Disk Drive Replacement

This section describes how to replace a failed disk drive in the Solace PubSub+ 3530 or 3560 appliance.

A PubSub+ 3530 or PubSub+ 3560 has four system drive bays that are accessible from the front of the appliance. The drives are fully hot swappable, which allows you to remove and install a disk drive without powering down the appliance or removing it from its rack.

ESD/Safety Precautions

Before replacing appliance disk drives, review the following safety warning.

To prevent equipment damage from Electrostatic Discharge (ESD), always ensure that the appliance chassis is electrically connected to earth ground. Always follow ESD prevention procedures when removing and replacing power supplies. Use an anti-static wrist strap, or another anti-static device. If no wrist strap or mat is available, touch the metal part of the chassis to ground yourself.

Tools and Equipment Requirements

  • Electrostatic discharge wrist strap
  • Anti-static mat
  • Replacement disk drive (supplied by Solace)

Replacing PubSub+ 3530/3560 Disk Drives

To replace a disk drive in a Solace PubSub+ 3530 or 3560, perform the following steps:

Step 1: Shut Down Failed Disk Drive

To shut down a failed disk drive, use the following procedure:

  1. Enter the following command at the User EXEC Level prompt: 
    solace> enable
    solace#

    The Solace CLI is now at the Privileged EXEC level.

  2. Recommended: As a precaution, enter the following Privileged EXEC command to copy the current configuration database file from the event broker to an external SFTP or SCP server (as applicable) for backup:

    Ensure that you have an SFTP or SCP server on the event broker network to which you have IP connectivity. 

    solace# copy current-config sftp://[<username>@]<ip-addr>/<remote-pathname>

    or

    solace# copy current-config scp://[<username>@]<ip-addr>/<remote-pathname>

    Where:

    <username> is the CLI user name and <ip-addr> and <remote-pathname> point to the location where the current configuration database file is copied onto the SFTP or SCP server

  3. To confirm that the disk drive has failed, enter the following User EXEC command: 
    solace# show disk

    An example display from show disk User EXEC command with a failed disk drive is:

    Reload required:               No


Attribute                        Disk 1               Disk 2
------------------ -------------------- --------------------
Device Model        INTEL SSDSC2BB120G4  INTEL SSDSC2BB120G4
Capacity (MB)                    114473               114473
Admin State                     enabled              enabled
Operational State                failed                   up

RAID in fully redundant state: Yes

Attribute                        Disk 3               Disk 4
------------------ -------------------- --------------------
Device Model
Capacity (MB)
Admin State                     shutdown              shutdown
Operational State                   down                  down

RAID in fully redundant state: No

    Note which disk drive has failed.

  4. To shut down the failed disk drive, enter the following CONFIG commands: 
    solace# configure
    solace(configure)# hardware
    solace(configure/hardware)# disk <disk-name> shutdown

    Where:

    <disk-name> is the number of the failed disk drive. Valid values are 1, 2, 3, or 4.

  5. To confirm that the failed disk drive is shut down, enter the following Global CONFIG command: 
    solace(configure/hardware)# show disk

    In the displayed output, the Admin State for the disk should be shutdown.

Step 2: Remove Failed Disk Drive

Do not attempt to remove a disk drive before completing Disk Drive Replacement. Removing a disk drive from an appliance without performing a disk shutdown results in service interruption and/or data loss.

To remove a failed disk drive from an appliance, follow this procedure:

  1. To reveal the available disk drives, pull down the top of the hinged portion of the front panel.
  2. Locate the disk drive you want to remove.
  3. Slide the release button to the right (that is, toward the disk drive) to open the latch on the system drive bay.

    Opening the System Drive Bay

    The latch springs open and the disk drive is released from the system drive bay.

  4. Pull the disk drive completely out of the system drive bay.

    Removing a Disk Drive

  5. Return the failed disk drive to Solace.

Step 3: Install New Disk Drive

To install a new Solace-supplied disk drive in an appliance, follow this procedure:

Only use a new Solace-approved replacement disk drive for this procedure. Attempting to use an existing disk drive from another appliance or any other source could result in communication problems and/or data loss.

  1. Insert the disk drive into the vacant appliance drive bay, and shut the latch on the system drive bay to return it to the full closed position.

    Install the new disk drive into the vacant bay only. Moving the operational disk drive to the vacant bay may result in communication problems or data loss within the event broker.

    The new disk drive is now locked into the appliance bay

  2. Close the hinged portion of the front panel.

Step 4: Enable and Rebuild New Disk

To enable and rebuild a new disk drive, follow this procedure:

  1. To enable the new disk drive, enter the following Hardware CONFIG command: 
    solace# configure
    solace(configure)# hardware
    solace(configure/hardware)# disk <disk-name> no shutdown

    Where:

    <disk-name> is the number of the failed disk drive. Valid values are 1, 2, 3, or 4.

  2. To confirm that the new disk drive is in waiting for rebuild mode, enter the following command: 
    solace(configure/hardware)# show disk

    An example display from show disk with a waiting for rebuild disk is:

    Reload required:               No

Attribute                        Disk 1               Disk 2
------------------ -------------------- --------------------
Device Mode         INTEL SSDSC2BB120G4  INTEL SSDSC2BB120G4
Capacity (MB)                    114473               114473
Admin State                     enabled              enabled
Operational State   waiting-for-rebuild                   up

RAID in fully redundant state: Yes

Attribute                        Disk 3               Disk 4
------------------ -------------------- --------------------
Device Model
Capacity (MB)
Admin State                     shutdown              shutdown
Operational State                   down                  down

RAID in fully redundant state: No

    It can take several seconds for the Operational State of a disk to change to waiting-for-rebuild.

  3. Enter the following commands: 
    solace(configure/hardware)# end
    solace# admin
    solace(admin)#
    solace(admin)# disk
  4. To start rebuilding the new disk drive, enter the following commands: 
    solace(admin/disk)# rebuild
  5. Optional: To shorten the rebuilding time, enter the following Disk Admin EXEC command, then y when prompted to continue: 
    solace(admin/disk)# rebuild-speed high
    Changing the disk rebuild time may impact performance of the appliance.
    This should only be done if the appliance is not passing traffic.
    Do you want to continue (y/n)? y
  6. To confirm that the new disk drive is recovering, enter the following commands: 
    solace(admin/disk)# exit
    solace(admin)# show disk

    It can take several seconds for the Operational State of a disk to change to recovering.

    An example display from show disk with a recovering disk drive is:

    Reload required:               No

Attribute                        Disk 1               Disk 2
------------------ -------------------- --------------------
Device Mode         INTEL SSDSC2BB120G4  INTEL SSDSC2BB120G4
Capacity (MB)                    114473               114473
Admin State                     enabled              enabled
Operational State            recovering                   up

RAID in fully redundant state: Yes

Attribute                        Disk 3               Disk 4
------------------ -------------------- --------------------
Device Model
Capacity (MB)
Admin State                     shutdown              shutdown
Operational State                   down                  down

RAID in fully redundant state: No