Random Short Take #15

Here are a few links to some random news items and other content that I recently found interesting. You might find them interesting too. Episode 15 – it could become a regular thing. Maybe every other week? Fortnightly even.

EMC – BRS Announcements – Q3 2013

Disclaimer: As part of my participation in EMC Elect 2013, EMC sometimes provides me with access to product briefings before new product announcements are made. I don’t want to turn this blog into another avenue for EMC marketing, and EMC are not interested in that either. Nonetheless, I’ve had the opportunity via various channels to actually try some of this stuff and I thought it was worth putting up here. I’ll reiterate though, I haven’t had the chance to verify everything for myself. This is more a prompt for you to go and have a look for yourself.

So, EMC made a few announcements around its BRS line today and I thought some of the Data Domain stuff was noteworthy. Four new models were released; here’s a table of speeds and feeds. Keep in mind that these are numbers published by EMC, not verified by me. As always, your mileage might vary.


In any case, the DD2500 is the replacement for the DD640, the DD4200 replaces the DD670, the DD4500 replaces the DD860 and the DD7200 replaces the DD890. One of the cooler parts of this announcement, in my opinion, is the improved archive support. This is something we’ve been investigating internally as part of our take the Centera out the back and shoot it  project. Here’s a screenshot of a marketing slide that includes a number of logos.

DD2Other aspects of the announcement include EMC Avamar 7 and NetWorker 8.1. The Avamar NDMP Accelerator now supports backup for Isilon, in addition to VNX, VNXe, Celerra and NetApp systems. Being a tape user, I’m also mildly excited about DD Boost over Fibre Channel support in NetWorker 8.1, although I’ve not had the chance to try it in our lab yet, so I’ll restrain my enthusiasm until I’ve had time to test it out.

In any case, have a chat to your local EMC BRS team about this stuff if you think it might work for you. You can also read more about it on EMC Pulse and the Reflections blog. When I’ve had a chance to test DD Boost over FC I’ll post it up here.


OT: My absence

To my three loyal readers, I must apologise for the relative paucity of blog posts recently. I’ve been consulting in a mid-large government department lately and haven’t done a lot of work that lends itself to this blog. Instead, I’ve been doing lots of pictures and lots of typing, and developing a plan of attack for their storage and data protection environment. It’s been a challenge, in so far as they have a bad habit of throwing storage at problems before checking if they’re really problems. To wit, the CX3-40f I configured last year with 5 DAEs is now fully populated with all 16 DAEs. There’s also a CX700 still doing its thing (almost fully populated), and a few other CLARiiONs performing other duties. I’ve also seen a few bizarre things happen there too, the strangest of which was when one of the sys admins uninstalled the e-mail archive program from the servers. This, in turn, deleted some C-CLIPs from the Centera, as they had no fixed policy on retention in place. Who’d have thought the Centera Backup and Recovery Module (CBRM) for EMC NetWorker actually worked? I could go on, but I think it’s best if I don’t.

So, my three month stint is almost up, but it looks like it will be extended another three months, and possibly extended again after that. So the blog posts may still be few and far between for the next little while, although I am hoping to start work on vSphere 4 shortly, and will no doubt have some stuff to write about Brocade / Cisco interop and how to make and break Cisco 9513 Directors. So, er, thanks for reading …

EMC NetWorker Library Configuration

Centuries ago, or at least in 2006, I worked at a fairly large EMC NetWorker site, looking after an ADIC (now Quantum) i2000 with 300 slots and 10 LTO-2 drives. It was a lot of fun when things went wrong, as they invariably did.  So occasionally I had to use NetWorker tools to perform configuration or rectification activities on said library d’enfer. I don’t for a moment pretend to be an expert with Networker. Nor am I necessarily super competent with backup and recovery in general. As I’ve mentioend previously, if you want to know about backup and recovery, buy a copy of Preston’s book. Fortunately for us, he has also started a NetWorker-focused blog, so I don’t need to pretend I know much about the product.

The one thing I did like about NetWorker was its extensive array of command-line tools. For a brief (and somewhat dated) overview of these commands, check out this site. But back to my story. I’ve been doing some migration work for a customer with ESX / NetWorker / some CLARiiONs and needed to introduce a new library to their backup host. For some reason, when we added the drivers for the new library, it tried (and failed) to update the drivers on the old library as well. As a result of this, NetWorker started having conniptions, and the two Dedicated Storage Nodes attached to the library weren’t particularly happy either. So the client decided we should tear it down and only present the old library to the NetWorker server. Not necessarily the path I would have preferred to take, as I’ll have to clean it up next weekend, but it was getting late on Sunday night and we were all keen to not have to deal with this problem.

So, the point of my story is that you can use a tool called jbconfig (the jukebox resource configuration tool) to configure libraries for NetWorker via the CLI. Note that you need to have drivers loaded on the host prior to configuring NetWorker – this ain’t no Backup Exec. So here’s an example of what you might see when running jbconfig, assuming that the library hasn’t been configured previously.

C:\Documents and Settings\user>jbconfig

Jbconfig is running on host backuphost.internal (Windows Server
2003 5.2),
and is using backuphost.internal as the NetWorker server.

1) Configure an AlphaStor Library.
2) Configure an Autodetected SCSI Jukebox.
3) Configure an Autodetected NDMP SCSI Jukebox.
4) Configure an SJI Jukebox.
5) Configure an STL Silo.
6) Configure a Microsoft Removable Storage Jukebox.

What kind of Jukebox are you configuring? [1] 2
14484:jbconfig: Scanning SCSI buses; this may take a while …
These are the SCSI Jukeboxes currently attached to your system:
1) scsidev@2.2.1: Standard SCSI Jukebox, DELL / PV-132T
2) scsidev@2.6.1: Standard SCSI Jukebox, IBM / 3573-TL
Which one do you want to install? 1
Installing ‘Standard SCSI Jukebox’ jukebox – scsidev@2.2.1.

What name do you want to assign to this jukebox device? pv132t
15814:jbconfig: Attempting to detect serial numbers on the jukebox and drives ..

15815:jbconfig: Will try to use SCSI information returned by jukebox to configur
e drives.

Turn NetWorker auto-cleaning on (yes / no) [yes]?

The following drive(s) can be auto-configured in this jukebox:
1> LTO Ultrium-2 @ 2.2.2 ==> \\.\Tape0
2> LTO Ultrium-2 @ 2.2.3 ==> \\.\Tape1
These are all the drives that this jukebox has reported.

To change the drive model(s) or configure them as shared or NDMP drives,
you need to bypass auto-configure. Bypass auto-configure? (yes / no) [no] yes
Is (any path of) any drive intended for NDMP use? (yes / no) [no]
Is any drive going to have more than one path defined? (yes / no) [no] yes

You will be prompted for multiple paths for each drive.
Pressing <Enter> on a null default advances to the next drive.

Please enter the device path information in one of the following formats:

\\.\Tape0 –for local path or
host:device-path –for remote node or NDMP device(s) or
host:drive-letter:directory path –for Windows disk file

Drive  1, element 256, system device name = \\.\Tape0,
local bus, target, lun value = 2.2.2,
ATNN=IBM     ULTRIUM-TD2     1110256727 model LTO Ultrium-2
Device path 1 ? [\\.\Tape0]
This path already exists in the database. Proceed with the same path?
(yes / no) [yes]
Device path 2 ? dsn01:\\.\Tape0
Device path 3 ? dsn02:\\.\Tape0
Device path 4 ?

Drive  2, element 257, system device name = \\.\Tape1,
local bus, target, lun value = 2.2.3,
ATNN=IBM     ULTRIUM-TD2     1110257072 model LTO Ultrium-2
Device path 1 ? [\\.\Tape1]
This path already exists in the database. Proceed with the same path?
(yes / no) [yes]
Device path 2 ? dsn01:\\.\Tape1
Device path 3 ? dsn02:\\.\Tape1
Device path 4 ?

Only model LTO Ultrium-2 drives have been detected.
Are all drives in this jukebox of the same model?  (yes / no) [yes]
A Dedicated Storage Node can backup only local data to its devices.
Should dsn02 be configured as a Dedicated Storage Node? (yes / no) [no] yes
A Dedicated Storage Node can backup only local data to its devices.
Should dsn01 be configured as a Dedicated Storage Node? (yes / no) [no] yes
Found device \\.\Tape0 already configured in NetWorker.
Delete existing definition? (yes / no) [no] yes
Found device \\.\Tape1 already configured in NetWorker.
Delete existing definition? (yes / no) [no] yes

Jukebox has been added successfully

The following configuration options have been set:

> Jukebox description to the control port and model.
> Autochanger control port to the port at which we found it.
> Networker managed tape autocleaning on.
> At least one drive was defined with multiple paths.  All such drives are
defined with a hardware identification as well as a path value to avoid
confusion by uniquely identifying the drive.  The hardware identification
for all drives which have one is always ‘autochanger_name – Drive #’ where
“autochanger_name” is the name you gave to the autochanger that was
just defined, and the # symbol is the drive number.
> Barcode reading to on.
> Volume labels that match the barcodes.
> Slot intended to hold cleaning cartridge to 22.  Please insure that a
cleaning cartridge is in that slot
> Number of times we will use a new cleaning cartridge to 50.
> Cleaning interval for the tape drives to 6 months.

You can review and change the characteristics of the autochanger and its
associated devices using the NetWorker Management Console.

Would you like to configure another jukebox? (yes/no) [no]

Note that even though we’d deleted the library from the NMC previously, the definitions for Tape0 and Tape1 were still present in the NetWorker configuration. I’m trying to recall whether if I deleted the library using jbedit the results would be the same.

Once you’ve setup the library using jbconfig, you should run jbverify to “check jukebox/device configurations in NetWorker”. This will identify whether you got the order of devices correct, particularly for the Dedicated Storage Nodes. Of course, one way to make sure you have everything matched up is to run inquire on each attached node prior to launching into a jbconfig session. But sometimes you don’t realise these things until you’ve done it a few times …