Nasuni Data Migration Service

Moving data using Nasuni's Data Migration Service

Nasuni's Data Migration Service is a robust data mover that works directly from the Filer. The service simplifies bulk data migration, automatically adjusting the ingest rate to expedite data transfer into Nasuni.

Step 1: Locate the menu item in the Filer called “services," then choose "Data Migration.” This is where you’ll start.

Step 2: The migration service supports multiple sources and targets for migrations. If you’re migrating from a number of sources, you can queue them up to run back to back. But let’s begin with the simplest case: a single migration. The first thing you'll need to do is add a migration source.

Step 3: Select a source to migrate. Your source can be a CIFS share from any file server with or without Active Directory. Migrating data from one Nasuni share to another is also possible. After entering your source information, click "Add."

Step 4: On the Current Migrations home screen Click "Add Migration."

Step 5: Fill in a descriptive name for the migration job. We'll call ours "Daily Numbers."

Step 5: Select the source that you would like to migrate. It can be the whole share (the root of the share) or any subfolder:

Now choose the destination volume. We recommend migrating to an unexported portion of your volume so users are not accessing/changing data as you’re migrating into that portion of your volume. You'll then pick a destination folder within that volume. Here you can choose from one of your existing folders or the root of the volume.

Click "Save." You'll see a dialogue box asking you if you wish to begin your migration, click "Create Migration" and your migration will automatically start. Please note: migrations will overwrite any files/folders at the destination path.

Step 6: Your migration will now begin. Progress is displayed on the "Current Migrations" screen.

Note: The migration makes two passes of the file tree. The first makes sure it can access the tree and calculates how many files need to be copied. The second is the actual process of copying the files from the source to the destination. During the second pass the migration service intelligently manages the cache space such that it does not stop because the cache becomes full. The result is that you can migrate terabytes of data into a Filer with a very small cache in a single operation. 

The migration service copies files. It does not move them, so your source is not modified during the migration process. 

After the first pass is complete and the migration begins, the estimated time to complete will be updated. This is a very approximate guess of how long the process will take based on the number of files to migrate, how many have been migrated at each point, and how long it has taken to migrate those files. As the migration progresses, this estimate becomes more and more accurate, but it can also oscillate a bit if your file sizes vary greatly.

Tip: While the migration is underway you can add additional migrations that will run when the first one finishes.

In addition to the progress bar, you can also view the migration log at any time. To do so, you either click through the “log” link on the status screen or navigate to a text file within the target migration directory of your volume. There you’ll find a file named something like “migration-log-2011-08-19_12398.txt”. You can view it with Notepad, Word, or other tools, and it will look like this:

This file is in a CSV format if you need to work with it, and it contains the detailed record of the migration: the source, the target, and the status of each file and directory involved.

Upon completion, you'll see a screen that looks like this:

And you’ll see new options: “remove” and “schedule.” The first, remove, forgets about the migration. Remove is appropriate for one-time migration events that you will not want to run again. Schedule is for re-running a migration. This may come in handy. If any errors happen during the first migration – if, for example, users change files on your source as you’re migrating - and we are not able to automatically recover, the log will indicate the errors. In this case, you might want to rerun the migration to try the copy again.

The second case where it may be helpful is if you’re periodically migrating data from one source to the Filer for backup/DR reasons. When you rerun a migration, the Filer intelligently migrates the data such that it will only copy files that have changed. It won’t start everything all over again. And it will also update any ACLs/permissions that have changed.

If you have feedback or questions about this guide, please email us – feedback@nasuni.com