Scalability Enhancements (#14)

* forcing restart of webserver every 10m since gunicorn keeps dying :(
* adding more max connections and shared memory to postgres to connect  more workers
* Increase sql_alchemy_pool_size
* Adding subnet configuration to allow for more ip addresses
* removing dag accidentally checked in from previous merge

docker/config/airflow.cfg

@@ -33,7 +33,7 @@ sql_alchemy_conn = postgresql+psycopg2://airflow:airflow@postgres/airflow
 
 # The SqlAlchemy pool size is the maximum number of database connections
 # in the pool.
-sql_alchemy_pool_size = 5
+sql_alchemy_pool_size = 300
 
 # The SqlAlchemy pool recycle is the number of seconds a connection
 # can be idle in the pool before it is invalidated. This config does

docker/docker-compose-CeleryExecutor.yml

@@ -19,6 +19,7 @@ services:
         deploy:
             placement:
                 constraints: [ engine.labels.infrakit-role == manager ]
+        command: postgres -c max_connections=1000 -c shared_buffers=4096MB
 
     webserver:
         image: wongwill86/air-tasks:latest
@@ -50,6 +51,10 @@ services:
         deploy:
             placement:
                 constraints: [ engine.labels.infrakit-role == manager ]
+            restart_policy:
+                condition: any
+                delay: 10m
+                window: 120s
 
     flower:
         image: wongwill86/air-tasks:latest
@@ -313,3 +318,11 @@ secrets:
         external: true
     ssl_certificate_key:
         external: true
+
+networks:
+    default:
+        attachable: true
+        ipam:
+            driver: default
+            config:
+                - subnet: 10.0.0.0/16