.NET Tutorials, Forums, Interview Questions And Answers
Welcome :Guest
Sign In
Win Surprise Gifts!!!

Top 5 Contributors of the Month
david stephan
Gaurav Pal
Post New Web Links

Is a 3 Node Cluster the way to go for a high availability environment?

Posted By:      Posted Date: May 22, 2011    Points: 0   Category :

Good Morning,

My company currently has a 2 Node SQL 2008 Active / Passive cluster. Due to the sensitivity and ugency of our industry, we have to maintain a 100% uptime for our clients. The issue here is that we cannot do any maintenance that involves failovering the cluster (patching, SPs, etc), because of the time it takes to failover (even a few seconds is critical).

We are considering going to a 3 Node Active / Active / Passive cluster, with the theory that since one Node will always be active, we can failover to the passive Node without any disruption. However, we're not sure if this would be the case, and if it isn't, would there be something we can do instead to keep uptime at 100% w/o the risk of our environments being outdated?

Thank you all in advance,

Nick Gargano, Sr. DBA

View Complete Post

More Related Resource Links

Setting up 2 node Active-Active SQL cluster as High availability and database mirroring at database

Hi guys
Please need your inputs and suggestions. I have always worked as junior DBA and have setup the HA but never planned for it. Please need some support here.

We have one main application: Application1 ...which is the sending data to all other application for billing, reporting, processing the orders and other stuff. This application is working right now on two Server.

Server1:  SQL Server 2005 Enterprise Edition  (Total 16 Gb RAM , SQL Server ( 4-12 GB AWE Enabled) and  2Processors (4 core each) and 836 GB of HD )

WCF service with WAS and High Availability Cluster on Server 2008


I have a WCF service using one-way net.msmq transport to deliver messages to a high availability cluster.

I have followed a few articles I have found on setting up WCF in a cluster environment and essnetially have the following queues


Cluster: private$\MI/ManagementInformationService.svc

Node A: private$\MI/ManagementInformationService.svc

Node B: private$\MI/ManagementInformationService.svc

Although I am not sure why I need the local queues.

I've installed the WCF application into IIS7 on Node A and Node B. I;ve configured both nodes to the following address:

address="net.msmq://{cluster endpoint}/private/MI/ManagementInformationService.svc"

So, I configure my Web Servers to deliver the MI to the clustered endpoing via WCF net.msmq binding.

At this point, everything works and as messages hit the clustered queue, they are picked up by one of the nodes and processed. However, when I fail over the cluster, the second node does not automatically spring into life, as I would expect it to with WAS. However, if I browse the WSDL for my service on the now-current node, then it does indeed spring into life and process the messages off the queue.

So I turn on WCF tracing, IISReset fire up the WSDL, IISReset (to flush the WCF Trace log) then see the following:

A warning: Cannot detec

Cloud Computing: Patterns For High Availability, Scalability, And Computing Power With Windows Azure


Here we examine the typical cloud platform architecture and some common architectural patterns, along with their implementation on the Windows Azure offering from Microsoft.

Joshy Joseph

MSDN Magazine May 2009

High Availability: Keep Your Code Running with the Reliability Features of the .NET Framework


Reliability requires the capacity to execute a sequence of operations in a deterministic way, even under exceptional conditions. This allows you to ensure that resources are not leaked and that you can maintain state consistency without relying on application domain unloading (or worse, process restarts) to fix any corrupted state. Unfortunately, in the.NET Framework, not all exceptions are deterministic and synchronous, which makes it difficult to write code that is always deterministic in its ability to execute a predetermined sequence of operations. In this article Stephen Toub will show you why, and explore features of the .NET Framework 2.0 that help you to mitigate these situations and write more reliable code.

Stephen Toub

MSDN Magazine October 2005

SQL Server 2005 - Installation in Cluster Environment - "Service SQL Server Browser (SQLBrowser) fai

Windows Server 2003 R2 Enterprise Edition SP2 32-bit All updates and patches installed Two-node cluster environment configured. Cluster functions perfectly. Trying to install SQL Server 2005 Enterprise Edition (32-bit) as a clustered application. Installation proceeds fine until it begins to install the SQL Server Database Services product. Status window says "Starting Services. Service: SQL Server Browser". Error message is presented: "Service 'SQL Server Browser' (SQLBrowser) failed to start. Verify that you have sufficient privileges to start system services". SQL Server Service Account is in the Administrators group on both local systems, and is a domain account. Account used for the installation is in the Administrators group on both local systems, and is a domain account. Have cancelled out of the install and rebooted both servers numerous times. Exact same error continues to occur. Cluster is active and all resources are owned by node #1 - the node where SQL Server Installation is being run. In the event viewer, I see: Event ID: 10 Description: The SQLBrowser service was unable to establish SQL instance and connectivity discovery". I am also seeing errors like: Source: SQLBrowser. The SQL configuration for SQL is inaccessible or invalid. Source: SQLBrowser. The Configuration of the AdminConnection\TCP protocol in the SQL instance xxxxxx i

Network failure in Cluster Environment,will it stop the SQL Server Instances?

Hi All I have problem with cluster Environment that is there is network failure in Cluster Environment,will it stop SQL Server Instances Event viewrer Information 6:22:09 The MSSQLSERVER service entered the stopped state. 6:22:05 The MSSQLSERVER service was successfully sent a stop control. 6:22:04 The SQLSERVERAGENT service entered the stopped state. 6:21:43 The node lost communication with cluster node 'MSSQL0002' on network 'LAN'. 6:22:02 The SQLSERVERAGENT service was successfully sent a stop control. 6:22:02 Cluster resource 'SQL IP Address1(SQLCLUSTER)' in Resource Group 'SQL' failed. 6:22:02 The TCP/IP interface for Cluster IP Address 'Cluster IP Address' has failed. 6:22:02 The TCP/IP interface for Cluster IP Address 'Cluster IP Address' has failed. 6:22:02 The TCP/IP interface for Cluster IP Address 'MSDTC IP address' has failed. 6:22:02 The TCP/IP interface for Cluster IP Address 'MSDTC IP address' has failed. 6:22:02 The TCP/IP interface for Cluster IP Address 'SQL IP Address1(SQLCLUSTER)' has failed. 6:21:58 The TCP/IP interface for Cluster IP Address 'SQL IP Address1(SQLCLUSTER)' has failed. 6:21:58 Cluster network 'LAN' is operational (up). All available server cluster nodes attached to the network can communicate using it. 6:21:58 The interface for cluster node 'MSSQL0002' on network 'LAN' failed. If the condition persist

SQL 2008 64 Bit 2 node cluster

I wanted to get some feedback on how much memory and CPUs I needed for each node in a 2 Node Cluster. The operating System will be Windows 2008 R2 64 bit Enterprise and the SQL version will be SQL 2008 R2 64 Bit. It will have a total of 8 instances of SQL Server. 2 of those instances will be SQL Server Reporting Services ( 1 SQL Native and the other SharePoint enabled). This cluster will have both nodes in a virtual enviroment. Any feedback would be greatly appreciated.

Adding second node to SQL Server cluster fails.

I have successfully created a MSCS cluster and I have the SQL cluster up and running on the first node.   When I put the install cd in the second node and go through the install it fails with the error SQL Server Setup has encountered the following error:  Object reference not set to an instance of an object. I'm assuming I get this error because on the cluster node configuration screen the "cluster network name" box is blank.   It won't let me select a name or type a name. The SQL cluster is up and running on the first node.   I can ping the sql cluster network name from both boxes.   The setup shows the installed node and the node I"m trying to add, so I don't see any problem. I have Windows 2008 R2 on both nodes and I'm installing SQL Server 2008 R2.   Any idea how I can get past this error?    Thank you.

Which high availability strategy for multi tenant architecture

Hi, I run a multi-tenant application : each client runs the same code but each one hase its own database for personnal parameters and data, that makes about 40-50 databases and growing (i hope so) I want to set up a high availability strategy, for that, I have two VMs each one running SQL Server 2008 Entreprise. I had started to configure log shipping, but I was told that it was not a good idea when there are many dbs involved. How can I choose between HA strategies ? What tests do I have to do to see if i'm still elligible so log shipping (number of transactions per hour ?) If log shipping is not an option, what should I choose ? Failover cluster ? Thanks ! Don't hesitate to ask if you need more information.

in-place upgrade in cluster environment(2000 to 2008),How to do compatabilty Test ???

Hi all As of I know there are three tools are involved in Upgrade process. 1.UpgradeAdvisor 2.RML utility 3.SQL Server Upgrade Assistant. I want do compabilty test...which tools in Best for that...  Thanks in advanceSNIVAS

Adding thrid Node to SS2K8 cluster fails on adding fileshare

I am attempting to add a third node to a two node cluster. I am using the SS2K8 SP1 slipstream I used to successfully install the other two nodes. The new node of the cluster is:  Hardware: Dell R610 blade  OS: Windows Server 2008 R2 Enterprise 64 bit SQL: SQL Server 2008 SP1 64 bit Enterprise. The two node system has been running for months without problems. How do I fix a failure to add a file share for filestream because the value of group was null (Bold Italics below)? I have not found anything on the web that addresses this problem. Brad The bootstrap summary file woes content(s): Detailed results:   Feature:                       Database Engine Services   Status:                        Failed: see logs for details   MSI status:                    Passed   Configuration status:          Failed: see details below   Configuration error code:      0x11428655   Configuration error description: Value cannot be null.  Parameter name: group &nbs

SQL node remove failed on Cluster :The state of your SQL Server installation was not changed after t

HI guys , Have you faced this .I am not able to remove the SQL Server 2008 ENT node .The error is : Overall summary:   Final result:                  Failed: see details below   Exit code (Decimal):           -2068643838   Exit facility code:            1203   Exit error code:               2   Exit message:                  The state of your SQL Server installation was not changed after the setup execution. Please review the summary.txt logs for further details.   Start time:                    2010-09-12 09:40:40   End time:                      2010-09-12 09:48:26   Requested action:              RemoveNode The details.txt shows the same error in a different way : Running Action: CloseUI 2010-09-12 09:44:44 Slp: Stop action skipped in UI

network configuration and best practice for 2 node cluster

For our 2 node cluster i have set the following: Node 1 nic 1 :,, Node 1 nic 2:, node 2 nic 1:,, node 2 nic 2:, Is this a configuration of good practice?? how does the cluster know which subnet to use as the private and which one to use as the public? The cluster IP should belong to the public subnet, for ex: But what about the MSDTC?? should it have an Ip of the public subnet or private subnet? Thanks.

SQL Management Studio not on Passive Node SQL 2008 Cluster

Hello, I would like to be able to use SQL Management Studio on the passive node in our cluster if there's a failover and Passive Node becomes Active Node.  It seems like the Add Node install didn't load SQL Management Studio on our passive node. Any help is appreciated. Thanks,  Morrisos

Error Adding 2nd Cluster Node - Account Name Box Greyed Out


I am in the process of setting up a 2008 R2 SQL cluster using SQL 2008 R2.  I have the first node up and running, and all is well.  Going through the process to the add the 2nd node, I am stuck.

At the part where you configure the SQL service accounts, the SQL server agent account is greyed out.  Hovering your mouse over it shows that field as "Read-Only"  The box for the password you can type in, so I went ahead and put in the password I used for the account on the first node.  (it's a domain account with the corect privledges).  On the database engine section it lets me select my domain account and then I put in the password.  Clicking next gives me an error stating that the "Credentials you provided for the SQL Server Agent service are invalid."  

I know the password is correct, but without the accout name box being greyed out, I am not sure it is pointing to the right account.  Other people too are having this issue:  http://social.msdn.microsoft.com/Forums/en-US/sqlsetupandupgrade/thread/aae80395-7d29-4a07-a1b7-29959e78d382   What is marked as a solution on that thread really isn't.  Follow up posts show other people at the same spot I am.

SharePoint High Availability Topology



We are planning high availabliglty for sharepoint and hae come up with following topology;

1) 4 FE Web Servers (Web + Application Services) - NLB for load balancing.

2) 2 Backend Servers to store SharePoint DB and Report Server D.

3) Mirrirong between the 2 Backend server and witness on another server.

4) I am not sure whether to put SSRS ReportServer Instance on the Web FE boxes or 2 BE Server. If i put SSRS Report Server on 2 BE servers then i have to configure NLB on the BE boxes. Will there be any issues if mirroring and NLB is configured on the same machine.

Your input will be really helpful.



SQL 2005 SP3 w/KB970892 installed on Cluster can't add new node.


We have SQL 2005 Enterprise Cluster, the cluster is running SP3 with the KB970892 hotfix installed. Using Cluster administrator "add new" node, then via Add\Remove programms add SQL node, all is good to this point. However the new node has the RTM EXE and if I try to run the SP3 from the active node w/KB I can't select the MSSQLServer option because "The product instance MSSQLServer been patched with more recent updates", if I try to fail the server over to the new node and install the SP3 I get "failed, no passive nodes updated" and the RTM node also does not update.



ASP.NetWindows Application  .NET Framework  C#  VB.Net  ADO.Net  
Sql Server  SharePoint  Silverlight  Others  All   

Hall of Fame    Twitter   Terms of Service    Privacy Policy    Contact Us    Archives   Tell A Friend