diff --git a/docs/deploy.rst b/docs/deploy.rst
index 1e642c7..9fd944b 100644
--- a/docs/deploy.rst
+++ b/docs/deploy.rst
@@ -1,4 +1,10 @@
 Deploy
 ========
 
-This is where you describe how the project is deployed in production.
+The production deployment for gvaldap is performed using saltstack and consists
+of the following steps:
+
+* installation of native dependencies
+* setup of a virtualenv
+* installation of gvaldap production dependencies inside the virtualenv
+* setup of celery worker under control of supervisord
diff --git a/docs/install.rst b/docs/install.rst
index f0cb69d..23a17ee 100644
--- a/docs/install.rst
+++ b/docs/install.rst
@@ -68,3 +68,17 @@ For production:
     $ pip install -r requirements.txt
 
 .. index:: celery, worker, ldap queue
+
+Running the Celery worker
+=========================
+
+gvaldap uses the `Celery`_ distributed task queue system. The gvaldap logix is
+executed by a celery worker. After all dependencies are installed you can go
+into the gvaldap directory and run the celery worker with:
+
+.. code-block:: sh
+
+    $ cd gvaldap
+    $ celery -A gvaldap worker -Q ldap -l info
+
+.. _Celery: http://www.celeryproject.org/