From d0447fb69ffb7de57d8b4c6b3a7c8b6db07a6e13 Mon Sep 17 00:00:00 2001
From: hailin <hailin.zhao@gdzx.xyz>
Date: Fri, 20 Feb 2026 00:13:47 -0800
Subject: [PATCH] fix: use node/python HTTP healthchecks instead of wget

wget returns error on 404, but services are healthy (just no root
endpoint). Using node http.get for NestJS services (accepts any
non-5xx response) and python urllib for voice-service.

Also upgraded api-gateway depends_on to service_healthy.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 deploy/docker/docker-compose.yml | 32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/deploy/docker/docker-compose.yml b/deploy/docker/docker-compose.yml
index 100ada9..5917cd8 100644
--- a/deploy/docker/docker-compose.yml
+++ b/deploy/docker/docker-compose.yml
@@ -43,19 +43,19 @@ services:
       - "18001:8001"
     depends_on:
       auth-service:
-        condition: service_started
+        condition: service_healthy
       agent-service:
-        condition: service_started
+        condition: service_healthy
       ops-service:
-        condition: service_started
+        condition: service_healthy
       inventory-service:
-        condition: service_started
+        condition: service_healthy
       monitor-service:
-        condition: service_started
+        condition: service_healthy
       comm-service:
-        condition: service_started
+        condition: service_healthy
       audit-service:
-        condition: service_started
+        condition: service_healthy
     healthcheck:
       test: ["CMD", "kong", "health"]
       interval: 10s
@@ -86,7 +86,7 @@ services:
       - JWT_REFRESH_SECRET=${JWT_REFRESH_SECRET:-dev-jwt-refresh-secret}
       - AUTH_SERVICE_PORT=3001
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3001/api/v1/auth || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3001/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -120,7 +120,7 @@ services:
       - AGENT_ENGINE_TYPE=claude_code_cli
       - AGENT_SERVICE_PORT=3002
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3002/api/v1/agent || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3002/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -152,7 +152,7 @@ services:
       - REDIS_URL=redis://redis:6379
       - OPS_SERVICE_PORT=3003
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3003/api/v1/ops || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3003/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -185,7 +185,7 @@ services:
       - VAULT_MASTER_KEY=${VAULT_MASTER_KEY:-dev-vault-key}
       - INVENTORY_SERVICE_PORT=3004
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3004/api/v1/inventory || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3004/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -215,7 +215,7 @@ services:
       - REDIS_URL=redis://redis:6379
       - MONITOR_SERVICE_PORT=3005
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3005/api/v1/monitor || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3005/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -248,7 +248,7 @@ services:
       - TWILIO_PHONE_NUMBER=${TWILIO_PHONE_NUMBER}
       - COMM_SERVICE_PORT=3006
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3006/api/v1/comm || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3006/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -280,7 +280,7 @@ services:
       - REDIS_URL=redis://redis:6379
       - AUDIT_SERVICE_PORT=3007
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3007/api/v1/audit || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3007/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -304,7 +304,7 @@ services:
       - KOKORO_MODEL=${KOKORO_MODEL:-kokoro-82m}
       - DEVICE=${VOICE_DEVICE:-cpu}
     healthcheck:
-      test: ["CMD-SHELL", "curl -sf http://localhost:3008/docs || exit 1"]
+      test: ["CMD-SHELL", "python3 -c \"import urllib.request; urllib.request.urlopen('http://localhost:3008/docs')\""]
       interval: 30s
       timeout: 5s
       retries: 3
@@ -326,7 +326,7 @@ services:
       - NEXT_PUBLIC_API_BASE_URL=/api/proxy
       - NEXT_PUBLIC_WS_URL=ws://localhost:18000
     healthcheck:
-      test: ["CMD-SHELL", "wget -qO- http://localhost:3000/ || exit 1"]
+      test: ["CMD-SHELL", "node -e \"require('http').get('http://localhost:3000/',r=>{process.exit(r.statusCode<500?0:1)}).on('error',()=>process.exit(1))\""]
       interval: 30s
       timeout: 5s
       retries: 3