Persistent template

PostgreSQL

A single PostgreSQL server in your project, deployed from the postgresql-persistent template. Data lives on a persistent volume; you operate the database.

Persistent volume for data
Credentials in a Secret
Single instance
Cluster-internal by default

Deploy PostgreSQL Talk to an engineer

No credit card required · Free during the Alpha phase · View pricing

Overview

A relational database for applications in the same project.

Relational storage

PostgreSQL for transactional workloads — SQL, JSONB, full-text search and the extensions shipped in the image.

Dev, test and small production

One server per template instance. Suited to development, CI, test environments and workloads that tolerate a restart.

You operate it

You own backups, upgrades and tuning. For PostgreSQL operated by the platform, with replication and failover, see Databases.

Deployment

An OpenShift template: one set of parameters, processed into the objects below.

Template

postgresql-persistent plus your parameters

Secret

User, password, database name

PersistentVolumeClaim

Sized by VOLUME_CAPACITY, mounted as the data directory

Server + Service

One pod, Recreate strategy, port 5432

Web console — Add → Database → PostgreSQL. Choose the persistent template, set the volume size and database name; leave user and password empty to have them generated.
CLI — oc new-app --template=postgresql-persistent with -p parameters. List every parameter with oc process --parameters.
Use the persistent template. The postgresql-ephemeral variant keeps data in the pod's filesystem and loses it on every restart.
Single writer — the volume is mounted by one pod at a time, so the server is recreated, not rolled, on change. Expect a short interruption on restart or redeploy.

bash — oc

# see the template's parameters
$ oc process --parameters postgresql-persistent -n openshift

# deploy with a 10 GiB data volume
$ oc new-app --template=postgresql-persistent \
    -p DATABASE_SERVICE_NAME=postgresql \
    -p POSTGRESQL_DATABASE=app \
    -p VOLUME_CAPACITY=10Gi

# open a psql session in the pod
$ oc rsh deployment/postgresql psql

Lifecycle

What the template does not do for you.

Backups — not automated. Schedule pg_dump, run through oc exec or a CronJob, and store the dumps outside the database's own volume.
Upgrades — the server version follows the image tag set by POSTGRESQL_VERSION. A major-version change needs a migration of the data directory: dump and restore, or the image's POSTGRESQL_UPGRADE mode for a step from the previous major version. Take a dump first.
Availability — one instance, no streaming replication or automatic failover. If the workload needs either, use the managed PostgreSQL service.
Capacity — grow the volume by editing the PVC's requested size; set CPU and memory limits with MEMORY_LIMIT or on the workload.

Platform integration

Standard Kubernetes objects in your own project.

Block storage — the PVC is provisioned from the cluster's default storage class.
Connections — applications in the project connect to postgresql:5432 (the DATABASE_SERVICE_NAME). No Route is created, so the database is not reachable from outside the cluster.
Secrets — inject the generated credentials into your application with oc set env --from=secret/postgresql instead of copying them.
Pairs with — Keycloak, which needs an external database, and Node.js applications in the same project.

Deploy PostgreSQL with a persistent volume.

From the console's database catalogue, or with oc new-app --template=postgresql-persistent.

Deploy PostgreSQL Managed databases

Hosted in the Netherlands · EU data residency · 100% renewable energy

Far far away, behind the word moun tains, far from the countries Vokalia and Consonantia, there live the blind texts. Separated they live in Bookmarksgrove right at the coast of

759 Pinewood Avenue
Marquette, Michigan
info@domain.com
Online Support
906-624-2565
Mon-Fri 8am-5pm

Purchase Now