SCRUBBING ARRAY

Question

Recently I came across scrubbing requirement for one of my client.

Technically speaking..

Once a clariion LUN unbinded & binded with NEW LUN size, Do you still think that EMC or any hackers can retrive data from clariion ?

If NO, then why EMC has a standard for scrubbing ? charging for it ? what involved here?

what scrubbing does from EMC?

customer's solution from us today..

we are using freeware scrubbing tool for customer which i am still not answered why we are doing this activities?.. 1) unbind all existing LUNs and delete all servers connections 2) bind NEW BIG (R1)LUN size and assign to scrubbing server 3) start freeware tool to scrubb data on claiion ..

can you help me understand about it ?

dynamox · Answer

is this a 'shared' array between multiple customers ? Not sure why somebody would do that unless you are getting rid of the array and need to make sure no data is left behind (see EMC certified erasure service).

ARSNMJ · Answer

This is dedicated storage for a customer.

dynamox · Answer

take a look at Glen's comment regarding 'background zeroing' https://community.emc.com/message/414156

kelleg · Answer

That's only one pass - disk zeroing - scrubbing is more like overwriting many, many times to ensure that you can't take the physical patters to someone and have them read the data - you can do this

glen

dynamox · Answer

but EMC can't scrub on per LUN bases, have to employ host based utilities ?

Pasta_Fazole · Answer

I agree with kelleg's comment that a single pass with zeros will not be sufficient to prevent the recovery of the data.

http://en.wikipedia.org/wiki/Data_remanence1

Is this the best method:

Attach a server running http://www.killdisk.com/ or some other utility to the luns that must be erased.

Is there a tool native to the Clariion that will do this?

Are there better utilities out there that will do this?

nandas · Answer

Destroying the existing LUN and then binding of new LUN will ensure that the old data is completely gone as Flare re-initializes the space. Please refer to the EMC Primus solution emc87647.

However the best way to destroy the data securely is to available the data eraser service available from EMC - which is called "Certified Data Erasure for CLARiiON" - Model Number: PS-BAS-CXDE. Please contact your local EMC team for the same. You may also explore more on this on Powerlink per the following navigation -

Home > Services > SPD TS Kits and Service Offering Index > TS Kits > Customer Support Services > Data Erasure and Disk Retention> Certified Data Erasure for CLARiiON

I have also seen customers to take out the disks and re-arrange their physical location within same type of DAE - but please do not alter the Flare Disks (Disk 0 to 5 on the first enclosure) as it may cause the array in unusable condition.

Hope this helps,

Sandip

hersh1 · Answer

Not a free one nor helpful if you don't use cisco switches. But I believe the latest version of Cisco SANOS (5.X) you can attach storage to it will perform DoD certified wipes (or other algorithms) of the disks. Have to purchase a specific license key and have a multiservice module installed in the switch.

Kavurisravan · Answer

We are seeing a difference between storage-side and host-side volume used capacity on PowerMax 950F. These are very old thin volumes.
Will scrubbing help reclaim unused space? Are host-side UNMAP/TRIM settings mandatory for effective data reclaim?
Looking for best practices to validate backend space reclamation.

DELL-Sam L · Answer

Hello Kavurisravan,

Here are the steps that you want to do to start space reclamation on the PowerMax 950F.

1. Select the Storage System: Access the management interface for your PowerMax 950F.

2. Navigate to Volumes: Go to STORAGE > Volumes.

3. Choose Volume Type: Select the appropriate tab for the volume type you wish to reclaim.

4. Start Space Reclamation:

· Select one or more volumes.

· Click on Reclaim to open the Start Allocate/Free/Reclaim dialog box.

· Choose Start Allocate/Free/ and then select Reclaim Volumes to reclaim unwritten tracks and those written completely with zeros. This option is available on systems running Enginuity 5876 or higher, or HYPERMAX OS 5977 or higher.

· Optionally, you can select Reclaim persistent capacity to reclaim tracks that are unwritten or zero-based, even if marked as persistent.

To stop space reclamation, select one or more volumes, click Reclaim, and follow the prompts to stop the operation.

For more detailed information on thin pools and thin provisioning concepts, refer to the relevant procedures in the PowerMax documentation.

Here is also a link to Unisphere for PowerMax Version 9.0.0 Online Help guide which has some additional assistance for thin volumes. https://dl.dell.com/content/docu88874_unisphere-for-powermax-9-0-0-online-help.pdf?language=en-us

CLARiiON

SCRUBBING ARRAY

Was this post helpful?