FAQs - Using ArchiverFS With A Distributed File System
Home | Pricing | Download | Documentation | Release Notes | Customers
Compatability:
Just as with all fundamental Microsoft technologies, ArchiverFS is 100% compatible with all versions of DFS. Due to the way ArchiverFS archives at the file system level and the fact that it doesn't use any client side software or low level file system drives, it integrates seamlessly with DFS.
Archiving from a DFS share \ namespace:
When creating an archiving job that targets a DFS share \ namespace as a live location to archive files from, it is important to consider how ArchiverFS works e.g. everything is based on UNC paths.
If you add the UNC path to an underlying physical replica as a live location instead of the domain based UNC path then all archiving activity for that location will use the underlying UNC path of the replica.
A such it is important that you always use the domain based path. If you use a UNC path to a local server in a DFS setup with multiple real servers \ replicas supporting a domain based DFS root, then when files are archived they will be recorded as having been moved from the specified local server UNC path and not the domain based DFS root path. This can cause multiple undesirable behaviors.
So remember, when creating an archive job that targets a DFS namespace:
\\Domain\DFS Root\Folder (optional)
\\Server\Share\Folder (optional)
A note about replication:
If you are using DFS then there is a good chance you are using replication, and if you are then you are probably servicing more than one geographical location with your file system.
When you run your first archive job you will probably archive a significant proportion of your file system to archive storage.
If this is the case then pleae consider that DFSwill have a lot of data to replicate. As such when running the first archive job on a DFS based file system we recommend you:
- Disable the maintenance job.
- Consider doing the initial archiving in stages e.g. run the archive job with it set to archive all unused files over 10 years old, then once that has completed re-run it set to archive everything over 9 years old and repeat gradualy working down to the desired file retention age.
- Keep an eye on your DFS replication queues.
- Once you have reached the desired file retention age and your DFS replication queues have cleared, enable the Maintenance job.
A DFS based destination:
If you have multiple geographical locations to service and they all access a single DFS root that is replicated across all locations, then you might consider archiving to a DFS root and replicating it to all of your sites.
While this would require second line storage to be available at all locations, it means that when users access migrated files via stubs\links their request will be served via the local network and not over the WAN. Depending on the speed of the WAN connection and the size of the file this can speed up the retrieval process.
Further information:
If you have any questions please don't hesitate to contact us via [email protected], we would love to hear from you!
The internet is full of guidse and videos that explain how to set up DFS. In adition to them, we would recomend taking a look at some of Microsoft's official documentation: https://docs.microsoft.com/en-us/windows-server/storage/dfs-namespaces/dfs-overview