5V0-35.21 Exam Dumps : VMware vRealize Operations Specialist

 A vRealize Operations cluster in a Continuous Availability configuration consists of three node types: primary node, primary replica node, and witness node. These node types are deployed across two fault domains, which are separate physical locations that can tolerate failures of an entire rack or site. The witness node is deployed in a third location to monitor the network connectivity between the two fault domains and prevent split-brain scenarios1.

The primary node is the first node that is deployed in the cluster and acts as the master node. It is responsible for managing the cluster configuration, user interface, alerting, and reporting. The primary node is deployed in fault domain 11.

The primary replica node is the second node that is deployed in the cluster and acts as the backup node for the primary node. It is responsible for replicating and synchronizing the data and configuration from the primary node. The primary replica node can take over the functions of the primary node in case of a failure. The primary replica node is deployed in fault domain 21.

The witness node is the third node that is deployed in the cluster and acts as the arbitrator node. It is responsible for monitoring the network connectivity and availability of the primary node and the primary replica node. The witness node can detect a split-brain situation, which occurs when the network connection between the two fault domains is lost and both nodes assume the role of the master node. The witness node can resolve the split-brain situation by making one of the nodes offline to avoid data inconsistency. The witness node is deployed in a third location that is independent of the two fault domains1.

The other options are not node types that are configured in a Continuous Availability configuration. A secondary node is a node that is added to the cluster to increase the capacity and performance of the cluster. A secondary node can be deployed in either fault domain, but it is not required for Continuous Availability1. A remote collector node is a node that is deployed outside the cluster to collect data from remote or isolated data sources. A remote collector node can be deployed in any location, but it is not part of the Continuous Availability configuration2. A secondary replica node is not a valid node type in vRealize Operations.

References: 1: About vRealize Operations Continuous Availability 2: Remote Collector Nodes

According to the VMware vRealize Operations Reference Materials, the High Availability feature in vRealize Operations cluster must be used for this requirement. The High Availability feature enables the cluster to tolerate one node failure without losing any data or functionality. The High Availability feature works by creating a replica node for each data node in the cluster, and synchronizing the data between them. The replica node acts as a standby node that can take over the role of the data node in case of a failure. The replica node can also serve as a load balancer to distribute the workload among the data nodes. The other options are incorrect because they do not meet the requirement or they are not supported by vRealize Operations. The Remote Collector option is not a cluster feature, but a type of node that can be used to collect data from remote sources and forward it to the cluster. The Remote Collector node does not store any data or perform any analysis, and it cannot replace a data node in case of a failure. The Continuous Availability option is a cluster feature that enables the cluster to tolerate multiple node failures without losing any data or functionality. However, the Continuous Availability feature requires at least two physical locations, and it is not supported for a single location. The Stretched Cluster option is not a cluster feature, but a configuration that allows a cluster to span across two physical locations. The Stretched Cluster option can provide high availability and disaster recovery, but it also requires additional network and storage configuration, and it is not supported for a single location. References: vRealize Operations Cluster and Node Maintenance; Adding High Availability to vRealize Operations; Adding Continuous Availability; Stretched Cluster

Making sure the NFS server allows read-write access to the shared location by the root system account on each cell is the correct way to enable the collection of log bundles from all cells at once. This is because VMware Cloud Director uses the vmware-vcd-support script to collect host log information as well as Cloud Director logs, and this script requires the NFS server to have read-write access to the shared location by the root system account on each cell1. Installing vRealize Log Insight agent on each cell to redirect logs to a centralized location is not a valid option, as vRealize Log Insight is a separate product that does not integrate with VMware Cloud Director. Configuring a syslog server on each of the cells, and requesting the logs from the syslog server is not a feasible option, as it would require manual intervention and coordination to collect the logs from multiple sources. Mapping the log location on the primary cell to a shared folder on each of the other cells in the server group, ensuring they all have read/write rights on the primary is not asufficient option, as it does not address the NFS server configuration that is required for the vmware-vcd-support script to work properly. References: Collecting diagnostic information for VMware Cloud Director (1026312). Preparing the Transfer Server Storage for VMware Cloud Director on Linux.