Merge pull request openshift#2385 from iamemilio/diskRequirements

Document disk speed requirements

Author: openshift-merge-robot

docs/user/openstack/README.md

@@ -13,6 +13,7 @@ In addition, it covers the installation with the default CNI (OpenShiftSDN), as
     - [Worker Nodes](#worker-nodes)
     - [Bootstrap Node](#bootstrap-node)
     - [Swift](#swift)
+    - [Disk Requirements](#disk-requirements)
     - [Red Hat Enterprise Linux CoreOS (RHCOS)](#red-hat-enterprise-linux-coreos-rhcos)
     - [Neutron Public Network](#neutron-public-network)
   - [OpenStack Credentials](#openstack-credentials)
@@ -90,6 +91,10 @@ openstack role add --user <user> --project <project> swiftoperator
 openstack object store account set --property Temp-URL-Key=superkey
 ```
 
+### Disk Requirements
+
+Etcd is run on your control plane nodes, and it has disk requirements that need to be met to ensure the stability of your cluster. It is important that you make sure that the ephemeral disk that gets attached to instances of your chosen flavor meets [these requirements](https://github.com/etcd-io/etcd/blob/master/Documentation/op-guide/hardware.md#disks). If this is not the case, then please refer to the known issue documentation [here](known-issues.md#boot-from-volume-support).
+
 ### Red Hat Enterprise Linux CoreOS (RHCOS)
 
 Get the latest RHCOS image [here](https://mirror.openshift.com/pub/openshift-v4/dependencies/rhcos/pre-release/latest/). The installer requires a proper RHCOS image in the OpenStack cluster or project:

docs/user/openstack/known-issues.md

@@ -30,3 +30,7 @@ clusterNetwork: 10.128.0.0/14
 The [account in url](https://docs.ceph.com/docs/master/radosgw/config-ref/#swift-settings) option must be enabled in Ceph RGW in order to use public-readable containers needed for temporary URLs.
 
 This setting wasn't enabled by default [prior to OSP15](https://bugs.launchpad.net/tripleo/+bug/1826894). We are working on removing the need for Swift temp-url in [PR #2311](https://github.com/openshift/installer/pull/2311).
+
+## Boot From Volume Support
+
+OpenStack clouds may have disk pools of various speeds, and the pool that the ephemeral disk is assigned from may not be ideal for customers to deploy OpenShift on. In order to address this, and the issues that may arise from it, we plan to support booting instances from volumes in an upcoming release. If this issue affects you, you can track the lates progress in [this bugzilla](https://bugzilla.redhat.com/show_bug.cgi?id=1754070).