Mapping a drive

I've always wondered what is the difference in terms of mapping a non-persistent drive from the run command and mapping a drive from Windows Explorer? Even mapping from the same computer to a remote folder, it seems like mapping a drive from the run command works about twice the amount of time. I thought both methods use SMB/Netbios, but there must be some difference in the underlying authentication...

Thanks!

 

I believe the underlying process is much the same. You get more options from the command line (for example, being able to map as an alternative user), but I'm surprised that you get a performance difference once the mapping is established.

By the way, I'd suggest that you do not map a non-persistent drive unless you have a very good reason to (for example, a program that requires access via a mapping)

When you map a drive Windows will keep that connection alive and search ahead for permissions and other background tasks. These processes mean that accessing a mapped drive is usually quicker than accessing via the share name (where authentications and permissions have to be established every time you connect).

However, if you map to a share that you cannot connect to via the network, Windows can spend a lot of effort trying to maintain the non-existent connection. This can cause significant performance reduction to Windows Explorer and other application windows that use the Windows Explorer.

If you only connect to a non-persistent drive occasionally, consider creating a shortcut in Network Neighbourhood rather than a mapping.

 

The credentials are passed at the time the session is created. There is no difference between a GUI non-persistent MAP and a command line NET USE. Nor is the session re-authenticated during its use; the token is valid for the life of the session.

 

Issues...

Thank you...Can someone identify what type of issue(s) would allow authentication using one method and not the other. This is a common issue I have in several of my LAN segments.

Thanks!

 

What do you mean "issue" here ?

 

Issue with issue


Jason,
Sorry for being vague. By "issue ", I mean for example: Type of user account on the machine establishing the mapped drive, association to workgroup/domain, use of DNS server, etc. In other words, assuming the same exact machine is used in the same environment, mapping to the same remote folder/drive, are there elements of a network or configuration settings on the machine originating the mapped drive that would create different performance/success vis-a-vis the networking of drives using non-persistent run command or GUI-based WExplorer? Again, it seems like regardless of these different network environments or even user accounts, I almost never have problems mapping drives using the run command, but on the same machine, to the same shared drive/folder, I won't be able to map using the GUI-based WE method. This has also been the issue between W2k, XP, 2k professional and 2k3. So, I didn't think it was OS related... THANKS.

 

No.

The only difference that comes to mind is that the GUI is going to, unless you take the extra steps, use your cached credentials in Protected Storage for the connection.

(As will NET USE if no Domain/username/password is passed)

If you are used to passing username, Domain, and password credentials when using NET USE, these may in fact differ from what the user has on their keyring in protected storage: http://support.microsoft.com/?kbid=281567

As to network speed or other considerations, there are none between NET USE and the GUI mapping methods. In either method the authentication token is valid for the length of the logon session by the user: http://support.microsoft.com/kb/281249/