in NST) to the existing configuration '.

Can you try inserting one of these drive to the existing DAEs (If any free slot is available ),and check the status.

Vipin V.K

Three replies above

A few thoughts when it comes to troubleshooting this (assuming of course that there aren't any actual hardware failures or as suggested above, inappropriate reinitialization of the drives)

1) Firstly is the disk supported for the model of array

2) The minimum FLARE code required met

3) Verifying that the DAE model supports the disk

4) Making sure the disk is rated for the back-end bus speed set on the arbitrated loop

To answer the bullet points above, always look to the "Disk and FLARE OE Matrix" guides for the model of the array.  Therefore, in reference to your CX3-20, the guide you will want to refer to is:

"EMC® CX3 UltraScale Series Disk and FLARE OE Matrix"

This specific guide is available on PowerLink via the following breadcrumb trail:

Home > Support > Technical Documentation and Advisories > Hardware/Platforms Documentation > CLARiiON CX3 UltraScale Series Systems > CLARiiON CX3-20 > General Reference

Therefore, stepping through each of the points above:

1) Firstly is the disk supported for the model of array

YES, part number: 005048807 is listed

2) The minimum FLARE code required met

YES, you are running at least FLARE 24 as noted above

3) Verifying that the DAE model supports the disk

???, cross-referencing the prefix in the model number above with the table below, requires that the DAE be of type:

a) DAE2

b) DAE2P

c) DAE3P (DAE4P)

d) and not a DAE2-ATA

4) Making sure the disk is rated for the back-end bus speed set on the arbitrated loop

???, is the back-end bus set to 2Gbps which is the maximum rated speed for that drive (and not 4Gbps for instance)?

Ahh, just found this.  This explains my situation exactly.

https://powerlink.emc.com/nsepn/webapps/btg548664833igtcuup4826/public/CLARiiON_All_Models/FAQs/cX4_cx3_bus_speeds.htm

Replying from my other account but in regards to point 4, the drive is being marked amber before its even hooked up to a back end bus. 

I actually just picked up some brand new 300GB 2Gb/sec CX-2G disks that are supported as well and i have the same issue.

Do i have to set 2Gb/sec on the bus in navisphere before these are recognized?

Sorry for the late reply but it looks like you found the answer on your own.  The link appears to be unique to your PowerLink session at the time as it doesn't come up for me; however, I'll assume that it talks about, the fact, that the back-end bus must be set to the lowest common denominator.

Therefore, if you have even a single disk that is rated for only 2Gbps but all else (disk or DAE) happens to support both 2Gbps and 4Gbps, for instance, the bus must be configured to accommodate the slowest common speed.  With this in mind, if available, one advantage of having multiple back-end buses allow you to separate such hardware.  Whether using Navisphere or Unisphere, this is accomplished via the wizard by the same name:

"Backend Bus Speed Reset Wizard"

As you'll see when you initiate the wizard, you'll want to plan accordingly as noted in the screenshot below: