Monitoring the PostgreSQL Cluster

eG Enterprise offers a special-purpose monitoring model for the PostgreSQL Cluster with the help of which administrators can keep a constant check on the availability and overall performance of the cluster.

Figure 1 depicts the layer model of a Mongo DB Cluster.

Figure 1 : Layer model of the PostgreSQL Cluster

Each layer of Figure 1 above is mapped to tests that provide deep insights into how the PostgreSQL Cluster manages connections to it, cluster availability, number of nodes, replication status, connections and in the process provides effective pointers to how the server can be fine-tuned to perform better.

The metrics reported by these tests help answer the following performance questions:

• What is the current status of the PostgreSQL cluster and its nodes?

• Are any cluster nodes currently unavailable or in an unhealthy state?

• Has any node in the cluster been restarted recently?

• Is the replication lag time or size too high for any replica node?

• Are any replica nodes lagging significantly behind the primary?

• Has the master node changed unexpectedly?

• Are the read/write and read-only roles correctly configured across nodes?

• Are too many active or waiting connections being observed on the database?

• Is the database responding to connection and query requests within acceptable time limits?

• Are any nodes currently under maintenance?

• Has any node experienced frequent or recent role switches?

• Is the database serving too many queries or fetching large volumes of records?

• Are all nodes running continuously, or is uptime fluctuating across the cluster?

• Are the replicas operating in hot standby mode as expected?

• Is replication enabled and functioning correctly across all nodes?

This section will deal with the first layer of the layer model only, as the Network Layer have been discussed elaborately in the Monitoring Unix and Windows Servers document.