Helm installation failure with opensearch as visibilitystore and postgres as defaultstore

[prashanthzen]

Is it possible to deploy temporal with postgres as primary datastore and opensearch as visibility datastore? I'm trying to deploy helm chart (1.22.4 version) using the values.yaml added in this message, but I keep running into the following error:

Error: INSTALLATION FAILED: execution error at (temporal/charts/temporal/templates/server-deployment.yaml:37:28): Please specify cassandra port for visibility store

values.yaml:

cassandra:
  enabled: false
elasticsearch:
  enabled: false
  external: true
  host: "es.mgmt.xxx.com"
  port: "443"
  version: "v7"
  scheme: "https"
  logLevel: "error"
  username: "temporal_visibility"
  password: "<placeholder>"
  visibilityIndex: temporal-visibility
schema:
  setup:
    enabled: true
  update:
    enabled: false
debug: true
server:
  replicaCount: 3
  podAnnotations:
    linkerd.io/inject: enabled
    config.linkerd.io/proxy-cpu-request: "10m"
    config.linkerd.io/proxy-memory-limit: 1Gi
    config.linkerd.io/proxy-memory-request: 25Mi
  metrics:
    annotations:
      enabled: true
    serviceMonitor:
      enabled: true
  dynamicConfig:
    limit.blobSize.error:
    - value: 268435456
    limit.historySize.error:
    - value: 268435456
    system.transactionSizeLimit:
    - value: 268435456
    frontend.namespaceCount:
    - value: 10000
    frontend.keepAliveMaxConnectionAge:
    - value: 72h
    frontend.keepAliveMaxConnectionAgeGrace:
    - value: 60s
    frontend.keepAliveMaxConnectionIdle:
    - value: 72h
    frontend.shutdownDrainDuration:
    - value: 30s
    matching.shutdownDrainDuration:
    - value: 30s
    history.shutdownDrainDuration:
    - value: 30s
    matching.longPollExpirationInterval:
    - value: 10s
    system.enableReadVisibilityFromES:
    - value: true
      constraints: {}
  config:
    logLevel: "debug,info"
    numHistoryShards: 512
    persistence:
      defaultStore: default
      default:
        driver: sql
        sql:
          driver: postgres12
          host: db-console-pg.xxx.com
          port: 5432
          database: temporal
          user: temporal
          existingSecret: pg-temporal
          maxConns: 20
          maxConnLifetime: "1h"
  frontend:
    metrics:
      annotations:
        enabled: false
      serviceMonitor:
        enabled: true
    service:
      membershipPort: 7933
    podAnnotations:
      linkerd.io/inject: enabled
      config.linkerd.io/proxy-cpu-request: "10m"
      config.linkerd.io/proxy-memory-limit: 1Gi
      config.linkerd.io/proxy-memory-request: 25Mi
      config.linkerd.io/skip-outbound-ports: 7933,7934,7935,7936,7939
      config.linkerd.io/skip-inbound-ports: "7933"
    resources:
      requests:
        cpu: 100m
        memory: 256Mi
      limits:
        memory: 1Gi
  history:
    metrics:
      annotations:
        enabled: false
      serviceMonitor:
        enabled: true
    service:
      membershipPort: 7934
    podAnnotations:
      linkerd.io/inject: enabled
      config.linkerd.io/proxy-cpu-request: "10m"
      config.linkerd.io/proxy-memory-limit: 1Gi
      config.linkerd.io/proxy-memory-request: 25Mi
      config.linkerd.io/skip-outbound-ports: 7933,7934,7935,7936,7939
      config.linkerd.io/skip-inbound-ports: "7934"
    resources:
      requests:
        cpu: 100m
        memory: 256Mi
      limits:
        memory: 1Gi
  matching:
    metrics:
      annotations:
        enabled: false
      serviceMonitor:
        enabled: true
    service:
      membershipPort: 7935
    podAnnotations:
      linkerd.io/inject: enabled
      config.linkerd.io/proxy-cpu-request: "10m"
      config.linkerd.io/proxy-memory-limit: 1Gi
      config.linkerd.io/proxy-memory-request: 25Mi
      config.linkerd.io/skip-outbound-ports: 7933,7934,7935,7936,7939
      config.linkerd.io/skip-inbound-ports: "7935"
    resources:
      requests:
        cpu: 100m
        memory: 256Mi
      limits:
        memory: 1Gi
  worker:
    metrics:
      annotations:
        enabled: false
      serviceMonitor:
        enabled: true
    service:
      membershipPort: 7939
    podAnnotations:
      linkerd.io/inject: enabled
      config.linkerd.io/proxy-cpu-request: "10m"
      config.linkerd.io/proxy-memory-limit: 1Gi
      config.linkerd.io/proxy-memory-request: 25Mi
      config.linkerd.io/skip-outbound-ports: 7933,7934,7935,7936,7939
      config.linkerd.io/skip-inbound-ports: "7939"
    resources:
      requests:
        cpu: 100m
        memory: 256Mi
      limits:
        memory: 1Gi
web:
  metrics:
    annotations:
      enabled: false
    serviceMonitor:
      enabled: true
  replicaCount: 3
  podAnnotations:
    linkerd.io/inject: enabled
    config.linkerd.io/proxy-cpu-request: "10m"
    config.linkerd.io/proxy-memory-limit: 1Gi
    config.linkerd.io/proxy-memory-request: 25Mi
  resources:
    requests:
      cpu: 10m
      memory: 256Mi
    limits:
      memory: 1Gi
  ingress:
    className: "ingress-zero"
    enabled: true
    hosts:
      - temporal.xxx.com
    tls:
      - hosts:
        - temporal.xxx.com
        secretName: ingress-self-signed-tls
prometheus:
  enabled: false
grafana:
  enabled: false

[robholland]

You need to set

postgresql:
   enabled: true

[VLZZZ]

@robholland I'm not sure but I believe that my issue #508 is about the same thing

BTW I've tried adding

postgresql:
   enabled: true

as proposed but it doesn't help as I belive the issue is affecting different part of the chart mechanics