Search This Blog

Tuesday, October 26, 2010

Failover Clustering Server 2008 R2

Today we are going to build windows Failover cluster using Windows Server 2008 R2 Ent Edition. Failover Clustering is now available in Windows 2008 Ent. and Window 2008 Data Center edition.
Before Beginning the section is divided into following.

1.General Intro
2. Requirements (Hardware and Software)
3. Step by Step deployment
4. Special Consideration* [These can be included anywhere between step 1 to 3]

General Intro
A failover cluster is a group of independent nodes that provide high availability for application and services. The cluster servers generally called Nodes are connected by physical cables and by software. In an event of server failure other node takes charge of services, this process is called failover. During this process user distruption is minimum.

A failover cluster can be used to provide high availability for many application and services like file server, print server, database, mail server, dhcp etc.

Usually in a failover cluster env a shared storage is used between the nodes and it's connected via high speed gigabit ethernet, Fibre Channel, HBA with Multi path I/O* recommmeded, iSCSI

Regardless of type of service that we will be hosting on failover cluster we need following hardware, software, service account and network infrastructure.

Servers: It's recommended to have identical server hardware which is mentioned on HCL
Network adapters and cables: In the network infrastructure that connects your cluster nodes, avoid having single points of failure. There are multiple ways of accomplishing this. You can connect your cluster nodes by multiple, distinct networks. Alternatively, you can connect your cluster nodes with one network that is constructed with teamed network adapters, redundant switches, redundant routers, or similar hardware that removes single points of failure.

For SAS and Fibre Channel: If you are using Serial Attached SCSI or Fibre Channel, in all clustered servers, all components of the storage stack should be identical. It is required that the multipath I/O (MPIO) software and Device Specific Module (DSM) software components be identical. It is recommended that the mass-storage device controllers—that is, the host bus adapter (HBA), HBA drivers, and HBA firmware—that are attached to cluster storage be identical.

Storage:  A compatiable storage with Windows Server 2008

 At least 2 physical disk or volumes at hardware level.
Partitions formatted with NTFS
If you are planning to use SAN then makes sure LUNs are isolated using Zoning or Masking.
Software Requirements:
Network Settings(same across cluster like flow control, Duplex mode etc) and IP address
DNS for name resolution
Domain Role: All cluster servers in same domain. The recommened role is member server*.

Account for administrering. The domain admin account can  be used but not necessary. we can achieve the same with domain user account with special consideration The account needs to be a member of local admin group on all cluster servers and Should have permission to create computer objects and read all properties permissions in the domain.

No comments:

Post a Comment