ONTAP intentionally will shut down a controller if an NVRAM card
isn’t being charged. This is for data protection. If you do
not have power to your NVRAM card and a dirty shutdown were to occur, data loss
could happen so the idea is to do a clean shutdown so that all dirty pages are
written to disk. In the case of a cluster, being taken over is a great
solution because the controller can keep running until you get a
replacement NVRAM card and can give control back to the controller with a
working NVRAM card.
There are unsupported ways of stopping the shutdown behavior,
but they are not published and not supported because of the data loss
possibility. NetApp will not recommend or support running a controller in
a config that, by design, could lose data. I strongly suggest running the
cluster in takeover mode until the new card is in and charged. It’s
precisely for reasons like this that people buy clustered configurations, it
removes single points of failure such as an NVRAM card.
-- Adam Fox
Systems Engineer
adamfox@netapp.com
From: Zeeshan
[mailto:xischaune@yahoo.com]
Sent: Wednesday, February 04, 2009 2:32 PM
To: Toasters List
Subject: NVRAM Battery Failure
|
Hello all... |