PowerScale A3000, A300, H7000,

H700 3.5-inch Drive Replacement

Guide
Replace a drive
This guide applies to the following Archive or Hybrid node hardware
configurations that contain 3.5" drives:

 A3000
 A300
 H7000
 H700

 Replacing a drive
 Gather logs
 Check for the latest drive support package
 Smartfail the drive
 Install the drive replacement
 Gather logs
 Returning a failed part
 Update the Install Base
 Where to get help
 You can replace a failed drive in the field.
 CAUTION Do not remove a drive sled while the Do Not Remove LED is
illuminated on the sled. Removing a sled with that LED illuminated
might interrupt operations and lead to data loss.
 Never remove more than one drive sled at a time. Removing multiple
drive sleds at once may lower the protection level of the cluster, put
data at risk, and lead to the interruption of client workflows.
 NOTE When you insert a new drive, OneFS automatically formats the
drive and adds it to the cluster. If ARR (Automatic Replacement
Recognition) has been disabled, you must add the drive manually with
the following command:
 isi devices drive add <bay-number> --node-lnn <node-number>

Gather logs
Before you begin any maintenance on a cluster, start gathering cluster
logs.

Cluster logs provide snapshots of the cluster, which you can review to ensure that
maintenance is successful.
# isi diagnostics gather is a utility that collects configuration and log
information from a cluster and automatically uploads it to Dell for
processing. This command runs in the background so that interruption of
the user session doesn't cause problems for the gather process.

If your environment has a firewall in place that blocks the uploading of the
log information, contact the Support Engineer assisting on the case for a
temporary FTP link to which to upload the logs.

1. Open a secure shell (SSH) connection to any node in the cluster and log in.
2. Gather cluster logs by running the following command:

# isi diagnostics gather start

3. View the gather status with the following command:

# isi diagnostics gather status

Check for the latest drive support package

The drive support package updates drive configuration information on the
node.

The drive support package contains the latest firmware versions for
PowerScale qualified drives.

NOTE Installation instructions can be found in the release notes.

1. Before you install a new drive, go to the Support page.

2. Check your version of the firmware:

isi_dsp_install --latest

3. Check for the latest version of the drive support package.

4. If necessary, download the package and update the firmware.

Smartfail the drive

Before removing a drive from a node, ensure that the drive has been
smartfailed.

NOTE A drive smartfail can take hours or days, depending on drive size, amount of
data on the drive, and overall cluster utilization.

1. Open a secure shell (SSH) connection to the node that contains the drive you
want to remove. Type the command:

ssh <cluster-name>-<node-number>
 2. Confirm the node number and bay number of the drive that you want to
remove. Type the following command:

isi devices drive list

NOTE Determine the node number and bay number of the drive to be
removed. If the drive has already been smartfailed automatically, skip the
next step. A smartfailed drive displays a status of REPLACE or, if the node
was restarted after the smartfail, a status of EMPTY. A smartfailed SED drive
might display a status of ERASE. This status also indicates that the drive is
ready to be replaced.

3. If the drive you intend to replace was not already smartfailed, then using the
node number and bay number of the drive, smartfail the drive by typing the
following command:

isi devices drive smartfail <bay-number>--node-lnn<node-number>

If the drive does not respond to the smartfail command, do not

remove it from the node. Contact Dell Technologies Technical
Support.

4. Confirm that the drive was smartfailed by repeating step 2.

CAUTION Do not replace the drive until you have confirmed that the smartfail
process is complete.

Install the drive replacement

Remove the failed drive and install the replacement drive.

The following drive map indicates the locations of the drives in the drive
sleds.
  Remove the front bezel
 Remove the drive sled containing the failed drive
 Remove the failed drive
 Install the new drive
 Install the drive sled
 Install the front bezel

Remove the front bezel

You can access drive sled bays and the front panel display assembly by
removing the front bezel.

There is a release button on each side of the front bezel. To remove the bezel, press
in the bezel release buttons while pulling the bezel away from the node.
Remove the drive sled containing the failed drive
After you remove the front panel, locate and remove the drive sled that
contains the failed drive.

NOTE If the suspend button is pressed and drives are detected, the node attempts
to rediscover the sled and rejoin its drives after 1 hour. If the suspend button is
pressed and drives are not detected or the sled is still removed, the node
automatically smartfails the drives after 15 minutes.

1. Locate the drive sled that contains the failed drive.

A service LED is lit on the sled that contains the failed drive.

NOTE If you need to remove one of the four drive sleds in the upper left
corner of the chassis, swing the display assembly out of the way by gently
pulling the right end of the display assembly away from the chassis.

2. Press the circular gray button on the lower left corner of the sled face.

After you press the button, the Do Not Remove LED on the face of the drive
sled, a hand with a slash through it, begins to blink as the system prepares
the sled for removal. When the drive sled is safe to remove, the Do Not
Remove LED turns off.

WARNING Do not proceed until the Do Not Remove LED is off. If the LED is lit,
that means that the system is currently accessing the drives in the sled. If
you remove the sled while that LED is lit, you will interrupt processes that
protect your data.

3. On the lower left corner of the drive sled face, slide the handle release tab to
the left to free the sled handle.
4. Grab the base of the handle as close to the drive sled face as possible. Pull
the sled away from the chassis until you are able to grab it with your
fingertips.

WARNING To avoid breaking the sled handle, do not pull on the tip of the
handle and only pull until you are able to grab the sled with your fingertips.

5. Pull the drive sled out of the chassis.

WARNING As you pull the sled away from the chassis, support the back end of
the sled to make sure it doesn't fall when it clears the chassis.
Remove the failed drive
Locate the failed drive and remove it from the drive sled.

1. After you remove the drive sled from the chassis, remove the drive sled top
cover to access the drives:
a. Open the blue sled cover latches at the back of the drive sled.

Press the ends of both latches toward the outside of the sled and
swivel the latches away from the sled to release the cover.

b. Slide the sled cover toward the back of the sled to release the tabs
from the base.
 c. Lift the cover straight up to remove it from the base.

2. Remove the drive from the drive sled.

Pull straight up to remove both the drive and the paddle board connected to
the drive.
Install the new drive
Install the new drive by inserting it in the drive sled.

Some drives might come equipped with bumpers to prevent movement

when installed. The bumpers are indicated by callout 3.

1. Install the drive in the drive sled.

Push the drive and paddle card straight down into the empty connector in the
drive sled.
 placeholder image 3.5" drive, deep chassis

2. Install the drive sled cover.

a. Insert the tabs on the drive sled top cover into the slots in the drive
sled.

b. Slide the cover forward to lock it onto the sled.

 c. Swing the blue sled cover latches back into place against the drive sled
and press until you feel the latches click.

Install the drive sled

Install the drive sled into the open drive sled bay it was removed from.

1. Make sure the drive sled handle is open before inserting the drive sled.
2. With two hands, slide the drive sled into the same bay you removed it from.
3. Push the drive sled handle back into the face of the sled to secure the drive
sled in the bay.
 4. If you replaced one of the four drive sleds in the upper left corner of the
chassis, swing the display assembly back into place against the chassis until
you feel it click into place.

Install the front bezel

Install the front bezel to protect the drive sleds and display assembly.

NOTE It is possible that the bezel frame can hit the disk sled buttons and de-
activate the drive sleds when repositioning the frame, which protrudes slightly. If
the disk sled buttons are hit, it can go undetected. Be very careful when
repositioning the bezel and keep the PuTTY SSH open for a final check after the
bezel is replaced and locked.

1. Align the front bezel with the front of the chassis and push until you feel the
bezel snap into place.
2. Run isi devices drive list to confirm that the drive state is healthy.
o If drive state is healthy, proceed.
o If drive state is used, format the replaced drive.
3. Argument :isi devices drive format <bay> Ex.If you replaced location
Bay A0 # isi devices drive format A0
4. You are about to format drive bayA0, on node lnn 4.
5. Are you sure? (yes/[no]):
issue "yes".
Gather logs
After you complete maintenance on a cluster, gather cluster logs.

You must collect cluster logs after all maintenance. Cluster logs provide snapshots
of the cluster that you can review to make sure that maintenance is successful.
# isi diagnostics gather is a utility that collects configuration and log
information from a cluster and automatically uploads it to Dell for
processing. This command runs in the background so that interruption of
the user session doesn't cause problems for the gather process.

If your environment has a firewall in place that blocks the uploading of the
log information, contact the Support Engineer assisting on the case for a
temporary FTP link to which to upload the logs.

1. Ensure your power supply is providing power by running the following

command:

# isi diagnostics gather start

2. Check the status of the gather by running the following command:

# isi diagnostics gather status

3. Check that there are no warnings related to node being powered down:

# isi status.