登入
 
首頁文件庫學習園地下載支援社群論壇
IT 專業人員的技術資源 >
> 
> 
How to migrate SQL server from one node to another?

已答覆 How to migrate SQL server from one node to another?

  • 2010年5月29日 上午 02:51
     
     
    登入以投票
    0

    Background:

    There are two nodes for the application server, and two for the database server (SQL server 2008 Standard).

    Each OS is Windows Server 2008 DC. The SQL nodes are writing/reading from a SAN.

    I want to move the SQL Servers from Nodes 1 and 2 to a less powerful machine.

    How do I do this in the most effective way?

    Part of me wants to just install SQL on the less powerful boxes and restore a backup to the engine.

     

     

所有回覆

  • 2010年5月29日 上午 09:16
    版主
     
     
    登入以投票
    0

    Is this an actual Failover Cluster(s) ?

     

     

     
  • 2010年5月30日 上午 01:46
     
     
    登入以投票
    0
    Sorry, yes; this is an actual failover cluster. There are two nodes for SQL, one primary and one failover. The database itself is stored on a SAN.
     
  • 2010年5月31日 上午 08:00
     
     提議的解答
    登入以投票
    0

    I am not sure if I understand, but it sounds to me like you want to move from the failover cluster to a different member server (not a cluster node).

    If so, then it should be simple enough to just remove the SQL from the failover cluster and then to attach the database to the new node.

    Good luck.

    Russ Kaufmann, MVP, MCSE: Messaging and Security, MCT, MCITP, MCTS and other stuff
     
  • 2010年5月31日 下午 03:33
     
     
    登入以投票
    0

    Actually, let me describe it a little bit better. There are four nodes in the cluster, nodes 1 and 2 are application servers (SharePoint) while nodes 3 and 4 are database servers (SQL).

    The database servers are more powerful than the application servers and I wan't to switch them: I want to move SQL to the less powerful nodes (1 and 2) and SharePoint to the more powerful nodes (3 and 4).

    It seems to me it would be as simple as installing SQL on nodes 1 and 2 then just attaching the old database to the new engine.

    Same with moving SharePoint, just reinstall on nodes 3 and 4? No?

     
  • 2010年6月1日 上午 05:09
     
     已答覆
    登入以投票
    0

    Actually, let me describe it a little bit better. There are four nodes in the cluster, nodes 1 and 2 are application servers (SharePoint) while nodes 3 and 4 are database servers (SQL).

    The database servers are more powerful than the application servers and I wan't to switch them: I want to move SQL to the less powerful nodes (1 and 2) and SharePoint to the more powerful nodes (3 and 4).

    It seems to me it would be as simple as installing SQL on nodes 1 and 2 then just attaching the old database to the new engine.

    Same with moving SharePoint, just reinstall on nodes 3 and 4? No?

    OK, now I get it.

    First off, SharePoint is generally considered to be two components, the Web Front-End and the SQL back-end. SPS is usually configured for HA by using NLB or a hardware load balancer that manages the connections between the front-ends. So, it really doesn't make sense to have SPS front-ends in a cluster.

    Second, the process for adding SQL to the two other nodes is based on the version. What version are we talking about?

    Russ Kaufmann, MVP, MCSE: Messaging and Security, MCT, MCITP, MCTS and other stuff http://infrastructurehelp.spaces.live.com
     
  • 2010年6月6日 下午 04:49
     
     
    登入以投票
    0

    SQL Server 2008 Standard.

     

    Actually, I was mistaken about the architecture. The correct architecture is as follows:

    SQL Server 2008 Standard on Node 1 and Node 2, both on a FC SAN.

    WSS/MOSS 2007 on two seperate servers with NLB. Note: the SP installation one one box is "physical" while the other two are virtual via hyper-v on second box.  

    I want to completely change the architecture in a different way now. I want to put SQL Server 2008 on a virtual machine via hyper-v, attach the existing databases, run setup, and point the WFE's to the new database server. Is that right?

    I also want to install one or more WFE/s on vm's. The way I understand it, I am basically doing an advanced setup where I'm adding to the farm rather than creating a new one. 

    Ideally I would like to put the SQL and WFE on the same VM for simplicity. Our server manufacturers documentation recommends having multiple WFE's, one for each 300 users. Personally I think this is BS.

    All the books I have read state that the NOPS formula is the best method for determining whether to have a stand alone installation or a farm, and if a farm how big of a farm is needed. Based on the NOPS formula, I could support many more times the number of users I have now.

    http://office.microsoft.com/en-us/sharepointportaladmin/HA100906131033.aspx

     
  • 2012年5月4日 上午 11:12
     
     
    登入以投票
    0
    Windows 2003 R2 Ent. 32bit cluster servers + storage moved to new hardware but not able to get online the 2nd node. help please

    Windows 2003 R2 Ent. 32bit is on cluster servers and storage which copied/image to new hardware, (took a image backup from old servers and restored the image to new hardware and storage. but only the last active server is working after some exercise but another server is not able to get online. The core application service (SQL is getting down in a second and moving back to the previous active server). Anybody faced this beofre or can any body help please? 

    Event ID:
    1071
    1069
    19019
    4185
    1069