Automatic giveback commands
Suggest changes
PDFs
By default, the take-over partner node automatically gives back storage when the off-line node is brought back on line, thus restoring the high-availability pair relationship. In most cases, this is the desired behavior. If you need to disable automatic giveback - for example, if you want to investigate the cause of the takeover before giving back – you need to be aware of the interaction of non-default settings.
If you want to… |
Use this command… |
|---|---|
Enable automatic giveback so that giveback occurs as soon as the taken-over node boots, reaches the Waiting for Giveback state, and the Delay before Auto Giveback period has expired. The default setting is true. |
|
Disable automatic giveback. The default setting is true. Note: Setting this parameter to false does not disable automatic giveback after takeover on panic; automatic giveback after takeover on panic must be disabled by setting the |
|
Disable automatic giveback after takeover on panic (this setting is enabled by default). |
|
Delay automatic giveback for a specified number of seconds (the default is 600). This option determines the minimum time that a node remains in takeover before performing an automatic giveback. |
|
How variations of the storage failover modify command affect automatic giveback
The operation of automatic giveback depends on how you configure the parameters of the storage failover modify command.
The following table lists the default settings for the storage failover modify command parameters that apply to takeover events not caused by a panic.
Parameter |
Default setting |
|---|---|
|
|
|
600 |
|
|
The following table describes how combinations of the -onreboot and -auto-giveback parameters affect automatic giveback for takeover events not caused by a panic.
|
Cause of takeover |
Does automatic giveback occur? |
|---|---|---|
|
reboot command |
Yes |
halt command, or power cycle operation issued from the Service Processor |
Yes |
|
|
reboot command |
Yes |
halt command, or power cycle operation issued from the Service Processor |
No |
|
|
reboot command |
N/A |
halt command, or power cycle operation issued from the Service Processor |
Yes |
|
|
reboot command |
No |
halt command, or power cycle operation issued from the Service Processor |
No |
The -auto-giveback parameter controls giveback after panic and all other automatic takovers. If the -onreboot parameter is set to true and a takeover occurs due to a reboot, then automatic giveback is always performed, regardless of whether the -auto-giveback parameter is set to true.
The -onreboot parameter applies to reboots and halt commands issued from ONTAP. When the -onreboot parameter is set to false, a takeover does not occur in the case of a node reboot. Therefore, automatic giveback cannot occur, regardless of whether the -auto-giveback parameter is set to true. A client disruption occurs.
The effects of automatic giveback parameter combinations that apply to panic situations.
The following table lists the storage failover modify command parameters that apply to panic situations:
Parameter |
Default setting |
|---|---|
|
|
(Privilege: Advanced) |
|
|
|
The following table describes how parameter combinations of the storage failover modify command affect automatic giveback in panic situations.
|
Does automatic giveback occur after panic? |
|---|---|
|
Yes |
|
Yes |
|
Yes |
|
No |
|
No |
|
A takeover can result from a failure not associated with a panic. A failure is experienced when communication is lost between a node and its partner, also called a heartbeat loss. If a takeover occurs because of a failure, giveback is controlled by the -onfailure parameter instead of the -auto-giveback-after-panic parameter.
|
|
|
When a node panics, it sends a panic packet to its partner node. If for any reason the panic packet is not received by the partner node, the panic can be misinterpreted as a failure. Without receipt of the panic packet, the partner node knows only that communication has been lost, and does not know that a panic has occurred. In this case, the partner node processes the loss of communication as a failure instead of a panic, and giveback is controlled by the -onfailure parameter (and not by the -auto-giveback-after-panic parameter).
|
For details on all storage failover modify parameters, see the ONTAP manual pages.