feat(cluster): expose replicationSlots on the CNPG Cluster (#877)

Surfaces CloudNativePG's `spec.replicationSlots` on the cluster chart so
chart users can enable synchronization of user-created logical
replication slots between the primary and standbys. With PostgreSQL 17+
failover slots, this lets CDC consumers (e.g. Debezium) survive a CNPG
failover without losing the slot.

To make logical decoding slots survive failover, users must:
- enable `cluster.replicationSlots.highAvailability.synchronizeLogicalDecoding`
- set `cluster.postgresql.parameters.hot_standby_feedback: "on"`
- set `cluster.postgresql.parameters.sync_replication_slots: "on"`
- create the CDC client's logical slot with `failover = true`

Requires CloudNativePG 1.27+ and PostgreSQL 17+ for native failover
slots. The block is omitted entirely when the value is empty, so
existing deployments are unaffected.

The non-default-configuration chainsaw test is extended with a
`replicationSlots` block and the matching PostgreSQL parameters, and
asserts they appear verbatim on the rendered Cluster CR.

Signed-off-by: Philippe Noël <philippemnoel@gmail.com>

Author: philippemnoel

charts/cluster/README.md

@@ -195,6 +195,7 @@ Kubernetes: `>=1.29.0-0`
 | cluster.primaryUpdateMethod | string | `"switchover"` | Method to follow to upgrade the primary server during a rolling update procedure, after all replicas have been successfully updated. It can be switchover (default) or restart. |
 | cluster.primaryUpdateStrategy | string | `"unsupervised"` | Strategy to follow to upgrade the primary server during a rolling update procedure, after all replicas have been successfully updated: it can be automated (unsupervised - default) or manual (supervised) |
 | cluster.priorityClassName | string | `""` |  |
+| cluster.replicationSlots | object | `{}` | Replication slot management. To make logical decoding slots survive failover for CDC consumers (e.g. Debezium), enable `highAvailability.synchronizeLogicalDecoding`, set `cluster.postgresql.parameters.hot_standby_feedback: "on"` and `cluster.postgresql.parameters.sync_replication_slots: "on"`, and ensure the CDC client creates its logical slot with `failover = true`. Requires CloudNativePG 1.27+ and PostgreSQL 17+ for native failover slots. See: https://cloudnative-pg.io/documentation/current/replication/#replication-slots |
 | cluster.resources | object | `{}` | Resources requirements of every generated Pod. Please refer to https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/ for more information. We strongly advise you use the same setting for limits and requests so that your cluster pods are given a Guaranteed QoS. See: https://kubernetes.io/docs/concepts/workloads/pods/pod-qos/ |
 | cluster.roles | list | `[]` | This feature enables declarative management of existing roles, as well as the creation of new roles if they are not already present in the database. See: https://cloudnative-pg.io/documentation/current/declarative_role_management/ |
 | cluster.securityContext | object | `{}` | Configure Container Security Context. See: https://cloudnative-pg.io/documentation/preview/security/ |

charts/cluster/templates/cluster.yaml

@@ -101,6 +101,11 @@ spec:
       {{- toYaml . | nindent 6 }}
     {{- end }}
 
+  {{- with .Values.cluster.replicationSlots }}
+  replicationSlots:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+
   {{- if not (and (empty .Values.cluster.roles) (empty .Values.cluster.services)) }}
   managed:
     {{- with .Values.cluster.services }}

charts/cluster/test/postgresql-cluster-configuration/01-non_default_configuration_cluster-assert.yaml

@@ -12,6 +12,12 @@ spec:
   postgresUID: 1001
   postgresGID: 1002
   instances: 2
+  replicationSlots:
+    highAvailability:
+      enabled: true
+      slotPrefix: "_cnpg_"
+      synchronizeLogicalDecoding: true
+    updateInterval: 30
   postgresql:
     ldap:
       server: 'openldap.default.svc.cluster.local'
@@ -23,7 +29,9 @@ spec:
           key: 'data'
         searchAttribute: 'uid'
     parameters:
+      hot_standby_feedback: "on"
       max_connections: "42"
+      sync_replication_slots: "on"
     pg_hba:
       - host all 1.2.3.4/32 trust
     pg_ident:

charts/cluster/test/postgresql-cluster-configuration/01-non_default_configuration_cluster.yaml

@@ -87,6 +87,12 @@ cluster:
        inRoles:
          - pg_monitor
          - pg_signal_backend
+  replicationSlots:
+    highAvailability:
+      enabled: true
+      slotPrefix: "_cnpg_"
+      synchronizeLogicalDecoding: true
+    updateInterval: 30
   postgresql:
     ldap:
       server: 'openldap.default.svc.cluster.local'
@@ -98,7 +104,9 @@ cluster:
           key: 'data'
         searchAttribute: 'uid'
     parameters:
+      hot_standby_feedback: "on"
       max_connections: "42"
+      sync_replication_slots: "on"
     pg_hba:
       - host all 1.2.3.4/32 trust
     pg_ident:

charts/cluster/values.schema.json

@@ -510,6 +510,11 @@
           "default": "",
           "type": "string"
         },
+        "replicationSlots": {
+          "description": "Replication slot management. To make logical decoding slots survive\nfailover for CDC consumers (e.g. Debezium), enable\n`highAvailability.synchronizeLogicalDecoding`, set\n`cluster.postgresql.parameters.hot_standby_feedback: \"on\"` and\n`cluster.postgresql.parameters.sync_replication_slots: \"on\"`, and ensure\nthe CDC client creates its logical slot with `failover = true`.\nRequires CloudNativePG 1.27+ and PostgreSQL 17+ for native failover slots.\nSee: https://cloudnative-pg.io/documentation/current/replication/#replication-slots",
+          "required": [],
+          "type": "object"
+        },
         "resources": {
           "description": "Resources requirements of every generated Pod.\nPlease refer to https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/ for more information.\nWe strongly advise you use the same setting for limits and requests so that your cluster pods are given a Guaranteed QoS.\nSee: https://kubernetes.io/docs/concepts/workloads/pods/pod-qos/",
           "required": [],

charts/cluster/values.yaml

@@ -390,6 +390,20 @@ cluster:
           # key: 'data'
         # searchAttribute: 'uid'
 
+  # -- Replication slot management. To make logical decoding slots survive
+  # failover for CDC consumers (e.g. Debezium), enable
+  # `highAvailability.synchronizeLogicalDecoding`, set
+  # `cluster.postgresql.parameters.hot_standby_feedback: "on"` and
+  # `cluster.postgresql.parameters.sync_replication_slots: "on"`, and ensure
+  # the CDC client creates its logical slot with `failover = true`.
+  # Requires CloudNativePG 1.27+ and PostgreSQL 17+ for native failover slots.
+  # See: https://cloudnative-pg.io/documentation/current/replication/#replication-slots
+  replicationSlots: {}
+    # highAvailability:
+    #   enabled: true
+    #   slotPrefix: "_cnpg_"
+    #   synchronizeLogicalDecoding: true
+    # updateInterval: 30
 
   # -- BootstrapInitDB is the configuration of the bootstrap process when initdb is used.
   # See: https://cloudnative-pg.io/documentation/current/bootstrap/