Subscribe

FAS3210 Active/Active in takeover mode (one controller dead): How to sanitize all disks remotely?

[ Edited ]

Dear community,

 

First of all, I have to mention that I am not very experienced with NetApp systems.

 

My task:

I want to decommission and sanitize all disks of this system. But so far all my attempts to do that remotely (cannot really go on site to move disks from one shelf to the other) failed.

 

Situation:

ControllerA, 12 disks, one agg0 with root volume vol0, works and has taken over Controller B

ControllerB, 12 disks, one agg0 with root volume vol0, dead, cannot use Ctrl-C to setup from sratch (at reboot reports "PANIC: Uncorrectable Machine Check Error at CPU1 ...")

I can see and maybe operate the disks of Controller B via the partner command. I have to note that the systems reports "Backplane Interconnect link down". Additionally I have currently powerd off ControllerB via the service processor. Ontap is NetApp Release 8.0.2P6

 

My solution would be:

1) ControllerA takes ownership of all disks of ControllerB and I sanitize all 12 disks of ControllerB

2) I create a new root volume on 3 of the sanitized disks

3) I sanitize the 12 disks of ControllerA

 

Issues:

Taking ownership of the ControllerB disks (all sort of disk assign/remove/fail..) or offlining/detroying ControllerB aggregate/volume already fails. Even in priv set advanced the system does not allow me to "destroy everything". For ControllerA I could do a re-setup via Ctrl-C plus option 4, but then still the question would be how to get access to the disks of ControllerB

 

Well. Does anybody have an idea how I could achieve this task? Prerequisite would be to not go on site.

 

Thanks in advance

Thomas

 

PS: access to both service processors is available

Re: FAS3210 Active/Active in takeover mode (one controller dead): How to sanitize all disks remotely

[ Edited ]

Have you disabled cluster take over?

Re: FAS3210 Active/Active in takeover mode (one controller dead): How to sanitize all disks remotely

Right now cluster takeover is enabled. I did disable it before. But this config then did not really help me in accessing the ControllerB disks.