Can I do this with NTFS volume mount points?

[Carson Dyle]

I have a home file server (Windows Server 2008 R2) that consists of 13 JBOD drives, each shared from the root folder with share names of D, E, F, G, etc.

I have each of these drives mapped to a drive letter on my desktop computer, but searching for files can sometimes be difficult. Could I use NTFS volume mount points on the desktop computer to give me what appears to be a single drive, so that I can use Windows Explorer to do file searches across all of the remote network drives?

 

[Zxian]

The Windows Search Service role in 2008 R2 should allow you to index all the drives on your server.

I believe that it should be plug-and-play with any Windows 7 or 8 clients.

I have a storage share that's mapped to Z: on my workstations. I can simply hit Win+F and type in a search string to find the files I want.

 

[Carson Dyle]

What exactly is a "storage share"?

 

[code65536]

No. NTFS reparse points (used for junctions and mounted volumes) is available only locally.

What you could do is mount them to a common directory on the server and then share that directory.

For example, my server has C:\Mount\D, C:\Mount\E, C:\Mount\M, C:\Mount\N, etc., into which I've mounted various drives. The Mount directory is then shared and accessible by my client computers.

 

[Zxian]

What exactly is a "storage share"?

It's a folder or volume (i.e. storage) that's shared.

On your 2008R2 machine, open up "Share and Storage Management" from your start menu. You'll be able to see all the volumes on your system and whether or not they have indexing enabled (which you can change via Right-Click, Properties, Advanced).

Any drives that show as indexed will be searchable from clients with Windows Desktop Search 4.0 installed (which I think is part of the regular WU packages).

 

[Carson Dyle]

No. NTFS reparse points (used for junctions and mounted volumes) is available only locally.

What you could do is mount them to a common directory on the server and then share that directory.

For example, my server has C:\Mount\D, C:\Mount\E, C:\Mount\M, C:\Mount\N, etc., into which I've mounted various drives. The Mount directory is then shared and accessible by my client computers.

Thanks. That sounds as though it would work.

 

[BonzaiDuck]

Maybe I misunderstand the issue or problem.

If you're happy with Win 2008 R2, you might want to look into software spawned initially to replace "drive extender" on Win Home Server v.1 when WHS-2011 was released (based on 2008 R2).

When it was announced that WHS-2011 would die without heirs, the add-in programmers probably already had their strategy in mind: support WHS-2011, Win Server 2008 R2, and all the later server OS's.

So, for instance, StableBit now comes in versions that also include applying the "drive-pool" concept to Win 7 and Win 8 as well as Server 2012, Essentials, SBS and WHS-2011.

If you ever thought that RAID left you "blind" when trying to recover an array, StableBit doesn't. Further, you can jump back and forth between StableBit v 1 and v 2 without mishap -- it is all managed quite well, and I proved it this last week. Even if drive labeling is inconsistent after an OS reinstall, either version of Stablebit will rebuild the drive pool without a hitch and without regard to what the drives were labeled before and after. (I also proved that in the last week or so.)

The price? Depends on your use. But I think you're talking about $25 or so, and the license will extend to the same hardware with changes that may even include processor replacement.

During these "tribulations" I mentioned obliquely, I bricked a good WHS server install on an SSD; then I bricked the HDD from which the SSD had been cloned; and then I bricked a drive image made with Acronis TM of that HDD. I finally just reinstalled the server OS from scratch.

But there was no problem with the data drive pool array. I had selected certain folders to be auto-duplicated by StableBit, but the duplication wasn't necessary for the current crisis. Moreover, both the old version and the new one recognized the selective folder duplication! They swiftly re-balanced the allocation between disks, and all was wonderful.

Also, you may be interested in "StableBit Scanner" which is a SMART monitoring and analysis software for all disks on your system, only including the drive pool array. Another $25.

 

[Carson Dyle]

Maybe I misunderstand the issue or problem.

I think you did.

 

[Zxian]

Thanks. That sounds as though it would work.

It still won't give you indexed results, if that's what you're looking for.

 

[Carson Dyle]

It still won't give you indexed results, if that's what you're looking for.

"Indexed" results? Do you mean the indexing of file contents? I don't have much need to search file contents, only search through file names.

Another thing I'd like to make easier is being able to run scripts or batch files starting at the root, across all drives.

 

[BonzaiDuck]

I think you did.

I think I see what you did there. I just don't understand why you chose to share each physical drive volume separately with a different "letter" to your client workstation. You might have spanned those volumes, then shared folders on the single resulting logical drive.

Or as I said, they could be in a drive pool, and you'd share folders from there.

But if I still don't see what you were doing, you don't need to take time to educate me about it.

 

[Zxian]

"Indexed" results? Do you mean the indexing of file contents? I don't have much need to search file contents, only search through file names.

Another thing I'd like to make easier is being able to run scripts or batch files starting at the root, across all drives.

By indexed, I mean that the search is handled by the indexing engine within Windows - the same indexer that gives you the search for items in your start menu and local files. If you've got files and folders in your user profile, they're indexed by default and easily searchable from the Start Menu or Win+F.

I'll give you my home setup example. I've got my storage array shared from my server and mounted on my workstation as Z drive. The server has the contents of this volume indexed (just like your user profile would be by default on Windows7/8). When I search for "The Fifth Element" from my workstation, the folder and MKV file stored on my server appear in the search results. The search query is passed from the client to the server, which then searches the index. The results are then passed back to the client machine and displayed on screen.

This happens to also index the file contents, but you do get filename based search along with that. Any non-indexed searching is going to be fairly slow on network drives.

As for your scripting situation, the python in me says "for drive in drives"...