Important Update: Cohesity Products Documentation
All Cohesity product documentation are now managed via the Cohesity Docs Portal: https://docs.cohesity.com/HomePage/Content/home.htm. Some documentation available here may not reflect the latest information or may no longer be accessible.
InfoScale™ 9.1 Cluster Server 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 the EMC SRDF agent's online function
If the state of all local devices in an RDF1 type device group is read-write enabled (RW) and the replication link is in the Consistent or Synchronized state, the agent creates a lock file on the local host. The lock file indicates that the resource is online.
If all the local devices are in the write-disabled (WD) state, the agent runs the symrdf command to enable read-write access to the devices.
Depending on SRDF/S and SRDF/A, the states can be different as follows:
For R2 devices in the SYNCHRONIZED or CONSISTENT state, the agent runs the symrdf failover command to make the devices writable.
For R1 devices in the FAILED OVER or R1 UPDATED state, the agent runs the symrdf failback command to make the devices writable.
For all devices in the PARTITIONED state, the agent runs the symrdf command to make the devices writable.
The agent runs the command only if the AutoTakeover attribute is set to 1 and if there are no dirty tracks on the local device. Dirty tracks indicate that an out-of-order synchronization was in progress when the devices became partitioned, rendering them inconsistent and unusable. If dirty tracks exist, the online entry point faults on timeout.
For R1 devices in the UPDINPROG state, the agent runs the symrdf command only after the devices transition to the R1 UPDATED state.
For R2 devices in the SYNCINPROG state, the agent runs the symrdf command only after the devices transition to the SYNCHRONIZED or CONSISTENT state.
The agent does not run any command if there is not enough time remaining for the entry point to complete the command.
In case of configurations with composite groups, if the LinkMonitor attribute is set to 1, the online function determines the RDF pair state at the composite group level based on the RDF pair states of the comprised device groups:
If all the device groups are in the Synchronized state, the composite group level state is considered to be Synchronized.
If all the device groups are in the Consistent state, the composite group level state is considered to be Consistent.
If some of the device groups are either in the Synchronized or the Consistent state, while some are in the Suspended state, the composite group level state is considered to be Suspended.
If some of the device groups are either in the Synchronized or the Consistent state, while some are in the Split state, the composite group level state is considered to be Split.
If some of the device groups are in the Split state, while some are in the Suspended state, the composite group level state is considered to be Suspended.
The agent brings the SRDF resource online only if the RDF pair state at the composite group level is either Synchronized or Consistent.
Configurations where some of the device groups are in the Synchronized state, while some are in the Consistent state, are not supported.
More Information