TechOnTip Weblog

Run book for Technocrats

SharePoint 2010 Small Farm Site Resilience Architecture

Posted by Brajesh Panda on June 14, 2010

Please Note:

SharePoint 2010 is just release 30days before. All promise features may not work correctly or if any product bug comes in-between team have to wait till next service pack to overcome the same.

SharePoint Architecture;

  • Backend SQL Infrastructure
    • Use Existing deployed SQL infrastructure (Deployed by
    • In Production Site-A use existing 2 Node Windows 2008 R2 Failover SQL Cluster for HA as 1st Level of HA to address Server Failure
    • In Production Site-B use exiting Mirror SQL Server for 2nd Level of HA for Production Site Failure
    • Every time one SharePoint database gets created in Site-A cluster, mirroring must configured manually to site B. SharePoint do not configure mirroring in application. It needs SQL Administration Skill & helps from DBA team.

       

  • SharePoint Front End
    • Two Front End Servers can be deployed
    • These servers will hold below set roles/features in both server giving redundancy
      • IIS 7 Role
      • Central SharePoint Administration Console (Port 10,000)
      • SharePoint Web Applications
      • SharePoint Virtual Directory
      • IIS Application Pools
    • These servers can hold Managed Services like Search etc. However at a time services can run on one server. If that server fails, SharePoint admin have to start services on other server. All services can run on a single front end server, can be divided among front end servers or dedicated extra application server can be used.
    • No need to start all Managed Services from the day one, it will create un necessary loads on farm. It should be on demand i.e. if YOU need that feature in future & if YOU developing applications using those features then should turned on. Else it will be management overhead

     

  • High Availability
    • Front End Servers should be load balanced using existing F5 Hardware Load Balancer hence seamless front end redundancy can be provided to end user
    • In general situation all Front end server from Site A & B will be doing transactions on Production Clustered SQL Databases
    • If one SQL server fails, databases will be transferred to other available SQL node providing complete redundancy.
    • For failover to Site B database in case of failure Managed Failover Database Partner must be configured for all Content & Service Application databases. This is a new feature in SharePoint 2010.
    • Before this SQL Admin must have configured Database Mirroring & production data must be stored in Site-B In Recovery SQL databases.
    • In case of Production Cluster SQL Database fails; SQL Administrator can initiate failover from Production Clustered SQL Server in Site A to Mirror SQL Server in Site B.
    • As soon as Databases get activated at Site B, Site A databases will move to in-recovery mode. SharePoint Front End & Managed Application Services will be do transactions in Site B database.
  • Backup/Restore
    • Backup Best Practices
      • Regular SQL Database Backup for Database Level Recovery in case logical database corruption
      • Use Disk to Disk to Tape backup solution
    • Web Application/Site Collection/Site level/Farm Level Backup/restore can be possible using Central Administration Tool & Powershell commands.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: