All installations require a central workspace known as the chef-repo. This is a place where primitive objects--cookbooks, roles, environments, data bags, and chef-repo configuration files--are stored and managed.
The chef-repo should be kept under version control, such as git, and then managed as if it were source code.
Knife is the command line interface for Chef. The chef-repo contains a .chef directory (which is a hidden directory by default) in which the Knife configuration file (knife.rb) is located. This file contains configuration settings for the chef-repo.
The knife.rb file is automatically created by the starter kit. This file can be customized to support configuration settings used by cloud provider options and custom knife plugins.
Also located inside the .chef directory are .pem files, which contain private keys used to authenticate requests made to the Chef server. The USERNAME.pem file contains a private key unique to the user (and should never be shared with anyone). The ORGANIZATION-validator.pem file contains a private key that is global to the entire organization (and is used by all nodes and workstations that send requests to the Chef server).
More information about knife.rb configuration options can be found in the documentation for knife.
A cookbook is the fundamental unit of configuration and policy distribution. A sample cookbook can be found in cookbooks/starter. After making changes to any cookbook, you must upload it to the Chef server using knife:
$ knife upload cookbooks/starter
For more information about cookbooks, see the example files in the starter cookbook.
Roles provide logical grouping of cookbooks and other roles. A sample role can be found at roles/starter.rb.
Now that you have the chef-repo ready to go, check out Learn Chef to proceed with your workstation setup. If you have any questions about Chef you can always ask our support team for a helping hand.
##Chef: Day2
Configuration management Other configuratoin management tools Companies requirement to choose which tool they have to use
All these conf.mgmt tools known as infrastructure as code (IFAC) -programatically provision and configure software components.
##Overview of Chef
Chef development kit: it is used by who writes the cookbooks. Chef server: where all the recipes and cookbooks are reside.
Node: machine which is managed by the chef server. Every node machine has to configure so that to which chef server it has to connect and communicate. Each node connects to only one chef server.
Chef-client: it runs on all the nodes, it is a agent runs on all the chef nodes. The nodes connects to the chef server on a regualr interval (default is every 30min) Chef & puppet are pull model conf.mgmt tools.
Run-list : list of cookbooks that needs to be run on a node. roles: Resource: everything that goes in a recipe is known as a resources. Search: searches a resource on a machine.
Resource Types: Networking Files etc..
Organizations: Organization can be a project. The chef server should have atleast one organization. Role: Types of server in your infrastructure. such as Load balancer, app server, db cache etc chef run maintains configuration drift ex: if the node is installed with a software which is not mentioned anywhere in the cookbook then the software will be removed.
once you bootstrap node to chefserver then you cannot install anything manually on that node.
knife: interface between chef-server and workstation.
----------------POST TEA BREAK-----------------
Work Station Setup: Login to the chef enterprise install chefdk on the workstation login to the chef.io and setup a hosted chef server. get a vm or physical machine to use as a chef node. Here we are using an aws instance as a chef node
aws: [email protected] eruser3@2012
ame shiva-ubuntu "bootstrapping a node means adding a node to the chef server so that the cookbooks can be run on that node through the chef server"
Ohai: system profile
-->cat /etc/chef/client.rb
add this line log_level :info
What's a cookbook? A cookbook is like a "package" for chef recipes It contains all the recipes, files, templates etc "chef generate cookbook apache"
Resources are executed in order
Multiphase Execution-Compile phase --During compile phase, Chef *Loads all cookbooks from Run list *Reads all Recipes to build resource collection The order of runlist is mandatory in cookfook fie.
Resource Collection:
Attribute precedence: attribute-->recipe-->environment-->
##Day-3
Problem Statement:
Data bag: A data bag can contain for items that represent info about your infrastructure that is not tied to single node ex: users, groups, applicatiion release information
Every cookbook has to go through these: create cookbook upload cookbook add to run list chef run
##Roles:
-webservers -dbservers -monitoring
###Base Role:
we can include run list in the base role.
##Environments: An organization could have multiple environments. ex: Dev Test Staging Prod etc..
##Chef Supermarket A communitry site where a huge number of pre-written cookbooks are available for free. use knife to work with the dhef supermarket site API download, extract, examine and implement cookbooks from the chef supermarket.
###CHEF SOLO ###CHEF KITCHEN ###CHEF APPLY ###HABITAT ###CHEF AUTOMATE