How to vertically scale

This guide explains how to adjust the CPU and memory resources of your PostgreSQL instance on Hikube, either via a predefined preset or with explicit values.

Prerequisites

• kubectl configured with your Hikube kubeconfig
• A PostgreSQL instance deployed on Hikube

Available presets

Hikube offers predefined resource presets to simplify sizing:

Preset	CPU	Memory
nano	250m	128Mi
micro	500m	256Mi
small	1	512Mi
medium	1	1Gi
large	2	2Gi
xlarge	4	4Gi
2xlarge	8	8Gi

warning

If the resources field (explicit CPU/memory) is defined, the resourcesPreset value is entirely ignored. Make sure to clear the resources field if you want to use a preset.

Steps

1. Check current resources

Review the current configuration of your instance:

kubectl get postgres my-database -o yaml | grep -A 5 -E "resources:|resourcesPreset"

Example output with a preset:

  resourcesPreset: micro
  resources: {}

Example output with explicit resources:

  resourcesPreset: micro
  resources:
    cpu: 2000m
    memory: 2Gi

2. Option A: Change the resource preset

To switch from one preset to another (for example from micro to large), apply a patch:

kubectl patch postgres my-database --type='merge' -p='
spec:
  resourcesPreset: large
  resources: {}
'

note

It is important to reset resources: {} when switching to a preset, so that the preset is properly applied. If resources contains explicit values, the preset is ignored.

You can also modify the full YAML manifest:

postgresql-scaled.yaml

apiVersion: apps.cozystack.io/v1alpha1
kind: Postgres
metadata:
  name: my-database
spec:
  replicas: 3
  resourcesPreset: large
  size: 20Gi

  users:
    admin:
      password: SecureAdminPassword

  databases:
    myapp:
      roles:
        admin:
          - admin

Then apply:

kubectl apply -f postgresql-scaled.yaml

3. Option B: Define explicit resources

For fine-grained control, define CPU and memory values directly:

kubectl patch postgres my-database --type='merge' -p='
spec:
  resources:
    cpu: 4000m
    memory: 4Gi
'

Or via the full manifest:

postgresql-custom-resources.yaml

apiVersion: apps.cozystack.io/v1alpha1
kind: Postgres
metadata:
  name: my-database
spec:
  replicas: 3
  resources:
    cpu: 4000m
    memory: 4Gi
  size: 20Gi

  users:
    admin:
      password: SecureAdminPassword

  databases:
    myapp:
      roles:
        admin:
          - admin

kubectl apply -f postgresql-custom-resources.yaml

tip

For PostgreSQL sizing, a good starting rule is to allocate shared_buffers at about 25% of total memory. Adjust PostgreSQL parameters accordingly via the postgresql.parameters section.

4. Verify the rolling update

After changing resources, the operator performs a rolling update of the PostgreSQL pods. Monitor the progress:

kubectl get po -w | grep postgres-my-database

Expected output (during the rolling update):

postgres-my-database-2   1/1     Terminating   0   45m
postgres-my-database-2   0/1     Pending       0   0s
postgres-my-database-2   1/1     Running       0   30s

Wait for all pods to be in Running state:

kubectl get po | grep postgres-my-database

postgres-my-database-1   1/1     Running   0   2m
postgres-my-database-2   1/1     Running   0   4m
postgres-my-database-3   1/1     Running   0   6m

Verification

Confirm that the new resources are applied:

kubectl get postgres my-database -o yaml | grep -A 5 -E "resources:|resourcesPreset"

Verify that the instance is functional:

kubectl get postgres my-database

Expected output:

NAME          READY   AGE   VERSION
my-database   True    1h    0.18.0

Going further

• PostgreSQL API Reference: complete documentation of resources, resourcesPreset parameters and preset table