September 6, 2013 | Written by: Diego Rosa Castilho
Share this post:
I have seen several environments running with the VMWare Virtual Center (VC) without backup or some solution created to avoid downtime. Thus, I will talk about two ways to keep your VC running even if you get a crash involving the system operational or application. Also, we used to get a lot of tools like Tivoli Asset Discovery for Distributed (TAD4D), IBM Tivoli Monitoring for Virtual Servers ITM6 and Server Resource Management SRM) integrated with Virtual Center. These applications will use VC to collect data.
According to VMWare, fault tolerance (FT) provides continuous availability for virtual machines by creating and maintaining a secondary virtual machine (VM) that is identical to, and continuously available to replace, the primary VM in the event of a failover situation. When a secondary VM is called upon to replace its primary VM counterpart, the secondary VM immediately takes over the primary VM’s role with the entire state of the virtual machine preserved. Applications are already running, and data stored in memory does not need to be reentered or reloaded. Therefore, you would need to be aware about the requirements.
The following vSphere features are not supported for fault tolerant virtual machines:
- Storage vMotion
- Linked clones
- Virtual Machine Backups – Snapshots based
Besides, the other option would be clustering your VC using Microsoft Cluster service (MCSC), once Microsoft servers provide three technologies to support clustering:
- Network Load Balancing (NLB)
- Component Load Balancing (CLB)
- Microsoft Cluster Service (MSCS)
Additionally, setup for failover clustering and Microsoft Cluster Service describes the types of clusters you can implement using virtual machines with Microsoft Cluster Service for Windows Server 2003 and Failover Clustering for Windows Server 2008.
First, make certain that you have the correct structure to implement MCSC. Check: vSphere MSCS Setup Checklist
Regarding VMWare structure you can create Cluster Virtual Machines on One Physical Host,Cluster Virtual Machines Across Physical Hosts, and Cluster Physical and Virtual Machines. I would suggest to always use clusters across physical hosts.
Whereas FT has particularities I would suggest using FT for small environment where you have a Virtual Machine with VC and VC’s database installed. Also, FT has a limit of one vCPU (Virtual Processor), it means, your VM can have only one vCPU attached. “http://www.vmware.com/pdf/vsphere5/r51/vsphere-51-configuration-maximums.pdf “
There is no comparison between FT and MCSC, both can be used to avoid downtime and keep your VC running in case of a crash. I would not be able to say which of them would be the best because there will be cases where FT can be a good idea and vice versa.
Still, it is important to know that you cannot depend only on FT or MCSC as your safe harbor. It would be better if you could combine them with a proper backup solution.
Furthermore, you can also get results by using tools like VMware vCenter Server Heartbeat and Fault Domain Manager (FDM).
Lastly, the arguments\comments above are related to vSphere 5.0,5.1.