Sunday, 6 April 2014

A Few Tweaks to My Clustered ONTAP 8.2.1RC1 Simulator Golden Image


Below I list out a few tweaks to my current Clustered Data ONTAP 8.2.1RC1 Simulator Golden Image(s*) for VMware Worksation.
*Golden Images if you’ve got more than 1 set of serial number and sysid.

Changing the serial number and sysid for the 2nd (and subsequent) node(s) in a SIM cluster has been covered before here - NetApp Simulator for Clustered ONTAP 8.2 RC Setup Notes. And that post (and many others) also contains the “Boot Menu - press Ctrl+C when prompted - and option 4 to initialize the disks.”

So, this post starts from a position that you already have a Golden Image(s) - added any additional NICs (see here), serial number/sysid settings have already been applied via the boot loader (as required), the wipe process has completed, the nodes come to the “Cluster setup wizard” and you’ve type exit, halted the node, and then taken a snapshot. This is your “pre-Cluster Setup Wizard” snapshot.

My starting place for this is a Golden Image with 7 NICs - this is to more closely replicate a FAS62XX with 6 onboard NICs and an e0M (onboard NICs are e0a,b,c,d,e,f).

Summary of Tweaks

1. (Via Console - Boot loader) Set bsdportname to e0g - this is for consistency with the SIM not having an e0M. The e0M on say a FAS62XX is effectively e0g in my SIM.

2. (Via Console - Boot loader) Set the VM to power off when you halt it - can’t see any reason for the SIM to stay on after you issue a halt, especially since it does not drop to a loader prompt…

3. (Via Console - Clustershell) Configure a temporary node-mgmt IP, give the admin account a password (so we can enable SSH), and enable ssh. This is so we can connect via SSH before running the Cluster Setup Wizard.

4. (Via SSH - Clustershell) Set up the ports so that they are like a FAS62XX with e0c and e0e for cluster (and node-mgmt for e0g which is simulating e0M.)

5. (Via SSH - Systemshell) Trash 1 of the 2 shelves of 14 x 1GB disks that originally comes with the SIM (not the shelf with the root aggregate’s disks on), then add 3 shelves of 14 x 4GB disks (could have gone for 9GB, all depends if you have the space on your system - see

6. (Via SSH - Clustershell) Finally, add 9 additional 1GB disks to the 3 x 1GB aggr0, and expand vol0 to less than 75% of the aggregate’s size (it’s a common issue with the SIM to experience the original < 1GB sized vol0 filling up.)

The Tweaks

## Console - Boot Loader ##

At the "Hit [Enter] to boot immediately, or any other key for command prompt" prompt, hit any key other than [Enter]

At the boot loader prompt (>) enter

setenv bootarg.bsdportname e0g
setenv bootarg.vm.no_poweroff_on_halt false

## Console - Clustershell ##

Welcome to the cluster setup wizard…
Do you want to create a new cluster or join an existing cluster?
Type "exit"

login: admin (there is no password)

At the clustershell (::>) enter the following replacing the IP Address, NetMask, node routing-group, and gateway, with entries of your choice

net int cre mgmt1 -role node-mgmt -home-node local -home-port e0g -address -netmask

net ro ro cre n10.0.0.0/24 -destination -gateway

sec log pass admin

Please enter your current password: [NO PASSWORD]
Enter a new password: *********
Enter it again: *********

sec log cre admin -application ssh -authmethod password -role admin

Now we can connect via SSH!

## SSH - Clustershell ##

login: admin
password: *********

net po mod -port * -role data
net po mod -port e0c,e0e -role cluster -node local
net po mod -port e0g -role node-mgmt -node local
sec log unlock diag
sec log pass diag

Enter a new password: *********
Enter it again: *********

set d

Warning: These diagnostic commands are for use by NetApp personnel only.
Do you want to continue? y


## SSH - Systemshell ##

login: diag
password: *********

At the systemshell (%) type (or copy and paste) the following commands

setenv PATH "${PATH}:/usr/sbin"
cd /sim/dev/,disks
sudo rm v0*
sudo rm ,reservations
cd /sim/dev
vsim_makedisks -h
sudo vsim_makedisks -n 14 -t 31 -a 0
sudo vsim_makedisks -n 14 -t 31 -a 2
sudo vsim_makedisks -n 14 -t 31 -a 3

Back in the clustershell (::*>) type

node reboot

Warning: Are you sure you want to reboot the node? y

## SSH - Clustershell ##

After reboot, login and run the following commands in the clustershell

login: admin
password: *********

sto agg add-disks -agg aggr0 -diskcount 9
node run local vol size vol0 +5632m
node halt

Warning: Are you sure you want to halt the node? y


After the node halts, take a snapshot!



If you mean to change mgmt1 during cluster setup, you should instead do this via the console before you run the Cluster Setup Wizard since otherwise you might encounter some issues… Also, if you change the subnet, remember to add the route in also.

Run the below to set the mgmt1 lif correctly before running Cluster Setup Wizard:

net int mod mgmt1 -home-node local -home-port ??? -address ?.?.?.? -netmask ?.?.?.?

net ro ro cre n?.?.?.?/?? -destination -gateway ?.?.?.?


  1. Great post. I wish NetApp would release a ready built image with the other serial number already configured. With that, we could completely script the building of a two-node Clustered ONTAP lab.

  2. Super helpful. Thank you for taking the time to do this. I've depended on your guides a ton.