Veritas Access Appliance 8.3 Solutions Guide for Enterprise Vault
- Introduction
- System Requirements
- Installing and configuring Enterprise Vault with Access Appliance
- Access Appliance features for Enterprise Vault archival storage
- Access Appliance archival policy configuration for Enterprise Vault
- Continuous replication failover and failback
- Troubleshooting
Continuous replication failover and failback
Typically, the source cluster drives a replication session. However, in some situations, it may be useful for the destination cluster to drive the replication session. Access Appliance supports a failover and a failback feature for continuous replication. This feature enables control of replication to be temporarily relocated from the source cluster to the destination (target) cluster.
Continuous replication failover and failback is useful for:
Planned failover
In cases where the source cluster is taken down for routine maintenance or for moving applications to another cluster, a planned failover procedure is available for moving replication from the source cluster to the destination cluster.
Unplanned failover or Disaster recovery
In cases where the source cluster fails unexpectedly, an unplanned failover procedure is availble for moving replication to the destination cluster.
With failover and failback, you can use the replication continuous failover command to move control from the source cluster to the destination cluster. You use the replication continuous failback to restore control to the source cluster.
Continuous replication failover does not automatically move the NFS or the CIFS share information that is associated with file system from the source cluster to the destination cluster. Share information has to be done manually.
CIFS shares should be removed before failover and failback operations. After failover and failback operation are complete, add the CIFS shares again.
Recommendations while configuring Access Appliance for continuous replication:
The number of RVGs configured under continuous replication on Access appliance should be restricted to 4.
The number of file systems under one RVG should be limited to 12.
Veritas recommends that you to add one file system at a time to an RVG. Once the file system is added and the status of continuous replication becomes consistent (up-to-date), add another file system to the RVGs.
Ensure that continuous replication IPs configured on Access Appliance (both primary and secondary sites) are properly resolved by the configured DNS server or by
/etc/hosts
file on both nodes of the primary and secondary sites.Addition of multiple file systems to an existing Replicated Volume Group is supported only through CLISH.
The failover and failback happens at the RVG level. So, it is recommended to keep all partitions (corresponding file systems on Access Appliance) of one Enterprise Vault store in one RVG. This reduces the need of failing over or failing back multiple RVGs.
For example:
If ev_vault1 uses share1 (closed partition) and share2(open partition) on Access Appliance which maps to file systems, fs1 (share1), and fs2 (share2).
If ev_vault2 uses share3 and share4 on Access Appliance which maps to file systems, fs3 (share3), and fs4 (share4).
In this scenario, you can create rvg1 for fs1 and add fs2 to rvg1. Similarly, you can create rvg2 for fs3 and fs4. In case, you have to failover ev_vault1, you can failover rvg1 and Enterprise Vault will be functional during disaster recovery. You are not required to do anything to rvg2 unless you want to failover ev_vault2 also.
If there are more than 4 Enterprise Vault stores in an environment, you are required to put the file systems (partitions) of different Enterprise Vault stores in one RVG as the maximum number of recommended RVGs on Access Appliance is 4.
The bandwidth available for replication between primary and secondary sites should be able to handle the rate at which data is being written on the primary site. With synchronous continuous replication, the writes on file systems on primary site are completed after it is replicated to the secondary site. Hence, Veritas recommends that you have a high bandwidth for replication with synchronous continuous replication.