knife download

Use the knife download subcommand to download roles, cookbooks, environments, nodes, and data bags from the Chef server to the current working directory. It can be used to back up data on the Chef server, inspect the state of one or more files, or to extract out-of-process changes users may have made to files on the Chef server, such as if a user made a change that bypassed version source control. This subcommand is often used in conjunction with knife diff, which can be used to see exactly what changes will be downloaded, and then knife upload, which does the opposite of knife download.

Syntax

This subcommand has the following syntax:

$ knife download [PATTERN...] (options)

Options

This subcommand has the following options:

--chef-repo-path PATH

The path to the chef-repo. This setting will override the default path to the chef-repo. Default: same value as specified by chef_repo_path in client.rb.

--concurrency

The number of allowed concurrent connections. Default: 10.

--cookbook-version VERSION

The version of a cookbook to download.

-n, --dry-run

Take no action and only print out results. Default: false.

--[no-]diff

Download only new and modified files. Set to false to download all files. Default: --diff.

--[no-]force

Use --force to download files even when the file on the hard drive is identical to the object on the server (role, cookbook, etc.). By default, files are compared to see if they have equivalent content, and local files are only overwritten if they are different. Default: --no-force.

--[no-]purge

Use --purge to delete local files and directories that do not exist on the Chef server. By default, if a role, cookbook, etc. does not exist on the Chef server, the local file for said role is left alone and NOT deleted. Default: --no-purge.

--[no-]recurse

Use --no-recurse to disable downloading a directory recursively. Default: --recurse.

--repo-mode MODE

The layout of the local chef-repo. Possible values: static, everything, or hosted_everything. Use static for just roles, environments, cookbooks, and data bags. By default, everything and hosted_everything are dynamically selected depending on the server type. Default: everything / hosted_everything.

Examples

The following examples show how to use this knife subcommand:

Download the entire chef-repo

To download the entire chef-repo from the Chef server, browse to the top level of the chef-repo and enter:

$ knife download /

Download the /cookbooks directory

To download the cookbooks/ directory from the Chef server, browse to the top level of the chef-repo and enter:

$ knife download cookbooks

or from anywhere in the chef-repo, enter:

$ knife download /cookbooks

Download the /environments directory

To download the environments/ directory from the Chef server, browse to the top level of the chef-repo and enter:

$ knife download environments

or from anywhere in the chef-repo, enter:

$ knife download /environments

Download an environment

To download an environment named “production” from the Chef server, browse to the top level of the chef-repo and enter:

$ knife download environments/production.json

or from the environments/ directory, enter:

$ knife download production.json

Download the /roles directory

To download the roles/ directory from the Chef server, browse to the top level of the chef-repo and enter:

$ knife download roles

or from anywhere in the chef-repo, enter:

$ knife download /roles

Download cookbooks and roles

To download all cookbooks that start with “apache” and belong to the “webserver” role, browse to the top level of the chef-repo and enter:

$  knife download cookbooks/apache\* roles/webserver.json