nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Failover takeover on node reboot

areti

Good afternoon!
I have NetApp 2750 ONTAP 9.7 Cluster mode (two HA nodes: netapp03-01 and netapp03-02)

When running the command "system node reboot -node netapp03-02"
I get a warning: ... To ensure continuity of service, use the "storage failover takeover" command.

Should I run the command "storage failover takeover" before rebooting the node?
Indeed, in HA mode, the "failover takeover" operation should be performed automatically when a node is disconnected or lost. Is not it?

node reboot.jpg

Find more posts tagged with

Accepted answers

paul_stejskal

A takeover is a reboot of the node being taken over from. I'm not sure if this is the question.

All comments

paul_stejskal

A takeover is a reboot of the node being taken over from. I'm not sure if this is the question.

a4ashu

Reboot will cause a panic mode takeover and wont be properly negotiated as it would had been when you run Storage failover which will reboot the node anyways .

I think best practice is if in Ha if you want to reboot node A .. do a takeover from node B there wont be future giveback issues

areti

Thanks for the answer, I want to clarify.
The "storage failover takeover" command invokes the takeover of the node, then the node is rebooted, then a giveback automatic occurs?

So, for a scheduled reboot, is it enough to run "storage failover takeover" command only? I shouldn't be running the "system node reboot" command. Right?

Are "storage failover takeover" command and "takeover node" execution from GUI equivalent?

paul_stejskal

Yes to all, and the first question also depends on if the automatic giveback is enabled.

a4ashu

Unfortunately I am not too suave with “storage failover takeover” via GUI

yes you’re right you don’t have to do a node reboot if it’s scheduled . Do a storage failover takeover and then wait for auto giveback this will ensure your aggregates are properly taken over and properly given back

node reboot involves issues sometimes

andris

Automatic giveback is enabled, by default. You might have noticed that the GUI gives you an "Action" to disable automatic giveback. This might be needed if you're not just rebooting the partner node - instead, you might need to be in control of when the giveback occurs.

The storage failover modify man page provides more info on this:

[-auto-giveback {true|false}] - Auto Giveback Enabled

This optionally specifies whether automatic giveback operations are enabled. An automatic giveback operation is invoked when one node of a failover pair is in takeover mode and the failed node is repaired and restarts. When the repaired node boots, the node in takeover mode detects this and initiates a giveback operation. The default setting is true. This parameter is not applicable if takeover was because of a disruption in the partner's operation. For those cases, use the -auto-giveback-after-panic parameter, instead.