locked
Problems with DFS Namespace Server Failover RRS feed

  • Question

  • Curious to know if someone has seen my DFS Namespace failover issue and might know of a resolution. I have found a couple of posts of folks with the same problem, but there never seems to be a resolution posted.

    I have two Server 2008 R2 Domain Controllers in my environment that are in the same site. They are both hosting DFS Namespace, no DFS Replication is setup. When I take a DC down for maintenance, some of the clients on my network lose the ability to get to the DFS Rootname. The clients are Windows 7 Pro machines that are all patched up.

    When I run dfsutil /spcinfo on them, the results are confirming that the client is holding onto the downed server as the "Active" server. 

    So how do I get my clients to automatically release the downed server, and use the DFS server that is still up for it's DFS communication needs? Preferably without rebooting or manually resetting cache? Those things just seem like fixing symptoms instead of the problem. 

    If it is of any consequence, WINS is not in use, and the clients and servers are on separate VLANs.

    Any assistance would be greatly appreciated, as I have been banging my head against the wall for awhile on this one.

    Thursday, April 25, 2013 3:53 PM

Answers

  • Curious to know if someone has seen my DFS Namespace failover issue and might know of a resolution. I have found a couple of posts of folks with the same problem, but there never seems to be a resolution posted.

    I have two Server 2008 R2 Domain Controllers in my environment that are in the same site. They are both hosting DFS Namespace, no DFS Replication is setup. When I take a DC down for maintenance, some of the clients on my network lose the ability to get to the DFS Rootname. The clients are Windows 7 Pro machines that are all patched up.

    When I run dfsutil /spcinfo on them, the results are confirming that the client is holding onto the downed server as the "Active" server. 

    So how do I get my clients to automatically release the downed server, and use the DFS server that is still up for it's DFS communication needs? Preferably without rebooting or manually resetting cache? Those things just seem like fixing symptoms instead of the problem. 

    If it is of any consequence, WINS is not in use, and the clients and servers are on separate VLANs.

    Any assistance would be greatly appreciated, as I have been banging my head against the wall for awhile on this one.

    There are many issues with DFS failover. Please follow this link on when it happens and when it does not:

    http://help.globalscape.com/help/wafs3/using_microsoft_dfs_for_failover.htm

    What Happens During Failover

    In DFS failover, clients attempt to access another target in a referral after one of the targets fails to respond or is no longer part of the namespace. Clients must access a domain-based namespace by using the format \\DomainName\RootName. If a client accesses a domain-based namespace directly on the root server (\\RootServer\RootName), root target failover does not occur. DFS failover is only performed when a client opens a file or folder. If a client has files or folders open and attempts to read or write to them when the target server is unavailable, the application will receive a failure on that operation.

    Switching Between Replicas During Failover

    Referrals are cached locally to maintain performance, and if replicas are available, all replicas are provided to the DFS client. The client chooses which referral to use as a failover and preference is still given to replicas within the same site as the client. After a referral is selected, a session setup is performed (credentials are passed to the new server if a prior connection does not exist). If the selected referral fails, a failover process begins. The speed and implications of the failover depend on what the client was doing at the time of the failure, how the failure occurred, and how tolerant of delays an application is.


    StarWind iSCSI SAN & NAS

    Thursday, April 25, 2013 9:17 PM
  • Hi,

    Please see if this article could help:

    How to configure DFS to use fully qualified domain names in referrals
    http://support.microsoft.com/kb/244380/en-us

    TechNet Subscriber Support in forum |If you have any feedback on our support, please contact tnmff@microsoft.com.

    Friday, May 3, 2013 8:52 AM

All replies

  • My understanding is that DFS can't be used for the high availability scenario you are trying to achieve. You'll need File Server Cluster to do that.

    http://technet.microsoft.com/en-us/library/cc731844(v=ws.10).aspx

    Thursday, April 25, 2013 4:16 PM
  • Ben,

    Thanks for taking the time to help out. I am skeptical that DFS doesn't allow the Namespace data to be hosted on two different servers. Everything I have read has lead me to believe that they have the ability to do this to some degree, especially if DFS-R is in use, and the clients have to go find another name server if the closest one goes down. 

    Keep in mind that I'm not trying have redundancy with the actual shares themselves, just the Domain-based DFS namespace data. I am reading that is supposed to be stored in AD, so I am thinking that one of the AD servers should be able to provide that information to the clients if one DC goes down. It just seems that some of my clients get stuck on the DC that I take down.

    Thanks again for any help.

    Owen

    Thursday, April 25, 2013 7:10 PM
  • Curious to know if someone has seen my DFS Namespace failover issue and might know of a resolution. I have found a couple of posts of folks with the same problem, but there never seems to be a resolution posted.

    I have two Server 2008 R2 Domain Controllers in my environment that are in the same site. They are both hosting DFS Namespace, no DFS Replication is setup. When I take a DC down for maintenance, some of the clients on my network lose the ability to get to the DFS Rootname. The clients are Windows 7 Pro machines that are all patched up.

    When I run dfsutil /spcinfo on them, the results are confirming that the client is holding onto the downed server as the "Active" server. 

    So how do I get my clients to automatically release the downed server, and use the DFS server that is still up for it's DFS communication needs? Preferably without rebooting or manually resetting cache? Those things just seem like fixing symptoms instead of the problem. 

    If it is of any consequence, WINS is not in use, and the clients and servers are on separate VLANs.

    Any assistance would be greatly appreciated, as I have been banging my head against the wall for awhile on this one.

    There are many issues with DFS failover. Please follow this link on when it happens and when it does not:

    http://help.globalscape.com/help/wafs3/using_microsoft_dfs_for_failover.htm

    What Happens During Failover

    In DFS failover, clients attempt to access another target in a referral after one of the targets fails to respond or is no longer part of the namespace. Clients must access a domain-based namespace by using the format \\DomainName\RootName. If a client accesses a domain-based namespace directly on the root server (\\RootServer\RootName), root target failover does not occur. DFS failover is only performed when a client opens a file or folder. If a client has files or folders open and attempts to read or write to them when the target server is unavailable, the application will receive a failure on that operation.

    Switching Between Replicas During Failover

    Referrals are cached locally to maintain performance, and if replicas are available, all replicas are provided to the DFS client. The client chooses which referral to use as a failover and preference is still given to replicas within the same site as the client. After a referral is selected, a session setup is performed (credentials are passed to the new server if a prior connection does not exist). If the selected referral fails, a failover process begins. The speed and implications of the failover depend on what the client was doing at the time of the failure, how the failure occurred, and how tolerant of delays an application is.


    StarWind iSCSI SAN & NAS

    Thursday, April 25, 2013 9:17 PM
  • Hi,

    As you said, if a server get down when a client is already connected to a folder target, as the cache is already created, it will take some time before clients starting finding a new target. You could set the TTL to a smaller number but it cannot help avoid the issue.

    The following 2 articles provided more detailed information:

    http://blogs.technet.com/b/filecab/archive/2006/01/20/417832.aspx

    http://blogs.technet.com/b/askds/archive/2011/08/19/3448148.aspx#dfsnspeed


    TechNet Subscriber Support in forum |If you have any feedback on our support, please contact tnmff@microsoft.com.

    • Proposed as answer by VR38DETTMVP Monday, April 29, 2013 4:36 PM
    Monday, April 29, 2013 7:00 AM
  • Thanks for the great reads Shaon Shan and VR38DETT. I now understand that there isn't going to be an immediate failover, and that caching occurs for a period of time. However, I have looked at the Cache Duration for the referrals, and I am set to the default TTL value for root referrals at 5 minutes and for link referrals at 30 minutes. So everything looks good there.



    Here is where something isn't working quite right. What I am struggling with is the fact that when I am taking my backup DC down, I have it in a down state for as  as 7 to 8 hours. And the clients are still hanging on to the downed server deep into the downed state, way beyond those TTL values. It is as if the clients aren't abiding by those TTL values for some reason, and are keeping the downed Namespace server cached way beyond the TTL values and refusing to refresh to the Namespace server that has been up the entire time.

    Based off of some of the articles provided, I noticed that NETBIOS is the default protocol used by DFS instead of DNS. Being that I have the DFS Clients and the DFS servers on a different VLAN and subnet, do you think that there is some necessary DFS NETBIOS communication that is being dropped?


    Tuesday, April 30, 2013 2:18 PM
  • Hi,

    Please see if this article could help:

    How to configure DFS to use fully qualified domain names in referrals
    http://support.microsoft.com/kb/244380/en-us

    TechNet Subscriber Support in forum |If you have any feedback on our support, please contact tnmff@microsoft.com.

    Friday, May 3, 2013 8:52 AM