FAS2020 Setup

[it_7]

Hiya,

We've just powered up our FAS2020 box and run through the initial CLI config. Now moved to System Manager and we're having trouble configuring the storage.

We have a dual controller model with 12x 450GB disks on RAID-DP with HS. If my understanding is correct, we lose 2 disks per controller for RAID-DP and 1 disk per controller for the spares. That should leave us with ~2.7TB of storage (6x 450GB). However, the default aggregate is over 250GB in size and it's telling us we can't create more as we don't have any disks available.

So what has happened to the other 2.5TB or so of storage?

Also, what is the recommended controller IP config for an active-active implementation? All on one subnet or all on seperate subnets? We are putting each controller port on dedicated switches.

Thanks

[adamfox]

If the disk you are trying to move from one controller to the other is a spare you can just use disk assign on the CLI, although you may need to -f flag to force it.  No need to take the system down.

However, if you want/need to do this through maint. mode, then be on the console during a reboot and press CTRL-C when prompted.

[chriskranz]

If both filers are up and running (the one that owns the disk, and the one you want to own the disk) you can definitely do this on the fly.

Firstly disable disk auto_assign (options disk.auto_assign off), then on the filer that currently owns the disk, do disk remove_ownership <disk_id>. Then on the other filer, do disk assign <disk_id> or disk assign all.

Obviously providing you have no data on this disk. If you do have data (even just an empty aggregate), you need to destroy this aggregate or reformat the filer.

[it_7]

OK, all back up and running.

One minor thing, when going to the interfaces section in System Manager, i'm getting the following error:

Failed to load network interface details. API invoke failed.

This is for both controllers. I note that after choosing option 4a in maint. mode and rebooting, the CLI was complaining about missing language files. The aggregates and initial volume have both been recreated as normal.

Is this because Data ONTAP now needs to be installed, as I also seem to have lost HTTP web access.

Thanks

[chriskranz]

After you do a 4a on the system, you'll need to re-install Data ONTAP again. The flash card data that is used to 4a the system doesn't contain all the files for the full Data ONTAP system, so you need to reinstall it again.

[it_7]

OK,  what is the latest version that's supported for install on a FAS2020, and how does one install it?

Thanks for all your help guys!

[chriskranz]

7.3.2 or 7.3.3 are probably your best bet. You can download them from the NOW site, and on the download page it includes some details on how to install it. It's all fairly straight forward, but feel free to come back if you have any questions or issues with it.

[it_7]

Super! I assume it's got to be done per controller?

[chriskranz]

Yup, will need to so that both nodes of the cluster are matched.

[it_7]

Sorry for being a bit dumb, but where do I extract the files to. It says to map a drive to C$, but C$ on what?

[chriskranz]

No prob!

That'll be C$ on the filer, the administration share. \\filer-name\c$ from your windows desktop or server.

You might need to license and enable CIFS first if you haven't added your licenses back in after doing the 4a. You can find your licenses (if you didn't take a note before) on the NOW site also.

Ta...

[it_7]

Yeah thought so.

Forgot about having to license CIFS, which I why I was failing at the first hurdle!

[it_7]

Hmmm added all my licenses in and still can't access c$. Does it need a reboot?

[emollonb2s]

Hi mate,

It's pretty easy follow this steps

if you can't find the C$ on your host you have to run the cifs setup to start the smb protocol on your filer.

if you have your system up and running just run from the CLI # cifs setup

This wizard will ask you all the configuration to start the smb protocol, if you have Active Directory it will ask you all the passwords and also it's very important to have the time of the Filer syncronized with the Active Directory, to set the date you have to run date and input the date like this date 20102005160233  can't be 5 mins apart of your AD

When you finish the wizard you have to see if you have the shares published doing the follow command.

# cifs shares

it will show you all the shares that you have. one of those have to be C$ then you can map using your AD user or using the local system admin user,

Remember before finishing the Cifs setup the filer will ask you if you want a local admin user, use Administrator and if you want to login with the administrator the user have to be BUILTIN\Administrator and your password.

Good luck!

It's easy

[aborzenkov]

There is no need to extract anything. Put .exe file on HTTP server and use software update command to install it directly from there (it could be software get/software install in older releases).

[it_7]

The instructions in the download say to extract the zip file to C$\etc on the filers, then run the download command on each filer to complete the process.

As i'm already part way through this process, ie just waiting for the download command to complete, i'll have to stick with this way!

[emollonb2s]

Another way to upgrade it's mapping the C$ on your PC and going to /etc/software (if this folder does not exist create it) then put your file .exe inside this folder and from the command cli type # software list  then you have to see the file that you put inside.

Then you can do # software install after finished

# download

then

# reboot

(if you have the cluster then cf disable )

then the filer will be upgraded

and cf enable.

[t_kumar]

I have a couple of  FAS2020(dual head) in clustere mode.  In one setup i see the IQN (of host) logged in both e0a and e0b while on the other cluster the iqn is logged only on e0b.

what is the best practice for HA?

[t_kumar]

iscsi was enabled on both e0a and e0b earlier which caused this. Now only e0b is enabled for iscsi and hence isolating the  iscsi traffic .