From 53f2ad3c1633edb1412a639491e1f2cd19f0356f Mon Sep 17 00:00:00 2001
From: Will Anderson <will@neuralplatform.ai>
Date: Sun, 26 Apr 2026 00:28:33 -0500
Subject: [PATCH] Fix Cloud Run probe ports and LB timeout_sec

- accounts: add ACCOUNTS_PORT=8080 env var (service defaults to 7753)
- api: add SERVER_PORT=8080 env var, change probe path to /actuator/health
- LB backends: remove timeout_sec (not supported for serverless NEG backends)
---
 servers/gcp/cloud-run-accounts.tf     | 12 ++++++++++++
 servers/gcp/cloud-run-api.tf          | 24 ++++++++++++++++++------
 servers/gcp/load-balancer-backends.tf |  6 ++----
 3 files changed, 32 insertions(+), 10 deletions(-)

diff --git a/servers/gcp/cloud-run-accounts.tf b/servers/gcp/cloud-run-accounts.tf
index 4c82315..794a0b6 100644
--- a/servers/gcp/cloud-run-accounts.tf
+++ b/servers/gcp/cloud-run-accounts.tf
@@ -43,6 +43,10 @@ resource "google_cloud_run_v2_service" "accounts_us" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "ACCOUNTS_PORT"
+        value = "8080"
+      }
       env {
         name = "ACCOUNTS_DATABASE_URL"
         value_source {
@@ -183,6 +187,10 @@ resource "google_cloud_run_v2_service" "accounts_eu" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "ACCOUNTS_PORT"
+        value = "8080"
+      }
       env {
         name = "ACCOUNTS_DATABASE_URL"
         value_source {
@@ -321,6 +329,10 @@ resource "google_cloud_run_v2_service" "accounts_apac" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "ACCOUNTS_PORT"
+        value = "8080"
+      }
       env {
         name = "ACCOUNTS_DATABASE_URL"
         value_source {
diff --git a/servers/gcp/cloud-run-api.tf b/servers/gcp/cloud-run-api.tf
index 6aadd9f..3fd013b 100644
--- a/servers/gcp/cloud-run-api.tf
+++ b/servers/gcp/cloud-run-api.tf
@@ -42,6 +42,10 @@ resource "google_cloud_run_v2_service" "api_us" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "SERVER_PORT"
+        value = "8080"
+      }
       env {
         name  = "ACCOUNTS_SERVICE_URL"
         value = "https://accounts.neurontechnologies.ai"
@@ -73,7 +77,7 @@ resource "google_cloud_run_v2_service" "api_us" {
 
       startup_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         initial_delay_seconds = 2
@@ -84,7 +88,7 @@ resource "google_cloud_run_v2_service" "api_us" {
 
       liveness_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         timeout_seconds   = 5
@@ -139,6 +143,10 @@ resource "google_cloud_run_v2_service" "api_eu" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "SERVER_PORT"
+        value = "8080"
+      }
       env {
         name  = "ACCOUNTS_SERVICE_URL"
         value = "https://accounts.neurontechnologies.ai"
@@ -170,7 +178,7 @@ resource "google_cloud_run_v2_service" "api_eu" {
 
       startup_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         initial_delay_seconds = 2
@@ -181,7 +189,7 @@ resource "google_cloud_run_v2_service" "api_eu" {
 
       liveness_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         timeout_seconds   = 5
@@ -233,6 +241,10 @@ resource "google_cloud_run_v2_service" "api_apac" {
         name  = "ENV"
         value = "production"
       }
+      env {
+        name  = "SERVER_PORT"
+        value = "8080"
+      }
       env {
         name  = "ACCOUNTS_SERVICE_URL"
         value = "https://accounts.neurontechnologies.ai"
@@ -264,7 +276,7 @@ resource "google_cloud_run_v2_service" "api_apac" {
 
       startup_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         initial_delay_seconds = 2
@@ -275,7 +287,7 @@ resource "google_cloud_run_v2_service" "api_apac" {
 
       liveness_probe {
         http_get {
-          path = "/health"
+          path = "/actuator/health"
           port = 8080
         }
         timeout_seconds   = 5
diff --git a/servers/gcp/load-balancer-backends.tf b/servers/gcp/load-balancer-backends.tf
index a7ab25e..f8ab5d2 100644
--- a/servers/gcp/load-balancer-backends.tf
+++ b/servers/gcp/load-balancer-backends.tf
@@ -57,8 +57,7 @@ resource "google_compute_backend_service" "accounts" {
   project               = var.project_id
   load_balancing_scheme = "EXTERNAL_MANAGED"
   protocol              = "HTTPS"
-  timeout_sec           = 30
-
+  # timeout_sec not supported for serverless NEG backends
   security_policy = google_compute_security_policy.marketing.self_link
 
   # No CDN for accounts — responses are user-specific and must not be cached
@@ -81,8 +80,7 @@ resource "google_compute_backend_service" "api" {
   project               = var.project_id
   load_balancing_scheme = "EXTERNAL_MANAGED"
   protocol              = "HTTPS"
-  timeout_sec           = 60  # API calls may be longer (AI inference routing)
-
+  # timeout_sec not supported for serverless NEG backends
   security_policy = google_compute_security_policy.marketing.self_link
 
   enable_cdn = false  # API responses are per-request; no CDN benefit