`helmfile` installation fails

[khteh]

$ helmfile sync
Adding repo cloudnative-pg https://cloudnative-pg.github.io/charts
"cloudnative-pg" has been added to your repositories

Adding repo cluster https://cloudnative-pg.github.io/charts
"cluster" has been added to your repositories

Upgrading release=cloudnative-pg, chart=cloudnative-pg/cloudnative-pg, namespace=postgresql-cluster
Release "cloudnative-pg" has been upgraded. Happy Helming!
NAME: cloudnative-pg
LAST DEPLOYED: Sat Aug 22 21:36:12 2026
NAMESPACE: postgresql-cluster
STATUS: deployed
REVISION: 3
TEST SUITE: None
NOTES:
CloudNativePG operator should be installed in namespace "postgresql-cluster".
You can now create a PostgreSQL cluster with 3 nodes as follows:

cat <<EOF | kubectl apply -f -
# Example of PostgreSQL cluster
apiVersion: postgresql.cnpg.io/v1
kind: Cluster
metadata:
  name: cluster-example
  
spec:
  instances: 3
  storage:
    size: 1Gi
EOF

kubectl get -A cluster

Listing releases matching ^cloudnative-pg$
cloudnative-pg	postgresql-cluster	3       	2026-08-22 21:36:12.222716868 +0800 +08	deployed	cloudnative-pg-0.29.0	1.30.0     

Decrypting secret /usr/src/kubernetes/PostgreSQL/PostgreSQL_Cluster/Secrets/postgresql_secret.yml.enc
Decrypting secret /usr/src/kubernetes/PostgreSQL/PostgreSQL_Cluster/Secrets/postgresql_superuser.yml.enc
Upgrading release=cluster, chart=cloudnative-pg/cluster, namespace=postgresql-cluster
Release "cluster" does not exist. Installing it now.


UPDATED RELEASES:
NAME             NAMESPACE            CHART                           VERSION   DURATION
cloudnative-pg   postgresql-cluster   cloudnative-pg/cloudnative-pg   0.29.0         22s


FAILED RELEASES:
NAME      NAMESPACE            CHART                    VERSION   DURATION
cluster   postgresql-cluster   cloudnative-pg/cluster                   1s

in ./helmfile.yaml: failed processing release cluster: command "/usr/sbin/helm" exited with non-zero status:

PATH:
  /usr/sbin/helm

ARGS:
  0: helm (4 bytes)
  1: upgrade (7 bytes)
  2: --install (9 bytes)
  3: cluster (7 bytes)
  4: cloudnative-pg/cluster (22 bytes)
  5: --version (9 bytes)
  6: 0.8.1 (5 bytes)
  7: --wait (6 bytes)
  8: --timeout (9 bytes)
  9: 300s (4 bytes)
  10: --create-namespace (18 bytes)
  11: --namespace (11 bytes)
  12: postgresql-cluster (18 bytes)
  13: --values (8 bytes)
  14: /tmp/helmfile3641506353/postgresql-cluster-cluster-values-b6c9d9c96 (67 bytes)
  15: --values (8 bytes)
  16: /tmp/helmfile4199551830/postgresql-cluster-cluster-values-84fd4899d7 (68 bytes)
  17: --values (8 bytes)
  18: /tmp/helmfile1095355560/postgresql-cluster-cluster-values-666ccbc9f9 (68 bytes)
  19: --values (8 bytes)
  20: /tmp/helmfile3041987700/postgresql-cluster-cluster-values-555c69478 (67 bytes)
  21: --values (8 bytes)
  22: /tmp/helmfile2994956756/postgresql-cluster-cluster-values-5d6c89cb4d (68 bytes)
  23: --values (8 bytes)
  24: /tmp/helmfile3999138337/postgresql-cluster-cluster-values-59865cddb9 (68 bytes)
  25: --values (8 bytes)
  26: /tmp/helmfile3387206898/postgresql-cluster-cluster-values-7794f98c74 (68 bytes)
  27: --reset-values (14 bytes)
  28: --history-max (13 bytes)
  29: 10 (2 bytes)

ERROR:
  exit status 1

EXIT STATUS
  1

STDERR:
  Error: execution error at (cluster/templates/cluster.yaml:14:6): Invalid cluster type!

COMBINED OUTPUT:
  Release "cluster" does not exist. Installing it now.
  Error: execution error at (cluster/templates/cluster.yaml:14:6): Invalid cluster type!

[dosubot]

@khteh The error Invalid cluster type! comes from the cluster.imageName template helper in the chart [1]. It's invoked at line 14 of cluster.yaml via cluster.image, which explains the error location you're seeing.

The chart accepts three valid values for the top-level type field [2]:

• postgresql (default)
• postgis
• timescaledb

If type is empty, misspelled, or set to anything else, the template fails with Invalid cluster type!. This check is skipped only when cluster.imageName is explicitly provided [3].

Since you're passing 7 values files via helmfile, the most likely cause is that one of them is overriding type with an invalid or empty value. To debug, you can render the merged values with a dry run:

helm upgrade --install cluster cloudnative-pg/cluster --version 0.8.1 \
  --values file1.yaml --values file2.yaml ... \
  --dry-run --debug

Look at the computed type value in the output. To fix it, either ensure type: postgresql is set in your values (or whichever type you need), or set cluster.imageName directly to bypass the type check entirely.

To reply, just mention @dosu.

---

Docs are dead. Just use Dosu.

[khteh]

I don't have type: defined in any of my manifests.

[dosubot]

@khteh If type is not defined in any of your values files, the chart's default of type: postgresql should be used [1]. However, if one of your values files explicitly sets type to an empty string ("") or null, that will shadow the chart default. In the template, the cluster.imageName helper only matches "postgresql" or "postgis" via eq checks and falls through to fail "Invalid cluster type!" for anything else — including nil or "" [2].

Since you mentioned using multiple --values files via helmfile, it's likely one of them is inadvertently overriding type. You can verify what your merged values resolve to:

helmfile --selector name=cluster template 2>&1 | grep "type:"

Adding an explicit type: postgresql to your primary cluster values file should resolve the issue.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.

[khteh]

To no avail. Same error.