We have a NetApp FAS 2240-4 model, 7-mode system, double-controller in single chassis, with no external storage, running Data ONTAP 8.1.3 . NETAPP2240A is the main controller, and auto reboots continually, so data on the storage is unreachable.
The second controller NETAPP2240B is the passive backup partner node, I expect it to takeover the main controller automatically but it does not.
If I use “takeover” command on NETAPPB, is there a possibility that it causes data loss? The data on the storage system is needed and I don’t want to risk the data.
I will appreciate if anyone with an experience on these old systems answers.
Chance of on-disk data loss from "takeover" is almost zero.
However, for data service to continue to work, the partner address setup on both nodes needs to be the same. If you'd like to post the uncensored output of "sysconfig -a", "ifconfig -a", "rdfile /etc/hosts" and "rdfile /etc/rc" from each node we could give you our opinion on if that is likely.
You are welcome to do a find and replace on your domain name, but that would be the limit of chances to the output.
I can get some output from ControllerB, but since the ControllerA autoreboots continually, I can not interrupt (Control+C brings only the boot menu) and can not get to the command line on this nodeB.
Assuming the configuration has been made properly, I consider applying “takeover” or “forcetakeover” commands on ControllerB. If the configuration is not OK, I don’t think there would be a chance to get the data anyway.
Since I can not get to the command line, I can not get any output from ControllerA, and not be sure of the configuration of ControllerA. Is it possible to get output from an auto rebooting controller?
Hi there, if you press Control-C right after the BIOS message, you will get a LOADER A> or B prompt and break the boot loop. At that point, performing "cf takeover" is fine - it'll either work or it won't.
When I press Control-C right after the BIOS message, I can’t get to LOADER A> prompt but can get and login to LOADER B> prompt. At that point, I performed "cf takeover" and there is a messsage;
“cf: takeover cannot be performed because of reason (partner halted in notakeover mode)”
Is this mentioned partner which is halted in notakeover mode, LOADER A, or is it itself LOADER B? If it is LOADER B, at that point is it possible to get it to “takeover mode”?
What about Force takeover command? It is said to be dangerous to perform it on the manual;
forcetakeover [-f] forcetakeover is dangerous and can lead to data corruption; in almost all cases, use cf takeover instead. Forces one node to take over its partner, even though the node detects an error that would otherwise prevent a takeover. For example, normally, if a detached or faulty Infiniband cable between the nodes causes the nodes’ NVRAM contents to be unsynchronized, takeover is disabled. However, if you enter the cf forcetakeover command, the node takes over its partner despite the unsynchronized NVRAM contents. This command might cause the node being taken over to lose client data. If you use the -f option, the cf command allows such a forcetakeover to proceed without requiring confirmation by the operator.