Cluster Server 7.4 Agent for EMC SRDF Configuration Guide - Windows
- Introducing the agent for EMC SRDF
- Configuring the agent for EMC SRDF
- Testing VCS disaster recovery support with EMC SRDF
- How VCS recovers from various disasters in an HA/DR setup with EMC SRDF
- Testing the global service group migration
- Testing disaster recovery after host failure
- Testing disaster recovery after site failure
- Performing failback after a node failure or an application failure
- Performing failback after a site failure
- Setting up fire drill
About preventing split-brain
Split-brain occurs when all heartbeat links between the primary and secondary hosts are cut. In this situation, each side mistakenly assumes that the other side is down. You can minimize the effects of split-brain by ensuring that the cluster heartbeat links pass through a similar physical infrastructure as the replication links. When you ensure that both pass through the same infrastructure, if one breaks, so does the other.
Sometimes you cannot place the heartbeats alongside the replication links. In this situation, a possibility exists that the cluster heartbeats are disabled, but the replication link is not. A failover transitions the original R1 to R2 and R2 to R1. In this case, the application faults because its underlying volumes become write-disabled, causing the service group to fault. VCS tries to fail it over to another host, causing the same consequence in the reverse direction. This phenomenon continues until the group comes online on the final node. You can avoid this situation by setting up your infrastructure such that loss of heartbeat links also means the loss of replication links.
To minimize the chances of split-brain, use the steward process.