diff options
author | Andrew Gauld <ag1282@att.com> | 2017-10-16 14:22:20 -0400 |
---|---|---|
committer | Andrew Gauld <ag1282@att.com> | 2017-10-16 14:27:21 -0400 |
commit | 8a47174c6180efe4c6572222d97dda46189ce736 (patch) | |
tree | 162d7031e70f5f116150c83f426db915868e1af9 | |
parent | d7350c6a9589ce1b386f85ef33f0688d01ce6406 (diff) |
Fill in CDAP blueprint documentation
Change-Id: Ie074327265dc1beafe8d7b05e0a27238f6253e98
Issue-Id: DCAEGEN2-165
Signed-off-by: Andrew Gauld <ag1282@att.com>
-rw-r--r-- | docs/sections/blueprints/cdap.rst | 133 |
1 files changed, 120 insertions, 13 deletions
diff --git a/docs/sections/blueprints/cdap.rst b/docs/sections/blueprints/cdap.rst index f63d5ecd..b707f972 100644 --- a/docs/sections/blueprints/cdap.rst +++ b/docs/sections/blueprints/cdap.rst @@ -1,23 +1,130 @@ CDAP ====================== -Overview of my blueprint and the part it plays in DCAE. +Note: This blueprint is intended to be deployed, automatically, as part of the +DCAE bootstrap process, and is not normally invoked manually. -Blueprint files ----------------------- +The ONAP DCAEGEN2 CDAP blueprint deploys a 7 node Cask Data Application +Platform (CDAP) cluster (version 4.1.x), for running data analysis +applications. The template for the blueprint is at +``blueprints/cdapbp7.yaml-template`` in the ONAP +``dcaegen2.platform.blueprints`` project. The ``02`` VM in the cluster +will be the CDAP master. -List where we can find the blueprints - -Parameters +Blueprint Input Parameters --------------------- -The input parameters needed for running the blueprint +This blueprint has the following required input parameters: -How To Run ---------------------- +* ``ubuntu1604image_id`` -Cfy command for running the blueprint + This is the OpenStack image ID of the Ubuntu 16.04 VM image that will be + used to launch the 7 VMs making up the cluster. -Additional Information ---------------------- -Any additional information that help other people understanding and using yor blueprint +* ``flavor_id`` + + This is the OpenStack flavor ID specifying the amount of memory, disk, and + CPU available to each VM in the cluster. While the required values will be + largely application dependent, a minimum of 32 Gigabytes of memory is + strongly recommended. + +* ``security_group`` + + This is the OpenStack security group specifying permitted inbound and + outbound IP connectivity to the VMs in the cluster. + +* ``public_net`` + + This is the name of the OpenStack network from which floating IP addresses + for the VMs in the cluster will be allocated. + +* ``private_net`` + + This is the name of the OpenStack network from which fixed IP addresses for + the VMs in the cluster will be allocated. + +* ``openstack`` + + This is the JSON object / YAML associative array providing values necessary + for accessing OpenStack. The keys are: + + * ``auth_url`` + + The URL for accessing the OpenStack Identity V2 API. (The version of + Cloudify currently being used, and the associated OpenStack plugin do + not currently support Identity V3). + + * ``tenant_name`` + + The name of the OpenStack tenant/project where the VMs will be launched. + + * ``region`` + + The name of the OpenStack region within the deployment. In smaller + OpenStack deployments, where there is only one region, the region is + often named ``RegionOne``. + + * ``username`` + + The name of the OpenStack user used as a credential for accessing + OpenStack. + + * ``password`` + + The password of the OpenStack user. (The version of Cloudify currently + being used does not provide a mechanism for encrypting this value). + +* ``keypair`` + + The name of the ssh "key pair", within OpenStack, that will be given access, + via the ubuntu login, to the VMs. Note: OpenStack actually stores only the + public key. + +* ``key_filename`` + + The full file path, on the Cloudify Manager VM used to deploy this blueprint, + of the ssh private key file corresponding to the ``keypair`` input parameter. + +* ``location_domain`` + + The DNS domain/zone for DNS entries associated with the VMs in the cluster. + If, for example, location_domain is ``dcae.example.com`` then the FQDN for + a VM with hostname ``abcd`` would be ``abcd.dcae.example.com`` and a DNS + lookup of that FQDN would lead an A (or AAAA) record giving the floating + IP address assigned to that VM. + +* ``location_prefix`` + + The hostname prefix for hostnames of VMs in the cluster. The hostnames + assigned to the VMs are created by concatenating this prefix with a suffix + identifying the individual VMs in the cluster (``cdap00``, ``cdap01``, ..., + ``cdap06``). If the location prefix is ``jupiter`` then the hostname of + the CDAP master in the cluster would be ``jupitercdap02``. + +* ``codesource_url`` and ``codesource_version`` + + ``codesource_url`` is the base URL for downloading DCAE specific project + installation scripts. The intent is that this URL may be environment + dependent, (for example it may, for security reasons, point to an internal + mirror). This is used in combination with the ``codesource_version`` input + parameter to determine the URL for downloading the scripts. There are 2 + scripts used by this blueprint - ``cdap-init.sh`` and + ``instconsulagentub16.sh`` These scripts are part of the + dcaegen2.deployments ONAP project. This blueprint assumes that curl/wget + can find these scripts at + *codesource_url/codesource_version*\ ``/raw/cloud_init/cdap-init.sh`` and + *codesource_url/codesource_version*\ ``/raw/cloud_init/instconsulagentub16.sh`` + respectively. For example, if codesource_url is + ``https://mymirror.example.com`` and codesource_version is ``rel1.0``, + then the installation scripts would be expected to be stored under + ``https://mymirror.example.com/rel1.0/raw/cloud_init/`` + +This blueprint has the following optional inputs: + +* ``location_id`` (default ``solutioning-central``) + + The name of the Consul cluster to register this CDAP cluster with. + +* ``cdap_cluster_name`` (default ``cdap``) + + The name of the service to register this cluster as, in Consul. |