NCDA Study Part 2.3 – Data ONTAP: Configuration

As always download the newest version of the blueprint breakdown from here.

This section will cover plexes.  A plex is composed of RAID groups and is associated with an aggregate.  An aggregate will typically have one plex, plex0, and in the case of SyncMirror will have two plexes, plex0 and plex1.  In this case plex1 will be a mirrored copy of plex0.  For the blueprint’s concern, this section will cover just verifying via the CLI and System Manager where you can see the plex information in Data ONTAP.  Later, using SyncMirror you’ll be able to see both plex0 and plex1 in that configuration.

We’re going to perform the following labs in both the CLI and System Manager where the capabilities to do in both exist.

  • Verify

Verify Plex – CLI

To look at the plex information, you have to look at the aggregate.  When you run the command below you will notice how the hierarchy works in regards to plexes, aggregates and RAID groups.  You can use either of these commands to get the same results.

aggr status –r
sysconfig –r

cli_plex_verify_step1

Looking at this output, you will see that the Plex for aggregate 0 is online, normal and active.  You will also notice that the output starts with the aggregate, then plex, then the RAID group.  This aggregate is aggr0, with plex0 inside, and composed of rg0.

Verify Plex – System Manager

Using System Manager to look at the plex is fairly straightforward.  Once you open and connect to the Data ONTAP device, expand the device, Storage, click on Aggregates and then click on the Disk Layout tab at the very bottom.  This will bring up the plex view. cli_plex_verify_step2

Advertisements

NCDA Study Part 2.2 – Data ONTAP: Configuration

As always download the newest version of the blueprint breakdown from here.

The next section on the configuration overview covers aggregates.  As discussed aggregates are composed of a plex which is made up of disk groups and represent the combined performance of the disk groups it’s composed of.  After initial configuration there will always be an aggr0 on the device.  This aggregate holds the root volume, /vol/vol0, and cannot be deleted.  You can however create another aggregate from the available disks.  Below are the steps for creating the various aggregate types.

We’re going to perform the following labs in both the CLI and System Manager where the capabilities to do in both exist.

  • Create 32/64bit aggregates – 1 RAID4 and 1 RAID-DP to show differences
  • Enable/disable Snapshots

Create 32bit Aggregate in RAID 4– CLI

  • SSH to the device and confirm that there are free disks to create an aggregate from
aggr status –r

We have 14 FCAL 15000 RPM drives free to create the aggregate

cli_aggr32_create_step1
  • Create a 32bit aggregate using only 7 of the 14 free disks.  This command below will create a 32bit aggregate named ‘newaggr32’, in a RAID group consisting of 7 drives in RAID 4, using 7 of the FCAL disks we see as spares.
aggr create newaggr32 –t raid4 –r 7 –B 32 –T FCAL 7

cli_aggr32_create_step2

Our new aggregate is created and is brought online

  • Verify aggregate has the appropriate settings
aggr status

cli_aggr32_create_step3

Create 64bit Aggregate in RAID-DP– CLI

  • SSH to the device and confirm that there are free disks to create an aggregate from
aggr status -r

cli_aggr64_create_step1

  • Create a 64bit aggregate using only 7 of the 14 free disks.  This command below will create a 64bit aggregate named ‘newaggr64’, in a RAID group consisting of 7 drives in RAID 4, using 7 of the FCAL disks we see as spares.
aggr create newaggr64 –t raid_dp –r 7 –B 64 –T FCAL7

cli_aggr64_create_step2

  • Verify aggregate
aggr status

cli_aggr64_create_step3

Create Aggregate – System Manager

Creating an aggregate in System Manager is pretty straightforward.  I’m not going to create each type within the GUI, but you’ll see the options when we’re going through.  Once you open and connect to the Data ONTAP device expand the device, Storage and click on Aggregates

  • Click on Create and click Next

osm_aggr_create_step1

  • Name the aggregate, choose the RAID type and the block format then click Next

osm_aggr_create_step2

  • Choose the controller you want to create it on.  This only applies if you have dual controllers in the array.  Choose the disk group and click Next.

osm_aggr_create_step3

  • Click on Select Disks… and type in the number of disks you want to use to create the RAID Group and click OK

osm_aggr_create_step4

  • Select the RAID Group Size and the number of disks in each group.  Remember, RAID Groups are composed of Parity and in the case of RAID-DP, Double-Parity drives.  The more RAID Groups, the more parity drives.  Depending on the model of controller and disk type, RAID Groups can be different sizes.  In our case we only want one RAID Group because we’re not nearing the maximums size the RAID Group can be with only 7 FCAL drives.  This also limits wasting drives for Parity when we don’t need them.  Also, remember that RAID Groups can’t be shrunk once they’re created.  You can only add disks to them or grow the group size up to the maximum.  Pay special attention when creating the groups with lots of disks that you keep the RAID Group sizes very close to the same.  This will keep performance consistent.  System Manager will give you a warning if you have disparate RAID Group sizes.

osm_aggr_create_step5

  • This is the correct setting or the setting that we want to use.  Click Next to continue.

osm_aggr_create_step6

  • Verify that the settings are correct and click on Create.   Data ONTAP will zero the disks and create the aggregate.

osm_aggr_create_step7

  • The aggregate is now completed and online.

osm_aggr_create_step8

  • Verify that all looks good

osm_aggr_create_step9

Enable/Disable Snapshots – CLI

The next section covers enabling and disabling Snapshots on the aggregate.  Aggregate Snapshots can be used to revert the entire aggregate back to a point in time.  While this is very rarely done, with the exception of a dire emergency type of scenario, it is a configurable option.  You can’t enable or disable Snapshot options for the aggregate from within System Manager.  It can only be done via the CLI.

By default when a new aggregate is created from the CLI and from System Manager, the Snap Reserve is 0% and the Snap schedule is 0 1 4@9, 14, 19.  This means that a snapshot will be taken 0 weekly, 1 daily, 4 per hour at 9am, 2pm and 7pm each day and that a Snapshot doesn’t have dedicated area wthin the aggregate to store the Snapshots, it will consume free space within the aggregate as much as it can get.  This is the default schedule and default Snap Reserve setting for Data ONTAP 8.x.  It can be modified according to whatever the needs are.

Snapshots are enabled by default when an aggregate is created.  You will not get a confirmation when you use this command.  Also, it will not delete any existing Snapshots either.  You’ll have to delete them by hand.

aggr options newaggr nosnap on
snap delete –A  newaggr nightly.0

cli_aggr_snap_step1

Deletes a Snapshot of the aggregate ‘newaggr’ named ‘nightly.0’.

You can effectively disable Snapshots by either using the on/true command above or by turning off the Snap schedule all together.

Snap sched –A newaggr
snap sched –A newaggr 0 0 0
snap sched –A newaggr

cli_aggr_snap_step2

Here are the commands to show the default Snapshot schedule for all aggregates:

snap sched –A

As you can see the default schedule is 0 1 4@9, 14, 19.  You can change the schedule to something else by using the following command:

snap sched –A newaggr 1 1 1@4,12,20

cli_aggr_snap_step3

1 weekly snapshot, 1 daily snapshot and 1 hourly snapshot at 4am, 12pm and 8pm.

NCDA Study Part 2.1 – Data ONTAP: Configuration

When you first boot up a NetApp array, you’ll need to go through the initial setup to configure the basics for Data ONTAP.   I covered the process of basic configuration back when I discussed building the Data ONTAP Simulator VMs.  So this first section will look very familiar.

We’re going to perform the following labs in both the CLI and System Manager where the capabilities to do in both exist.

  • Initial Setup of the controller
  • License the products

Initial Setup – CLI

The initial configuration of Data ONTAP prepares the device for remote connectivity and allows you to SSH into the device once the configuration is completed.  From that point, more in depth configuration needs to take place to prepare the device for service.  This part of the setup process is only able to be performed from the CLI of the device.  This is typically done using the cable provided by NetApp on the physical appliance or by using a blue Cisco roll-over cable if you don’t have the NetApp cable.  The cable is plugged into the serial IOIO port on the back of the first controller.  You can use any terminal emulator at that point pointed to the COM port on a laptop with the settings 9600baud 8N1.  Because I’m doing this in VMware, it’s easier because you just use the console window of the virtual machine.

  • Boot device
  • When prompted to access the Boot menu press CTRL+C
  • Press 4 to Zeroize and clean drives
  • Confirm with y
  • Confirm with y
  • System will reboot

Allow the wipe and disk assign process to take place.  When the system finishes zeroizing the disks and assigning ownership, it will then move into configuration mode.

  • Enter the hostname of the simulator (I named mine ONTAP1/2)
  • Take default on IPv6 of no
  • Take default on interface groups of no
  • Enter the IP address of the e0a interface of 192.168.175.10
  • Enter the Subnet Mask of 255.255.255.0
  • Take default on media type of auto
  • Take default on flow control of full
  • Take default on jumbo frames of no
  • Enter the IP address of the e0b interface of 192.168.175.11
  • Enter the Subnet Mask of 255.255.255.0
  • Take default on media type of auto
  • Take default on flow control of full
  • Take default on jumbo frames of no
  • Don’t enter an IP for e0c
  • Don’t enter an IP for e0d
  • Take the default on web interface setup of no
  • Enter the IP address of the default gateway of 192.168.175.51(our router’s IP address)
  • Don’t enter the IP address for an administrative host
  • Enter the hours of the timezone from GMT of -6 (Chicago)
  • Enter the name of where the filer is located of Chicago
  • Take the default for root directory of /vol/vol0/home/http
  • Take the default of DNS resolver of no
  • Take the default of NIS client of no
  • Press Enter
  • Take the default for ACP of no
  • Allow VM to finish configuration

Once this initial setup is complete, you can now login to the device via SSH.  From this point you can run the sysconfig command to get some basic information about the device.  If you’re device is an HA pair, you’ll need to configure the other controller using the same methods above using different IP addressing.

intial_config_step1

Licensing – CLI

The next step is to license the device.  This can be done using either the OnCommand System Manager or through the CLI.  It’s much faster to do it via the CLI as with System Manager, you have to add them one at a time.  Since we’re using the Data ONTAP Simulator, it comes with free licenses for use with this VM.  I typically put them into Notepad++ so they’re easier to add all of them at once.

7-mode Licenses
—————
License Key Purpose
a_sis MTVVGAF Deduplication (Advanced Single-Instance Storage)
cifs DZDACHD CIFS protocol
compression CEVIVFK Compression
disk_sanitization PZKEAZL Disk sanitization
http NAZOMKC HTTP protocol
fcp BKHEXNB Works but does not provide much functionality
flex_cache_nfs ADIPPVM FlexCache license
flex_clone ANLEAZL FlexClone license
iscsi BSLRLTG iSCSI protocol
nearstore_option ELNRLTG NearStore personality
nfs BQOEAZL NFS protocol
operations_manager CYLGWWF Operations manager
protection_manager CGUKRDE Protection manager
provisioning_manager UYNXFJJ Provisioning manager
smdomino RKBAFSN SnapManager for Domino*
smsql HNGEAZL SnapManager for SQL Server*
snaplock ZOJPPVM SnapLock WORM Compliance edition
snaplock_enterprise PTZZESN SnapLock WORM Enterprise edition
snapmanagerexchange BCJEAZL SnapManager for Exchange*
snapmanager_hyperv COIRLTG SnapManager for Hyper-V
snapmanager_oracle QZJTKCL SnapManager for Oracle
snapmanager_sap WICPMKC SnapManager for SAP
snapmanager_sharepoint UPDCBQH SnapManager for SharePoint
snapmirror DFVXFJJ SnapMirror between simulators
snapmirror_sync XJQIVFK Synchronous SnapMirror between simulators
snaprestore DNDCBQH SnapRestore
snapvalidator JQAACHD Oracle SnapValidator license
sv_linux_pri ZYICXLC Open Systems SnapVault from Linux clients*
sv_ontap_pri PVOIVFK SnapVault “primary” (source filer)
sv_ontap_sec PDXMQMI SnapVault “secondary” (destination filer)
sv_unix_pri RQAYBFE Open Systems SnapVault from UNIX clients*
sv_windows_pri ZOPRKAM Open Systems SnapVault from Windows clients*
syncmirror_local RIQTKCL SyncMirror (think RAID 4+1)
vfiler NQBYFJJ Multiple virtual Filers
vld JGFRLTG

The command to add all of these in is license add <license>.  So we end up with this as the script:

 license add MTVVGAF
 license add DZDACHD
 license add CEVIVFK
 license add PZKEAZL
 license add NAZOMKC
 license add BKHEXNB
 license add ADIPPVM
 license add ANLEAZL
 license add BSLRLTG
 license add ELNRLTG
 license add BQOEAZL
 license add CYLGWWF
 license add CGUKRDE
 license add UYNXFJJ
 license add RKBAFSN
 license add HNGEAZL
 license add ZOJPPVM
 license add PTZZESN
 license add BCJEAZL
 license add COIRLTG
 license add QZJTKCL
 license add WICPMKC
 license add UPDCBQH
 license add DFVXFJJ
 license add XJQIVFK
 license add DNDCBQH
 license add JQAACHD
 license add ZYICXLC
 license add PVOIVFK
 license add PDXMQMI
 license add RQAYBFE
 license add ZOPRKAM
 license add RIQTKCL
 license add NQBYFJJ
 license add JGFRLTG

From the CLI, paste in the commands and press enter.  This is what you will see:

cli_license_step1

Some of the licenses will already be part of the Simulator and some will not.  Some might even require a reboot of the controller to facilitate the activation.

Licensing – OnCommand System Manager

To perform this same task from OnCommand System Manager, you’ll need to open it up and connect to the controller.  System Manager requires Java Runtime 32bit and Adobe Flash Player to be installed before you can install the application.  The installation for System Manager is straightforward, take the default settings.

When you launch the System Manager application, you’ll need to add a controller:

  • Click the Add button once the console opens and input the IP address of the controller as well as the root user and login password, and click Add

osm_step1

  • Allow the controller to be found and added and then double click the newly added controller to connect to it

osm_step2

  • Expand the controller name, Configuration, System Tools and Licenses.  I’m going to add the CIFS license to the device by clicking on Add and then pasting in the license code from above.

osm_step3

Once the license is added, you’ll see it in the list of licenses.  All licenses have to be added one at a time via System Manager.  This makes it a very lengthy process adding all the licenses from above.  It’s much simpler to add them via the CLI and they can be bulk imported very quickly.

The move: Blogger to WordPress

I made the move to WordPress from Blogger.  I wasn’t happy with how little Blogger had in the way of options.  I’ve used WordPress before and it was a better choice. I extremely dislike the way that Blogger formats my posts.  I do the majority of my posting in Word and then copy/paste it in.  If I had bullets, it would completely jack them up.  It was much more work to go back and fix them.  So as it stands, here I am on this site for the time being.  We’ll see how it turns out.  Eventually I’d like to host my own from a server at my house.  We’ll see when I get around to that.  So please, bear with me as I find a theme and layout I like.  You’ll probably notice it changing a few times here and there.