Support setting Kafka storage volume

Not everyone wants Kafka data stored on a Docker volume. This
change allows a user to flexibly control where the data is stored.

Change-Id: I2ba8c7a85c7bf2564f954a43c6e6dbb3257fe902

2 files changed, 26 insertions, 0 deletions

diff --git a/doc/source/reference/logging-and-monitoring/index.rst b/doc/source/reference/logging-and-monitoring/index.rst
index a2cfbd0c9..4aa730c2b 100644
--- a/doc/source/reference/logging-and-monitoring/index.rst
+++ b/doc/source/reference/logging-and-monitoring/index.rst
@@ -10,6 +10,7 @@ logging and monitoring services available in kolla.
 
    central-logging-guide
    influxdb-guide
+   kafka-guide
    monasca-guide
    osprofiler-guide
    prometheus-guide
diff --git a/doc/source/reference/logging-and-monitoring/kafka-guide.rst b/doc/source/reference/logging-and-monitoring/kafka-guide.rst
new file mode 100644
index 000000000..2867e9f2f
--- /dev/null
+++ b/doc/source/reference/logging-and-monitoring/kafka-guide.rst
@@ -0,0 +1,25 @@
+.. _kafka-guide:
+
+============
+Apache Kafka
+============
+
+Overview
+~~~~~~~~
+
+`Kafka <https://kafka.apache.org/intro>`_ is a distributed stream processing
+system. It forms the central component of Monasca and in an OpenStack context
+can also be used as an experimental messaging backend in `Oslo messaging
+<https://docs.openstack.org/oslo.messaging/latest/admin/kafka.html>`_.
+
+Kafka
+~~~~~
+
+A spinning disk array is normally sufficient for Kafka. The data directory
+defaults to a docker volume, ``kafka``. Since it can use a lot of disk space,
+you may wish to store the data on a dedicated device. This can be achieved by
+setting ``kafka_datadir_volume`` in ``/etc/kolla/globals.yml``:
+
+.. code-block:: yaml
+
+    kafka_datadir_volume: /mnt/spinning_array/kafka/