Dell Equallogic RAID failover test

recently I deployed a Dell Equallogic PS6510 for customer and everthing is setup properly.
now we are doing a UAT test and customer would like to have a test on disk failure and raid rebuild..

I make a call to Dell Tech Support and they advise me that not recommended to simply pull out the disk for testing, but we are supposed to set the disk offline first.

unfortunately, they can't provide me the guide or command line to set the disk offline.
so wonder anyone here ever try this before?

and after pull out the disk, we will need to re-use the disk as hotspare. into same enclosure.
I believe that this might not possible because the disk already mark as failed disk to the controller.
Who is Participating?
hell_angelEngineerAuthor Commented:
i had tested, pull out the disk, HS kicked in. raid rebuild. after that i re-insert the disk, it regconized as HS
noxchoGlobal Support CoordinatorCommented:
Here is a good link where you can get the info from:
Quote from there:
Installing and Removing Disks

You can install disks in an array and replace failed disks without disrupting I/O operations. See the PS Series Hardware Maintenance manual for your array model for instructions for physically installing and removing disks.

It is important to note the following:

    Do not remove a disk that is not a spare. Removing a disk that is an active member of a RAID set will be handled as a disk failure. If you must remove a disk (for example, to install it in another array), remove a spare from a member that has two working spares if possible. Replace the spare as soon as possible.
    Do not leave a disk slot empty. Always install a disk or a blank carrier.
    Disks are delicate. Always use electrostatic protection when handling disks. When not installed in an array, always protect a disk from electrostatic discharge.
    Wait for a disk to spin down. Once you press the disk handle release button, wait 30 seconds before removing the disk.
    Use a uniform disk size. Although you can use disks with different sizes in an array, the smallest disk in the array will determine how much space is available. For example, if an array has 250 GB and 400 GB disks, only 250 GB of space will be available on each disk.

When installing disks in an array, there is a two-minute delay between the time you insert a disk and the time the disk is automatically configured into a RAID set. This time interval enables you to install multiple disks and configure all the disks simultaneously, which is more efficient than installing a single disk, configuring it, and then repeating the process.

For example, when you install a disk in an array, a timer starts. If no other disks are installed, the disk is configured after a delay of 2 minutes. If you install another disk before 2 minutes have elapsed, the timer is restarted.

Notes: If you install a disk during an ongoing RAID reconstruction or verification operation, the new disk will not be configured until the operation completes.

If you remove a disk from one array and then install the disk in a different array, you must clear the disk label before the disk can be automatically configured. Click the foreign disk link in the Disk Status window. (See Table 3: Disk Status.)

After installing a disk, check the status to ensure that the disk is either online or a spare, as shown in the Member Disks window (Figure 9: Member Disks - PS5000 Array ).

As far as I know right click on disk in RAID GUI and selecting - set OFFLINE makes them offline.
hell_angelEngineerAuthor Commented:
i'm actually gone through the link you sent.
im looking for guide how to simulate a disk failure. customer required at acceptance test to simulate in case active disk failed and does the HS is possible auto kick in and rebuild.

the problem is, if I pull out a working disk, and HS does kicked in, how am I suppose to erase the metadata on the disk that I pulled out and continue use it as fresh disk..
Network Scalability - Handle Complex Environments

Monitor your entire network from a single platform. Free 30 Day Trial Now!

noxchoGlobal Support CoordinatorCommented:
The metadata is stored on the HDD, normally at start. Simply erase the disk  on different machine (connect it internally to a workstation for example) using some disk wiper tool so that system sees it as new not initialized disk. Then you can simulate "new" disk.
As for force "disk failure" simulation - I don't think it is possible.
When you put the disk back it should show as foreign and you can click the foreign disk link and clear the label. It's possible that hot-removed will be added to the SMART data on the disk when you pull it out so don't do it too much or it'll be reported as pre-failure alert.
hell_angelEngineerAuthor Commented:
the info provided is not quite relevant to my incident
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.