about setting "profile" and "home directory"

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

here are two domain servers, how can I set profile and home directory for user so that when anyone of domain server downs, user still can get his profile and home directory from another domain server.thanks.

 

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

Hi,

could you clarify your question a bit further please?

Are both of these servers DC's or is one of them a member server that holds the user profiles and paths?

 

If they are both domain controllers in the same domain, then they will replicate active directory information between themselves. Unfortunately they do not replicate every folder on the server, only the "sysvol" folder. As the users folders don't reside in this folder they won't get replicated.

User profiles are not replicated between domain controllers because of the huge size involved. It would play havoc with replication in my opinion.

So in answer to your question, I don't think it is possible for the reason above and other reasons like the path to the home folders for example is fixed to a specific server.

 

you could always use a dfs share to store the profiles

 

That's a thought. Have you tried that Clyde?

 

thanks for your reply, I guess people use other way to solve this problem in real world.

 

no, it just seemed a good solution

 

There could be many different ways to solve your problem, however in order to help you further we need to know more about how you currently have the servers setup and what roles they perform.

 

Depending on the number and size of peoples home drive you could either use an external drive to store the home drives on or a NAS box if you have the cash.

For the profiles you could use roaming profiles but I think if the DC went down they would be useless.

 

my question is not good, actually I mean what solution can make users still can access their profiles and home folder when one of server ,which stores user's profile and home folder, is totally crashed.I thought this is common issue which administrators need concern.

I want to know how administrator deal with this problem in real life, for example email users "server is crashed, we need restore data for users, it may take one or two days.thanks for your patient". is this a acceptable solution in real life?

 

For that situation there are a number of strategies you can implement.

You could have a SAN, which stores the data across multiple arrays of disks.

You could implement DFS as Clyde suggested, this works simillar to a SAN in that data is stored in numberous places across the network.

You could choose to have a 'spare' online server which contains a backup of all user data and switch the paths over to the spare should the primary server go down.

These are just the three that come up off the top of my head.

 

thanks, which way is more practical or popular?

 

To be honest I think that it all comes down to cost.

SAN's are expensive to implement.
DFS can be cost effective, but only if you have enough servers to spread the data across them, and disk space!

Personally I am looking to implement simply a spare server in my organisation to act as a backup DC and to store a copy of the data, so that in the event of my DC or file server going down I am easily able to keep the network up. Yes it will not run as fast (the spare server will basicaly be a converted workstsation) but it will be up.

 

Technically your DC should never really go down. For DC you tend to use redundancy techniques like RAID, Two NIC's etc.

You could instead set up a 2nd server to heartbeat the first one meaning that if the DC goes down the heartbeat server takes over. With this setup when something is copied to one server it's copied to the heartbeat server.

There are bascially lots of different ways to do it but it depends on budget and in the real world like the company I work for we simply say to the user sorry but it knacked and we will fix it in however long it takes.

 

Hi all

My preference would be to not store profiles and user data on a domain controller. As long as we have at least one DC up our users will be able to locate the path to their data via the settings on the user object in AD users and computers, it might be a little slower if we have to go to a DC in a different site but we will still get the required path.

As far as the physical data is concerned if we need quick failover I would be looking at placing the information on an active passive cluster so if one node fails we can automatically failover and the users can get their data without interruption.

As mentioned SAN's and DFS are other options it comes down to the acceptable downtime Vs cost of proposed solution.

 

And of course if you go down the DFS route you'll have a lovely sofa to sit on in the server room and not have to pay anything till January 2007!!!!

We are talking about the same DFS aren't we?

I'll get my coat ....