You can reconfigure the hardware to return to a single-controller configuration.
Before you begin
The HA pair software must be disabled.
Steps
- Halt both nodes by entering the following command on each console:
halt
- Using the information you recorded earlier in the disk shelves or loops with mixed storage, physically move the disks to a disk shelf in a loop that belongs to the node that owns the disks.
For example, if the disk is owned by Node B, move it to a disk shelf in a loop that is owned by Node B.
Note: Alternatively, you can move the data on the disks using a product such as Snapshot software rather than physically moving the disks. See the
Data ONTAP Data Protection Online Backup and Recovery Guide for 7-Mode.
After moving the data from the disk, you can zero the disk and use the disk remove_ownership command to erase the disk's ownership information. See the Data ONTAP Storage Management Guide for 7-Mode.
- If you are completely removing one node so that all the disk shelves will belong to a single stand-alone node, complete the following substeps:
- Boot the node that you are removing into Maintenance mode as described in the Data ONTAP System Administration Guide for 7-Mode.
- Enter the following command to reassign all disk shelves so that they all belong to the remaining node:
disk reassign [-o old_name | -s old_sysid] [-n new_name] -d new_sysid
- Halt the node by entering the following command:
halt
- Turn off the power to each node, then turn off the power to the disk shelves and unplug them from the power source.
- Ground yourself, then remove the HA interconnect cables from both nodes.
See the hardware documentation for your system for details.
- Move or remove the adapter used for the HA interconnect:
If your system uses... |
Then... |
An HA interconnect adapter or an FC-VI adapter |
Remove the adapter from the system. |
An NVRAM5 or NVRAM6 adapter |
You might need to change the slot position of the adapter. See the Hardware Universe for details about expansion slot usage for the adapter. |
- Recable the system, depending on the type of system:
If you are converting a... |
Then... |
System with nonmirrored disks |
- Disconnect all cabling from the Channel B loop on the local node.
- Repeat for the partner node.
|
System with mirrored disks or a redundant Channel B loop |
- Connect the local node to the open Channel B loop in its local disk shelves, as described in the appropriate disk shelf guide.
- Repeat for the partner node.
|
- Power on the disk shelves, then the individual nodes, monitoring the system console for error messages during the boot process.
- Run all system diagnostics at the boot prompt by entering the following command on the system console:
boot_diags
- Unset the partner system ID by entering the following command at the prompt:
unsetenv partner-sysid
- Boot the node by entering the following command:
boot_ontap
- Check HA pair status by entering the following command:
cf status If the HA pair is disabled, you see the following message:
Non-HA mode.
- Repeat Step 1 through Step 12 for the partner node.