diff --git a/WebHostLib/__init__.py b/WebHostLib/__init__.py
index e4c2ab83c7..f856eea4c5 100644
--- a/WebHostLib/__init__.py
+++ b/WebHostLib/__init__.py
@@ -23,6 +23,17 @@ app.jinja_env.filters['any'] = any
 app.jinja_env.filters['all'] = all
 app.jinja_env.filters['get_file_safe_name'] = get_file_safe_name
 
+# overwrites of flask default config
+app.config["DEBUG"] = False
+app.config["PORT"] = 80
+app.config["UPLOAD_FOLDER"] = UPLOAD_FOLDER
+app.config["MAX_CONTENT_LENGTH"] = 64 * 1024 * 1024  # 64 megabyte limit
+# if you want to deploy, make sure you have a non-guessable secret key
+app.config["SECRET_KEY"] = bytes(socket.gethostname(), encoding="utf-8")
+app.config["SESSION_PERMANENT"] = True
+app.config["MAX_FORM_MEMORY_SIZE"] = 2 * 1024 * 1024  # 2 MB, needed for large option pages such as SC2
+
+# custom config
 app.config["SELFHOST"] = True  # application process is in charge of running the websites
 app.config["GENERATORS"] = 8  # maximum concurrent world gens
 app.config["HOSTERS"] = 8  # maximum concurrent room hosters
@@ -30,19 +41,12 @@ app.config["SELFLAUNCH"] = True  # application process is in charge of launching
 app.config["SELFLAUNCHCERT"] = None  # can point to a SSL Certificate to encrypt Room websocket connections
 app.config["SELFLAUNCHKEY"] = None  # can point to a SSL Certificate Key to encrypt Room websocket connections
 app.config["SELFGEN"] = True  # application process is in charge of scheduling Generations.
-app.config["DEBUG"] = False
-app.config["PORT"] = 80
-app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
-app.config['MAX_CONTENT_LENGTH'] = 64 * 1024 * 1024  # 64 megabyte limit
-# if you want to deploy, make sure you have a non-guessable secret key
-app.config["SECRET_KEY"] = bytes(socket.gethostname(), encoding="utf-8")
 # at what amount of worlds should scheduling be used, instead of rolling in the web-thread
 app.config["JOB_THRESHOLD"] = 1
 # after what time in seconds should generation be aborted, freeing the queue slot. Can be set to None to disable.
 app.config["JOB_TIME"] = 600
 # memory limit for generator processes in bytes
 app.config["GENERATOR_MEMORY_LIMIT"] = 4294967296
-app.config['SESSION_PERMANENT'] = True
 
 # waitress uses one thread for I/O, these are for processing of views that then get sent
 # archipelago.gg uses gunicorn + nginx; ignoring this option