I have two computer, the first one is the server (so WinXP) with a shared folder granted for user “test” and this is out of domain, it is just in WORKGROUP.
The second computer is a client computer (WS2012) in domain and i would like to access directly to the shared folder of the server without prompting username/password.

I tried to add a “test” user (with the same password) in localuser of the client and also in the domain, but the access fails.

I need this on.the.fly access because i need some file from share just with a file path
(ex. \server\path\file.txt) without prompting.

thank you.

You have a few options.

1. Create a local account on each machine with the exact same username and password. Have your batch job run as this account. Give this account permission to access the share. This is called pass-through authentication and it’s an ugly hack and holdover from the early days.

2. Give Everyone access and enable anonymous authentication. This is extremely insecure and you shouldn’t do this either.

3. The “correct” way to handle this is to join the server to the domain and use a domain account that has access to the share to run the batch job in question.

Check more discussion of this question.

I am trying to copy a 1.2Gb file between a Windows 7 workstation and a Windows Server 2003 system. That same server also hosts our enterprise’s shares, most of them anyway.

The copy from the workstation to the server times out. If not in the event logs, where might this timeout event be recorded? Furthermore, what tools exist, other than doing something like running Wireshark, that would let me see the details of the failure?

No, this sort of thing won’t be in the event logs; however, if the root cause was something (like a NIC losing connection or failing hard drive) that would be recorded, you could see if there’s anything correlated.

If the problem is being caused by the network, and if you have managed switches (assuming this is all on the same LAN), then you might see some errors on the port(s). That might be similar to what you could see in Wireshark from one host or the other, if you know how to read its output.

How are you doing the copy? Explorer drag-n-drop, or copy, xcopy, robocopy? I prefer to use the last one, because it has decent error codes, retry options, and logging. Try it with robocopy and see what results you get.

Check more discussion of this question.

For my company, I would like to create a single directory for each user in a server. Each directory must be personal, and I would like to create a desktop link to that directory. For example, in my server SERVER01 I need a simple structure like

users\johnsmithusers\robertgreenusers\johndoe

Each directory must have a limit of 5 gigabytes and must be accessible only to the single user and to the domain administrator. How to do in Active Directory? With a script?

You should look into using Folder Redirection. This way, your users will be able to use their normal folders like Documents and Desktop, and it will be transparently mapped to the folder on the server whenever they log in. It’s a part of the Group Policy settings in Active Directory.

Check more discussion of this question.

On Windows Server 2008 R2, when I run the following command I am getting the 1920. I’ve tried pretty much everything I am aware of and I can’t figure out what causes the error. When I try to map the same network share using the UI and same credentials, everything works fine.

net use * \\EAAA-12345\C$\ /USER:\\EAAA-12345\Administrator Passw0rd /PERSISTENT:NO

Anybody knows how to get rid of the 1920 error?

Remove the back-slash after the C$ and remove the back-slash between USER: and EAAA-12345.

So it should look like this:

net use * \\EAAA-12345\C$ /USER:EAAA-12345\Administrator Passw0rd /PERSISTENT:NO

Check more discussion of this question.

I’ve setup a server which I’ve connected to multiple PCs in my workplace. Sadly, data transfer speeds are at max 3 MB/sec per connection which works out slow for file transfers, especially when transferring large files. I’m using Windows filesharing and the server is a Windows Server 2008 (2 Ghz CPU, 1 GB RAM) and the client PCs mostly running Windows 7. How can I detect bottlenecks in my network and improve file sharing speed within the network?

Hoping the Windows Server 2008 is “Server 2008 r2″. Can understand a limitation of moving to “r2″ requires 64bit OS & CPU. Immediately noticing server has 1GB RAM, which would be low for a server. 2GHz CPU is ambiguous, for instance a 2GHz P4 versus a 2GHz Xeon E5-2650 would be a 50 fold performance difference. State of storage is not known, such as where the OS and fileshares reside physically and logically to the server.

From Microsoft:
http://technet.microsoft.com/en-us/windowsserver/bb414778.aspx

Hence, main concern would be the health and viability of the server before moving to physical and logical network components (switch, cables, VLANs, QoS, bandwidth, noise, etc).

Performance tuning instantiates with obtaining a local server baseline. To get started, and better form a basis of the sever situation, do check these solid guides for Server 2008′s own Performance Monitor. It is a big topic, an overview:
http://www.petri.co.il/performance-monitoring-easy-way-part-1.htm

http://www.petri.co.il/performance-monitoring-easy-way-part-2.htm

http://www.petri.co.il/performance-monitoring-easy-way-part-3.htm

http://www.petri.co.il/performance-monitoring-easy-way-part-4.htm

Check more discussion of this question.

From a windows208r2 client connected to a SMB/CIFS share, I move a file from one directory to another directory on the same network share.

Does windows send a “move” command to the share, or does windows copy the bytes locally, write them to the new location on the same share in the new location, and then delete the original file?

One would be much more efficient than the other, just don’t know how “smart” the client/server is for this type of share.

The short answer is: Meh.

On a file “Move” (AKA Cut and Paste) between the same share it will simply change the index on the file server (pretty much instant).

Unfortunately when you try to move between two different shares (Even on the same file server) or do a “copy” operation it does pull the data through the client PC (slow).

On any “move” operation is does copy each file before deleting the source file, so at least that’s good =)

Easy to test all of this by connecting to a share with a large folder and trying it out.

Check more discussion of this question.

if I’m accessing a local file via UNC path, for example //hostname/abc.txt, Will the traffic still go through the switch or it’s clever enough to figure out this is a local machine, i.e. will there be a significant speed impact if I’m accessing a file via a UNC path rather than its local path (d:\abc.txt, for example)

Thanks.

File sharing is based on ordinary TCP/IP, so the rules are no different than those that apply to other network applications. Hostnames from UNC paths will always be resolved first using DNS or NetBIOS and then the traffic will sent out based on a routing table. Since it will resolve to a local address, the traffic will never go out, unless you have some very strange rules on a machine (I am not sure if it is even possible to create such a setup).

As for the speed, I am not so sure if it is smart enough not to use TCP/IP (even if packets never leave the machine) when connecting to a local address. If it does use TCP/IP, local file access will be faster as it does not have the network socket overhead. You can test this with a packet analyser, like Wireshark (see this link to see how to set up a capture on a local interface).

Check more discussion of this question.

Scenario is a small network with a Server 2008 R2 running AD, DHCP, DNS, Print & File Sharing.

We’d like to build a backup server to continue operations uninterrupted should the main server go down. Configuring DHCP, DNS and AD is easy.

The question is how do I configure a ‘backup’ Network Drive, redirected users folders, and printers. Is there any MS documentation on this? All my Google hits are just on backup software.

Currently all our UNCs are \Server\resource . Is there a tool / MS component to create a Virtual Storage Pool that is synchronized across two (or more) servers, and our UNCs could be pushed out as \VirtualPool\resource ?

TIA

You can use a DFS namespace to obscure the path to the share so that it would appear as \\domain\users instead of \\server\users. You’d use DFS-R to replicate the shares between the two servers and then add them both to the namespace.

For printers, there’s not a lot you can do other than using group policy to map them and change that policy to the other server in the event of a failure (unless you get dedicated print servers and cluster them).

Check more discussion of this question.

Until today, we had a 2003 DC, all my folder mappings were done in the login script. (net use…)

we have now moved to 2008 R2, and I would like to create/manage the new mappings “The easy way”

If I have several mappings, to several groups (some are Read Only/ others are Read/Write)
What will be the easy way? Should I create a single GPO per OU (each of our departments has it’s own OU) – and add the drive mappings there (NOT folder mappings) ? *there is no reed/write option

Or should I keep managing that from the share folder properties itself? (right click on the share name on the NAS)

First things first, all security should be managed using NTFS permissions on the folder and share (Remember, these are 2 sets of permissions and are mutually exclusive) itself. To do this you’ll need security groups, and there is no way round that. Not mapping somebody to a drive is not ‘security’.

Secondly, to MAP the drives, I’d be using Group Policy Preferences. As above, it won’t handle the file and folder permissions, but it will let you map drives very easily and cleanly.

Finally, the beauty of Group Policy Preferences is that if your environment suits it, they will allow to you use 1 single GPO to map many different shares for many different users, using item level tar targeting to filter based on OU, Security Groups etc.

Seriously, if you’ve just moved to 2008 (And I kind of hope you meant 2008 R2 seeing as 2008 is already now 2 editions old) then read up on Group Policy Preferences. Oh, you may need to patch any XP / 2003 clients to get them to work.

Check more discussion of this question.

The company that I work for wishes to replace its windows servers with new linux servers but wants to keep Windows on employee desktops. The requirements are:

• The employee desktops should be able to authenticate with a central linux server.
• The employee desktops should have access to shared folders on a central linux server.
• Software applications should be able to use Active Directory.

My first thought was to use samba, but can samba meet the above requirements, if so, in what way? (details are appreciated, but I just need the general idea.)

Keep in mind that I am not very familiar with server setup (I am a web designer by profession).

In my workplace, we switched from Windows shared folder for network use to Ubuntu server for our 4 disc hardware raid. We do not use Active Directory (and I don’t know what it is) but we do use user permissions to access the raid and mount it as a network drive. You can install webmin and use that to help manage the server.

There are a couple tricks: setting up the user permissions and umask in smb.conf and then getting the shared folder’s permissions and “sticky bits” set correctly.

Check more discussion of this question.