You can change the parameters that make up the FEP. PostgreSQL provides two types of parameters: those that take effect immediately, and those that take effect after restarting FEP server process.
postgresql.conf
pg_hba.conf
pgaudit.conf
Note
For parameters that take effect immediately, operator will apply the change to all FEP Pods and reload the FEP server process automatically. There is no outage on the cluster.
For parameters that take effect after restarting FEP server process, operator will update the configuration files on all FEP Pods. However, users have to initiate a manual restart of FEP process on all the FEP Pods using the FEPAction CR. There is a momentary outage on the cluster and the users should perform this action at a time that has least disruption to the service.
You can change the amount of CPU and memory resources allocated to FEP server containers, FEP backup containers, or FEP pgpool2 containers by changing the FEPCluster CR. The operator will apply the change to the Statefulset. However, the users have to perform a restart of all the Pods for the new resource allocation to take effect.
Note
Changing resource allocation will not take effect immediately. The users have to restart all the Pods for new resource allocation to take effect. There is a momentary outage on the cluster and the users should perform this action at a time that has least disruption to the service.
You can extend the disk size mounted on the FEP server container by modifying the FEPCluster custom resource.
Disk expansion can also work with Prometheus to automatically expand based on disk usage.
Disk extensions must take advantage of disks that support the Kubernetes PVC extension.
