From 9335742140a2b4e8b408560c600a490b2749bd9d Mon Sep 17 00:00:00 2001
From: Aron Gunn <ritz_303@yahoo.com>
Date: Thu, 6 Jul 2017 12:35:38 -0500
Subject: [PATCH] doc/rbd: Added the iSCSI gateway requirements content

Fixes: http://tracker.ceph.com/issues/20437

Signed-off-by: Aron Gunn <agunn@redhat.com>
---
 doc/rbd/iscsi-requirements.rst | 53 ++++++++++++++++++++++++++++++++++
 1 file changed, 53 insertions(+)
 create mode 100644 doc/rbd/iscsi-requirements.rst

diff --git a/doc/rbd/iscsi-requirements.rst b/doc/rbd/iscsi-requirements.rst
new file mode 100644
index 0000000000000..84e456c52211f
--- /dev/null
+++ b/doc/rbd/iscsi-requirements.rst
@@ -0,0 +1,53 @@
+==========================
+iSCSI Gateway Requirements
+==========================
+
+To implement the Ceph iSCSI gateway there are a few requirements. Red
+Hat recommends using two nodes for a highly available Ceph iSCSI gateway
+solution and up to four Ceph iSCSI gateway nodes.
+
+For hardware recommendations, see the `Hardware Recommendation page <http://docs.ceph.com/docs/master/start/hardware-recommendations/>`_
+for more details.
+
+.. WARNING::
+    On the iSCSI gateway nodes, the memory footprint of the RBD images
+    can grow to a large size. Plan memory requirements accordingly based
+    off the number RBD images mapped. Each RBD image roughly uses 90 MB
+    of RAM.
+
+There are no specific iSCSI gateway options for the Ceph Monitors or
+OSDs, but changing the ``osd_client_watch_timeout`` value to ``15`` is
+required for each OSD node in the storage cluster.
+
+-  Online Updating Using the Ceph Monitor
+
+   ::
+
+       ceph tell <daemon_type>.<id> injectargs '--<parameter_name> <new_value>'
+
+   ::
+
+       # ceph tell osd.0 injectargs '--osd_client_watch_timeout 15'
+
+-  Online Updating on the OSD Node
+
+   ::
+
+       ceph daemon <daemon_type>.<id> config set osd_client_watch_timeout 15
+
+   ::
+
+       # ceph daemon osd.0 config set osd_client_watch_timeout 15
+
+   .. NOTE::
+    Update the Ceph configuration file and copy it to all nodes in the
+    Ceph storage cluster. For example, the default configuration file is
+    ``/etc/ceph/ceph.conf``. Add the following lines to the Ceph
+    configuration file:
+
+   ::
+
+        [osd]
+        osd_client_watch_timeout = 15
+
+For more details on setting Ceph's configuration options, see the `Configuration page <http://docs.ceph.com/docs/master/rados/configuration/>`_.
-- 
2.39.5