fix up steps

dash-apm-python/step_3.md

@@ -1,7 +1,7 @@
 Add the following environment variable to the `frontend`, `node`, `pumps`,
 and `sensors` services in `docker-compose.yml`.
 
-`DD_ANALYTICS_ENABLED=true`{{copy}}
+`DD_TRACE_ANALYTICS_ENABLED=true`{{copy}}
 
 Our services should look like:
 

dash-apm-python/step_4.md

@@ -19,4 +19,4 @@ Restart services `docker-compose up`{{execute interrupt}}
 
 Finally, open the service page for `frontend` to view the new metadata on traces.
 
-https://app.datadoghq.com/apm/resource/frontend/flask.request/
+https://app.datadoghq.com/apm/service/frontend/flask.request

dash-apm-python/step_6.md

@@ -2,24 +2,25 @@ With APM we can easily visualize where latencies are coming from.
 
 Open `sensors/sensors.py` file in the editor.
 
-Add the following to the `def refresh_sensors` endpoint.
+Add the following to the `def simulate_all_sensors` function.
 
 ```python
-time.sleep(2)
+db.session.execute('SELECT pg_sleep(2);')
 ```{{copy}}
 
 When we are done the function should look like:
 
 ``` python
-@app.route('/refresh_sensors')
-def refresh_sensors():
-    app.logger.info('Calling refresh sensor simulator')
-    sensors = simulate_all_sensors()
-
-    time.sleep(2)
-
-    return jsonify({'sensor_count': len(sensors),
-                    'system_status': sensors})
+@tracer.wrap(name='sensor-simulator')
+def simulate_all_sensors():
+    db.session.execute('SELECT pg_sleep(2);')
+    sensors = Sensor.query.all()
+    for sensor in sensors:
+        sensor.value = random.randint(1,100)
+    db.session.add_all(sensors)
+    db.session.commit()
+    app.logger.info('Sensor data updated')
+    return [s.serialize() for s in sensors]
 ```{{copy}}
 
 Restart our services `docker-compose up`{{execute interrupt}}