delay and timeouts for probes

rnd/autogpt_server/autogpt_server/util/service.py

@@ -33,7 +33,7 @@ class PyroNameServer(AppProcess):
     def run(self):
         try:
             print("Starting NameServer loop")
-            nameserver.start_ns_loop()
+            nameserver.start_ns_loop(host="0.0.0.0", port=9090)
         except KeyboardInterrupt:
             print("Shutting down NameServer")
 
@@ -77,8 +77,8 @@ class AppService(AppProcess):
 
     @conn_retry
     def __start_pyro(self):
-        daemon = pyro.Daemon()
-        ns = pyro.locate_ns()
+        daemon = pyro.Daemon(host="0.0.0.0")
+        ns = pyro.locate_ns(host="0.0.0.0", port=9090)
         uri = daemon.register(self)
         ns.register(self.service_name, uri)
         logger.warning(f"Service [{self.service_name}] Ready. Object URI = {uri}")

rnd/infra/helm/autogpt-server/Chart.yaml

@@ -7,4 +7,4 @@ type: application
 
 version: 0.1.0
 
-appVersion: "1.16.0"
+appVersion: "1.0.0"

rnd/infra/helm/autogpt-server/templates/NOTES.txt

@@ -6,16 +6,16 @@
   {{- end }}
 {{- end }}
 {{- else if contains "NodePort" .Values.service.type }}
-  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "autogpt_server.fullname" . }})
+  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "autogpt-server.fullname" . }})
   export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
   echo http://$NODE_IP:$NODE_PORT
 {{- else if contains "LoadBalancer" .Values.service.type }}
      NOTE: It may take a few minutes for the LoadBalancer IP to be available.
-           You can watch its status by running 'kubectl get --namespace {{ .Release.Namespace }} svc -w {{ include "autogpt_server.fullname" . }}'
-  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ include "autogpt_server.fullname" . }} --template "{{"{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}"}}")
+           You can watch its status by running 'kubectl get --namespace {{ .Release.Namespace }} svc -w {{ include "autogpt-server.fullname" . }}'
+  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ include "autogpt-server.fullname" . }} --template "{{"{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}"}}")
   echo http://$SERVICE_IP:{{ .Values.service.port }}
 {{- else if contains "ClusterIP" .Values.service.type }}
-  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/name={{ include "autogpt_server.name" . }},app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
+  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/name={{ include "autogpt-server.name" . }},app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
   export CONTAINER_PORT=$(kubectl get pod --namespace {{ .Release.Namespace }} $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
   echo "Visit http://127.0.0.1:8080 to use your application"
   kubectl --namespace {{ .Release.Namespace }} port-forward $POD_NAME 8080:$CONTAINER_PORT

rnd/infra/helm/autogpt-server/templates/service.yaml

@@ -4,6 +4,10 @@ metadata:
   name: {{ include "autogpt-server.fullname" . }}
   labels:
     {{- include "autogpt-server.labels" . | nindent 4 }}
+  {{- with .Values.service.annotations }}
+  annotations:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
 spec:
   type: {{ .Values.service.type }}
   ports:

rnd/infra/helm/autogpt-server/values.dev.yaml

@@ -2,7 +2,7 @@
 
 image:
   repository: us-east1-docker.pkg.dev/agpt-dev/agpt-server-dev/agpt-server-dev
-  pullPolicy: IfNotPresent
+  pullPolicy: Always
   tag: "latest"
 
 serviceAccount:
@@ -11,9 +11,11 @@ serviceAccount:
   name: "dev-agpt-server-sa"
 
 service:
-  type: LoadBalancer
-  port: 80
+  type: ClusterIP
+  port: 8000
   targetPort: 8000
+  annotations:
+    cloud.google.com/neg: '{"ingress": true}'
 
 ingress:
   enabled: true
@@ -22,12 +24,16 @@ ingress:
     kubernetes.io/ingress.class: gce
     kubernetes.io/ingress.global-static-ip-name: "agpt-dev-agpt-server-ip"
     networking.gke.io/managed-certificates: "autogpt-server-cert"
+    kubernetes.io/ingress.allow-http: "true"
   hosts:
-    - host: server.agpt.co
+    - host: dev-server.agpt.co
       paths:
-        - path: /*
-          pathType: ImplementationSpecific
-  tls: []
+        - path: /
+          pathType: Prefix
+      backend:
+        service:
+          name: autogpt-server
+          port: 8000
 
 resources:
   requests:
@@ -37,7 +43,24 @@ resources:
     cpu: 500m
     memory: 512Mi
 
-domain: "server.agpt.co"
+livenessProbe:
+  httpGet:
+    path: /docs
+    port: 8000
+  initialDelaySeconds: 30
+  periodSeconds: 10
+  timeoutSeconds: 5
+  failureThreshold: 6
+readinessProbe:
+  httpGet:
+    path: /docs
+    port: 8000
+  initialDelaySeconds: 30
+  periodSeconds: 10
+  timeoutSeconds: 5
+  failureThreshold: 6
+
+domain: "dev-server.agpt.co"
 
 cloudSqlProxy:
   image:

rnd/infra/helm/autogpt-server/values.yaml

@@ -1,6 +1,6 @@
 # base values, environment specific variables should be specified/overwritten in environment values
 
-replicaCount: 2
+replicaCount: 1
 
 image:
   repository: us-east1-docker.pkg.dev/agpt-dev/agpt-server-dev/agpt-server-dev
@@ -74,3 +74,14 @@ tolerations: []
 affinity: {}
 
 domain: ""
+
+cloudSqlProxy:
+  image:
+    repository: gcr.io/cloud-sql-connectors/cloud-sql-proxy
+    tag: 2.11.4
+  instanceConnectionName: ""
+  port: 5432
+  resources:
+    requests:
+      memory: "2Gi"
+      cpu: "1"

rnd/infra/terraform/environments/dev.tfvars

@@ -43,4 +43,13 @@ role_bindings = {
   "roles/iam.workloadIdentityUser" = [
     "serviceAccount:dev-agpt-server-sa@agpt-dev.iam.gserviceaccount.com"
   ]
+  "roles/compute.networkUser" = [
+    "serviceAccount:dev-agpt-server-sa@agpt-dev.iam.gserviceaccount.com"
+  ],
+  "roles/container.hostServiceAgentUser" = [
+    "serviceAccount:dev-agpt-server-sa@agpt-dev.iam.gserviceaccount.com"
+  ]
 }
+
+pods_ip_cidr_range     = "10.1.0.0/16"
+services_ip_cidr_range = "10.2.0.0/20"

rnd/infra/terraform/main.tf

@@ -30,11 +30,13 @@ module "static_ips" {
 module "networking" {
   source = "./modules/networking"
 
-  project_id   = var.project_id
-  region       = var.region
-  network_name = var.network_name
-  subnet_name  = var.subnet_name
-  subnet_cidr  = var.subnet_cidr
+  project_id             = var.project_id
+  region                 = var.region
+  network_name           = var.network_name
+  subnet_name            = var.subnet_name
+  subnet_cidr            = var.subnet_cidr
+  pods_ip_cidr_range     = var.pods_ip_cidr_range
+  services_ip_cidr_range = var.services_ip_cidr_range
 }
 
 module "gke_cluster" {

rnd/infra/terraform/modules/gke_cluster/main.tf

@@ -26,5 +26,10 @@ resource "google_container_cluster" "primary" {
 
   network    = var.network
   subnetwork = var.subnetwork
+
+  ip_allocation_policy {
+    cluster_secondary_range_name  = "pods"
+    services_secondary_range_name = "services"
+  }
 }
 

rnd/infra/terraform/modules/networking/main.tf

@@ -8,5 +8,15 @@ resource "google_compute_subnetwork" "subnet" {
   ip_cidr_range = var.subnet_cidr
   region        = var.region
   network       = google_compute_network.vpc_network.self_link
+
+  secondary_ip_range {
+    range_name    = "pods"
+    ip_cidr_range = var.pods_ip_cidr_range
+  }
+
+  secondary_ip_range {
+    range_name    = "services"
+    ip_cidr_range = var.services_ip_cidr_range
+  }
 }
 

rnd/infra/terraform/modules/networking/variables.tf

@@ -18,4 +18,12 @@ variable "subnet_cidr" {
   description = "The CIDR range for the subnet"
 }
 
+variable "pods_ip_cidr_range" {
+  description = "The IP address range for pods"
+  default     = "10.1.0.0/16"
+}
 
+variable "services_ip_cidr_range" {
+  description = "The IP address range for services"
+  default     = "10.2.0.0/20"
+}

rnd/infra/terraform/modules/static_ip/main.tf

@@ -1,6 +1,5 @@
-resource "google_compute_address" "static_ip" {
+resource "google_compute_global_address" "static_ip" {
   count        = length(var.ip_names)
   name         = "${var.project_id}-${var.ip_names[count.index]}"
-  region       = var.region
   address_type = "EXTERNAL"
 }

rnd/infra/terraform/modules/static_ip/outputs.tf

@@ -1,9 +1,9 @@
 output "ip_addresses" {
   description = "Map of created static IP addresses"
-  value       = { for i, ip in google_compute_address.static_ip : var.ip_names[i] => ip.address }
+  value       = { for i, ip in google_compute_global_address.static_ip : var.ip_names[i] => ip.address }
 }
 
 output "ip_names" {
   description = "List of full names of the created static IP addresses"
-  value       = google_compute_address.static_ip[*].name
+  value       = google_compute_global_address.static_ip[*].name
 }

rnd/infra/terraform/variables.tf

@@ -99,3 +99,15 @@ variable "role_bindings" {
   type        = map(list(string))
   default     = {}
 }
+
+variable "pods_ip_cidr_range" {
+  description = "The IP address range for pods"
+  type        = string
+  default     = "10.1.0.0/16"
+}
+
+variable "services_ip_cidr_range" {
+  description = "The IP address range for services"
+  type        = string
+  default     = "10.2.0.0/20"
+}