christinedraper / knife-topo

A knife plugin to manage topologies of nodes

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

knife topo

The knife-topo plugin allows you to create and update topologies consisting of multiple nodes using single knife commands, based on a JSON definition of the topology. The plugin:

  • creates a data bag for the topology
  • generates attribute file(s) in a topology-specific cookbook
  • sets and updates the run list, chef environment and properties of nodes
  • uploads the topology-specific cookbook and bootstraps nodes

You may find this plugin useful if you are regularly updating a system consisting of multiple nodes, and would like to manage its dynamic configuration (e.g. changing software versions) through a single (json) configuration file. It may also be useful if you are regularly bringing up multi-node systems with similar topologies but differences in their configuration details.

This plugin can be used in conjunction with the topo cookbook to configure dynamically deployed topologies of nodes (e.g. in AWS). In this approach, use knife-topo to set up the topology details. Use the 'topo' cookbook to let nodes pull their own detailed configuration from the topology data bag on the Chef server.

Changes from V1

Attribute setting strategy

V2 introduces the notion of an attribute setting strategy. Instead of specifying node attributes and cookbook attributes separately, you specify one set of attributes and the method by which they should be set on the node (e.g. 'direct_to_node' or 'via_cookbook').

The purpose of this change is to make it easier to support new methods of setting attributes such as policy files, and also making it easier to switch between methods.

Node type

A node type can be specified for a node. The node type is used by the 'topo' cookbook to identify the right configuration to use. It is also used in the 'via cookbook' method to support attributes that vary by node type.

One topology per file

To reduce complexity, V2 no longer supports multiple topologies in a JSON file.

V1 -> V2 Migration

V1 topology JSON files can be converted to V2 by importing them and then exporting them. knife-topo will auto-detect V1 format files on import. You can also explicitly specify the input format using knife topo import sometopo.json --input-format 'topo_v1'

Installation

Install knife-topo as a gem

  $ chef gem install knife-topo

Usage

Define each topology in a topology file. Import that file into your Chef workspace using knife topo import, then create the topology knife topo create, specifying the '--bootstrap' option if you want to bootstrap all of the nodes.

Update the topology file as the configuration changes, import those changes knife topo import and run knife topo update to update the topology.

Getting Started

Try out this plugin using a test-repo provided in the knife-topo github repository. Download the latest knife-topo release and unzip it, then follow the Instructions for the example.

The instructions assume you have chefDK installed and working with Vagrant and VirtualBox.

If you're the sort of person who just wants to jump in and try it, here's some hints.

Generate a topology file for a topology called test1 from existing nodes node1 and node2:

knife topo export node1 node2 --topo test1 > test1.json

Import a topology json file, generating all of the necessary artifacts in your workspace:

knife topo import test1.json

Create the topology using existing nodes:

knife topo create test1

Create the topology, bootstrapping new nodes in vagrant (you will need to add the host details for bootstrap to the file before importing):

knife topo create test1 --bootstrap -xvagrant -Pvagrant --sudo 

Topology File #

See the example topology file

The topology file contains a single topology. Each topology has some overall properties, an array of nodes and an array defining topology cookbook attributes.

Overall Topology Properties

    {
        "name": "test1",
        "node_type": "appserver",
        "chef_environment": "test",
        "tags": [ "testsys" ],
        "normal": {
          "owner": {
            "name": "Christine Draper"
          }
        },
        "nodes" : [
          ...
        ]
    }

The name is how you will refer to the topology in the knife topo subcommands.

The chef-environment and normal attributes defined here will be applied to all nodes in the topology, unless alternative values are provided for a specific node. The tags will be added to each node.

Attribute setting strategy

The default strategy for setting attributes is direct_to_node. In this strategy, normal attributes are set directly on the nodes when the topology is created, bootstrapped or updated. Attributes with other priorities are ignored.

The strategy field can be set to 'via_cookbook', in which case additional strategy_data can be provided to specify a cookbok and attribute filename.

    {
        "name": "test1",
        ...
        "strategy" : "via_cookbook",
        "strategy_data": {
          "cookbook": "topo_test1",
          "filename": "softwareversion",
      }
    }

In this strategy, the cookbook and attribute file are generated in the local workspace when the topology is imported, and uploaded to the server when the topology is created or updated. Attributes can have any valid priority.

Node List

Each topology contains a list of nodes.

    {
        "name": "test1",
        ...
        "nodes": [
           {
              "name": "buildserver01",
              "node_type" : "buildserver",
              "ssh_host": "192.168.1.201",
              "ssh_port": "2224",
              "chef_environment": "dev",
              "run_list": [ 
                "role[base-ubuntu]", 
                "ypo::db", 
                "recipe[ypo::appserver]"
              ],
              ... node attributes, see below ...,
              "tags": [ "build" ]
            },
            ...
        ]
    }

Within nodes, the name field is the node name that will be used in Chef. The fields chef_environment, run_list and tags will also be applied to the node in Chef. All of these fields are optional.

The node_type sets the node attribute normal['topo']['node_type']. This attribute is used in the 'via_cookbook' strategy to specify attributes that apply to only nodes of that type.

The ssh_host and ssh_port fields are used to bootstrap a node.

Node Attributes

Each topology may have attributes that are set on each node according to the attribute setting strategy. The attribute names and values are specified by priority ('default', 'normal', 'override').

	"nodes": [
		{
      "name": "buildserver01",
			"normal": 
			{			
				"nodejs": 
				{
					"version": "0.10.40"
				},

				"testapp": 
				{
					"version": "0.0.3"
				},

				"mongodb": 
				{
					"package_version": "2.6.9"
				}
			}
		}
	]

Subcommands

The main subcommands for knife topo are:

The additional subcommands can also be useful, depending on your workflow:

The topologies are data bag items in the 'topologies' data bag, so you can also use knife commands such as:

  • knife data bag show topologies test1 - Show details of the test1 topology data bag

Common Options:

The knife topo subcommands support the following common options.

Option Description
-D, --data-bag DATA_BAG The data bag to use for the topologies. Defaults to 'topologies'.

knife topo bootstrap

knife topo bootstrap TOPOLOGY

Runs the knife bootstrap command for each node in the topology that has the ssh_host attribute. Specified options will be passed through to knife bootstrap and applied to each node.

Options:

The knife topo bootstrap subcommand supports the following additional options.

Option Description
--overwrite Re-bootstrap existing nodes
See knife bootstrap Options supported by knife bootstrap are passed through to the bootstrap command

Examples:

The following will bootstrap nodes in the test1 topology, using a user name of vagrant, password of vagrant, and running using sudo.

$ knife topo bootstrap test1 -x vagrant -P vagrant --sudo

knife topo create

knife topo create TOPOLOGY

Creates the specified topology in the chef server as an item in the topology data bag. Creates the chef environment associated with the topology, if it does not already exist. Uploads the topology cookbook, if using the 'via_cookbook' method. Updates existing nodes based on the topology information. New nodes will be created if the bootstrap option is specified.

Options:

The knife topo create subcommand supports the following additional options.

Option Description
--bootstrap Bootstrap the topology (see topo bootstrap)
See knife bootstrap Options supported by knife bootstrap are passed through to the bootstrap command
--disable-upload Do not upload topology cookbooks
--overwrite Re-bootstrap existing nodes

Examples:

The following will create the 'test1' topology, and bootstrap it.

$ knife topo create test1 --bootstrap

The following will create the 'test1' topology but will not bootstrap it or upload topology cookbooks.

$ knife topo create test1 --disable-upload

knife topo delete

knife topo delete TOPOLOGY

Deletes the specified topology. Does not delete the nodes in the topology, but does remove them from the topology by removing the ['topo']['name'] attribute which is used by knife topo search.

knife topo export

knife topo export NODE [ NODE ... ] 

Exports the nodes into a topology JSON.

If the topology does not already exist, an outline for a new topology will be exported. The exported JSON can be used as the basis for a new topology definition.

If nodes are specified, these will be exported in addition to any nodes that are in the topology.

If no topology is specified, all defined topologies will be exported.

Options:

The knife topo export subcommand supports the following additional options.

Option Description
--topo Name of the topology to export (defaults to 'topo1')
--min-priority Only export attributes with a priority equal or above this priority.

Examples:

The following will export the data for nodes n1 and n2 as part of a topology called 'my_topo':

$ knife topo export n1 n2 --topo my_topo > my_topo.json

The following will create an outline for a new topology called
'christine_test', or export the current details if it already exists:

$ knife topo export --topo christine_test > christine_test.json

knife topo import

knife topo import [ TOPOLOGY_FILE ] 

Imports data bag items containing the topologies from a topology file into the local repo. If no topology file is specified, attempts to read from a file called 'topology.json' in the current directory. Generates additional artifacts (e.g. topology cookbook attribute file) where needed.

Examples:

The following will import the topology or topologies defined in the 'topology.json' file.

$ knife topo import topology.json

The following will import the 'test1' topology defined in the 'topology.json' file.

$ knife topo import topology.json test1

knife topo list

knife topo list

Lists the topologies that have been created on the server.

knife topo search

knife topo search [ QUERY ]

Searches for nodes that are in a topology and satisfy the query. With no options, this searches for nodes in any topology. Use --topo=topo_name to search within a specific topology. Use --no-topo to search for nodes in no topology.

knife topo search uses the ['topo']['name'] attribute to identify which nodes are in which topology.

Examples:

The following will search for nodes in any topology that have a name starting with "tst".

$ knife topo search "name:tst*"

The following will search for nodes in the "prod" chef environment that are not in a topology.

$ knife topo search "chef_environment:prod" --no-topo

The following will search for all nodes in the "systest" topology.

$ knife topo search --topo=systest

Options:

The knife topo search subcommand supports the following additional options.

Option Description
--topo Search for nodes in the specified topology
--no-topo Search for nodes that are not in any topology
See knife search Options supported by knife search are passed through to the search command

knife topo update

knife topo update [ TOPOLOGY ] 

Updates the specified topology. Creates or updates nodes identified in the topology, using information specified in the topology for the specific node.

If no topology is specified, all existing topologies will be updated.

Option Description
--bootstrap Bootstrap the topology (see topo bootstrap)
See knife bootstrap Options supported by knife bootstrap are passed through to the bootstrap command
--disable-upload Do not upload topology cookbooks

Examples:

The following will update the 'test1' topology.

$ knife topo update test1

The following will update all topologies in the 'topologies' data bag.

$ knife topo update

License

Author:: Christine Draper (christine_draper@thirdwaveinsights.com)

Copyright:: Copyright (c) 2014-2016 ThirdWave Insights, LLC

License:: Apache License, Version 2.0

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

About

A knife plugin to manage topologies of nodes

License:Apache License 2.0


Languages

Language:Ruby 98.4%Language:HTML 0.8%Language:JavaScript 0.8%