Merge pull request #457 from JakobStadlhuber/Readiness-Liveness-Probes

Readiness liveness probes

apps/api/src/controllers/liveness.ts

@@ -0,0 +1,6 @@
+import { Request, Response } from "express";
+
+export async function livenessController(req: Request, res: Response) {
+  //TODO: add checks if the application is live and healthy like checking the redis connection
+  res.status(200).json({ status: "ok" });
+}

apps/api/src/controllers/readiness.ts

@@ -0,0 +1,6 @@
+import { Request, Response } from "express";
+
+export async function readinessController(req: Request, res: Response) {
+  // TODO: add checks when the application is ready to serve traffic
+  res.status(200).json({ status: "ok" });
+}

apps/api/src/routes/v0.ts

@@ -7,6 +7,8 @@ import { crawlJobStatusPreviewController } from "../../src/controllers/status";
 import { searchController } from "../../src/controllers/search";
 import { crawlCancelController } from "../../src/controllers/crawl-cancel";
 import { keyAuthController } from "../../src/controllers/keyAuth";
+import { livenessController } from "../controllers/liveness";
+import { readinessController } from "../controllers/readiness";
 
 export const v0Router = express.Router();
 
@@ -23,3 +25,6 @@ v0Router.get("/v0/keyAuth", keyAuthController);
 // Search routes
 v0Router.post("/v0/search", searchController);
 
+// Health/Probe routes
+v0Router.get("/v0/health/liveness", livenessController);
+v0Router.get("/v0/health/readiness", readinessController);

apps/playwright-service/main.py

@@ -5,7 +5,7 @@ the HTML content of a specified URL. It supports optional proxy settings and med
 
 from os import environ
 
-from fastapi import FastAPI
+from fastapi import FastAPI, Response
 from fastapi.responses import JSONResponse
 from playwright.async_api import Browser, async_playwright
 from pydantic import BaseModel
@@ -39,14 +39,28 @@ async def shutdown_event():
     """Event handler for application shutdown to close the browser."""
     await browser.close()
 
+@app.get("/health/liveness")
+def liveness_probe():
+    """Endpoint for liveness probe."""
+    return JSONResponse(content={"status": "ok"}, status_code=200)
+
+
+@app.get("/health/readiness")
+async def readiness_probe():
+    """Endpoint for readiness probe. Checks if the browser instance is ready."""
+    if browser:
+        return JSONResponse(content={"status": "ok"}, status_code=200)
+    return JSONResponse(content={"status": "Service Unavailable"}, status_code=503)
+
+
 @app.post("/html")
 async def root(body: UrlModel):
     """
     Endpoint to fetch and return HTML content of a given URL.
-    
+
     Args:
         body (UrlModel): The URL model containing the target URL, wait time, and timeout.
-    
+
     Returns:
         JSONResponse: The HTML content of the page.
     """

examples/kubernetes/cluster-install/README.md

@@ -4,12 +4,12 @@
 2. Build Docker images, and host it in your Docker Registry (replace the target registry with your own)
    1. API (which is also used as a worker image)
       1. ```bash
-         docker build -t ghcr.io/winkk-dev/firecrawl:latest ../../apps/api
+         docker build --no-cache -t ghcr.io/winkk-dev/firecrawl:latest ../../../apps/api
          docker push ghcr.io/winkk-dev/firecrawl:latest
          ```
    2. Playwright 
       1. ```bash
-            docker build -t ghcr.io/winkk-dev/firecrawl-playwright:latest ../../apps/playwright-service
+            docker build --no-cache -t ghcr.io/winkk-dev/firecrawl-playwright:latest ../../../apps/playwright-service
             docker push ghcr.io/winkk-dev/firecrawl-playwright:latest
          ```
 3. Replace the image in [worker.yaml](worker.yaml), [api.yaml](api.yaml) and [playwright-service.yaml](playwright-service.yaml)

examples/kubernetes/cluster-install/api.yaml

@@ -15,16 +15,35 @@ spec:
       imagePullSecrets:
         - name: docker-registry-secret
       containers:
-      - name: api
-        image: ghcr.io/winkk-dev/firecrawl:latest
-        args:  [ "pnpm", "run", "start:production" ]
-        ports:
-        - containerPort: 3002
-        envFrom:
-        - configMapRef:
-            name: firecrawl-config
-        - secretRef:
-            name: firecrawl-secret
+        - name: api
+          image: ghcr.io/winkk-dev/firecrawl:latest
+          imagePullPolicy: Always
+          args: [ "pnpm", "run", "start:production" ]
+          ports:
+            - containerPort: 3002
+          envFrom:
+            - configMapRef:
+                name: firecrawl-config
+          #- secretRef:
+          #    name: firecrawl-secret
+          livenessProbe:
+            httpGet:
+              path: /v0/health/liveness
+              port: 3002
+            initialDelaySeconds: 30
+            periodSeconds: 30
+            timeoutSeconds: 5
+            successThreshold: 1
+            failureThreshold: 3
+          readinessProbe:
+            httpGet:
+              path: /v0/health/readiness
+              port: 3002
+            initialDelaySeconds: 30
+            periodSeconds: 30
+            timeoutSeconds: 5
+            successThreshold: 1
+            failureThreshold: 3
 ---
 apiVersion: v1
 kind: Service

examples/kubernetes/cluster-install/configmap.yaml

@@ -7,8 +7,7 @@ data:
   PORT: "3002"
   HOST: "0.0.0.0"
   REDIS_URL: "redis://redis:6379"
-  PLAYWRIGHT_MICROSERVICE_URL: "http://playwright-service:3000"
+  REDIS_RATE_LIMIT_URL: "redis://redis:6379"
+  PLAYWRIGHT_MICROSERVICE_URL: "http://playwright-service:3000/html"
   USE_DB_AUTHENTICATION: "false"
-  SUPABASE_ANON_TOKEN: ""
-  SUPABASE_URL: ""
-  SUPABASE_SERVICE_TOKEN: ""
+  HDX_NODE_BETA_MODE: "1"

examples/kubernetes/cluster-install/playwright-service.yaml

@@ -1,3 +1,10 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: playwright-service-config
+data:
+  PORT: "3000"
+---
 apiVersion: apps/v1
 kind: Deployment
 metadata:
@@ -15,13 +22,32 @@ spec:
       imagePullSecrets:
         - name: docker-registry-secret
       containers:
-      - name: playwright-service
-        image: ghcr.io/winkk-dev/firecrawl-playwright:latest
-        ports:
-        - containerPort: 3000
-        envFrom:
-        - configMapRef:
-            name: firecrawl-config
+        - name: playwright-service
+          image: ghcr.io/winkk-dev/firecrawl-playwright:latest
+          imagePullPolicy: Always
+          ports:
+            - containerPort: 3000
+          envFrom:
+            - configMapRef:
+                name: playwright-service-config
+          livenessProbe:
+            httpGet:
+              path: /health/liveness
+              port: 3000
+            initialDelaySeconds: 30
+            periodSeconds: 30
+            timeoutSeconds: 5
+            successThreshold: 1
+            failureThreshold: 3
+          readinessProbe:
+            httpGet:
+              path: /health/readiness
+              port: 3000
+            initialDelaySeconds: 30
+            periodSeconds: 30
+            timeoutSeconds: 5
+            successThreshold: 1
+            failureThreshold: 3
 ---
 apiVersion: v1
 kind: Service

examples/kubernetes/cluster-install/worker.yaml

@@ -15,10 +15,12 @@ spec:
       imagePullSecrets:
         - name: docker-registry-secret
       containers:
-      - name: worker
-        image: ghcr.io/winkk-dev/firecrawl:latest
-        envFrom:
-        - configMapRef:
-            name: firecrawl-config
-        - secretRef:
-            name: firecrawl-secret
+        - name: worker
+          image: ghcr.io/winkk-dev/firecrawl:latest
+          imagePullPolicy: Always
+          args: [ "pnpm", "run", "workers" ]
+          envFrom:
+            - configMapRef:
+                name: firecrawl-config
+          #- secretRef:
+          #    name: firecrawl-secret