python-jamf 0.4.7

Python wrapper for Jamf Pro API

python-jamf

This is a Python 3 utility for maintaining & automating Jamf Pro patch management via command-line. The idea behind it is to have a class that maps directly to the Jamf API (https://example.com:8443/api). The API class doesn't abstract anything or hide anything from you. It simply wraps the url requests, authentication, and converts between python dictionaries and xml. It also prints json.

Requirements

This utility has been tested on macOS 10.14, macOS 10.15, macOS 11, and CentOS 7.

The python-jamf project requires python3. Please make sure you have that by running the following command.

python --version

or

python3 --version

macOS does not include python3. You can get python3 with Anaconda or Homerew. For example, this is how you install python3 with Homebrew.

brew install python3

Installation

To install python-jamf globally:

sudo pip3 install python-jamf

To install it locally for the current user:

pip3 install python-jamf --user

If you have /usr/local/bin/plistlib.py make sure it is the python 3 version.

If you don't want to install python-jamf globally you will also need to install requests first.

pip3 install requests
pip3 install python-jamf

Test

To test your install, start python3's REPL.

python3

Create an api.

import jamf
api = jamf.API()

Enter your credentials (it is only interactive if you don't have a config file--see below).

Hostname (don't forget https:// and :8443): https://example.com:8443
username: james
Password:

Then pull some data from the server and print it out.

from pprint import pprint
pprint(api.get('accounts'))

You should see something like this.

{'accounts': {'groups': None,
			'users': {'user': [{'id': '2', 'name': 'james'},
								{'id': '1', 'name': 'root'}]}}}

Are you exited? Try getting these as well.

pprint(api.get('computers'))
pprint(api.get('computergroups'))
pprint(api.get('policies'))
pprint(api.get('categories'))

You can view all of the things you can query by going to this url on your jamf server. https://example.com:8443/api/

Config file

The config file can be setup several ways.

First, you can download jctl and run the setconfig.py script. Please see that project for instructions.

Or you can use the JSSImporter/python-jss configuration.

If you don't want to do either of these methods, this script will also look for /Library/Preferences/com.jamfsoftware.jamf.plist and grab the server from there and just ask for username and password.

Or you can specify it in code. By specifying any of the connection settings in code, the config file will be ignored. You either have to specify hostname, username, and password, or you have to pass in promt=True to get it to prompt if you don't specify one of the required parameters (hostname, username, password).

python3

This specifies all of the credentials

import jamf
api = jamf.API(hostname='https://example.com:8443', username='james', password='secret')

Or to prompt for the password, use this.

import jamf
api = jamf.API(hostname='https://example.com:8443', username='james', prompt=True)

Note, on Linux, the config file is stored as a plist file at ~/.edu.utah.mlib.jamfutil.plist

Using the API

The API script interacts with Jamf using the get, post, put, and delete commands in combination with the API resources. To see all of your resources, go to the following URL on your server. https://example.com:8443/api

The api can be interacted with via python3 shell. This is how you set it up.

python3

from pprint import pprint
import jamf
api = jamf.API()

Getting data

Note: The API get method downloads the data from Jamf. If you store it in a variable, it does not update itself. If you make changes on the server, you'll need to run the API get again.

Get any information from your jamf server using the classic api endpoints. This includes nested dictionaries.

pprint(api.get('accounts'))
pprint(api.get('buildings'))
pprint(api.get('categories'))
pprint(api.get('computergroups'))
pprint(api.get('computers'))
pprint(api.get('departments'))
pprint(api.get('licensedsoftware'))
pprint(api.get('networksegments'))
pprint(api.get('osxconfigurationprofiles'))
pprint(api.get('packages'))
pprint(api.get('patches'))
pprint(api.get('policies'))
pprint(api.get('scripts'))

Get all categories (and deal with the nested dictionaries)

categories = api.get('categories')['categories']['category']
category_names = [x['name'] for x in categories]
print(f"first category: {category_names[0]}")
pprint(category_names)

Get computer management information (this demonstrates using an id in the get request)

computers = api.get('computers')['computers']['computer']
pprint(computers[0])
pprint(api.get(f"computermanagement/id/{computers[0]['id']}"))
pprint(api.get(f"computermanagement/id/{computers[0]['id']}/subset/general"))

Getting computer groups and filtering using list comprehension filtering.

computergroups = api.get('computergroups')['computer_groups']['computer_group']
smartcomputergroups = [i for i in computergroups if i['is_smart'] == 'true']
pprint(smartcomputergroups)
staticcomputergroups = [i for i in computergroups if i['is_smart'] != 'true']
pprint(staticcomputergroups)
computergroupids = [i['id'] for i in computergroups]
pprint(computergroupids)

Posting data

Create a new static computer group. Note, the id in the url ("1") is ignored and the next available id is used. The name in the url ("ignored") is also ignored and the name in the data ("realname") is what is actually used.

import json
api.post("computergroups/id/1",json.loads( '{"computer_group": {"name": "test", "is_smart": "false", "site": {"id": "-1", "name": "None"}, "criteria": {"size": "0"}, "computers": {"size": "0"}}}' ))
api.post("computergroups/name/ignored",json.loads( '{"computer_group": {"name": "realname", "is_smart": "false", "site": {"id": "-1", "name": "None"}, "criteria": {"size": "0"}, "computers": {"size": "0"}}}' ))

Updating data

This changes the group "realname" created above to "new name".

import json
api.put("computergroups/name/realname",json.loads( '{"computer_group": {"name": "new name", "is_smart": "false", "site": {"id": "-1", "name": "None"}, "criteria": {"size": "0"}, "computers": {"size": "0"}}}' ))

This is how you'd get the id.

computergroups = api.get('computergroups')['computer_groups']['computer_group']
newgroup = [i for i in computergroups if i['name'] == 'new name']

And this is how to change the name by id.

api.put(f"computergroups/id/{newgroup[0]['id']}",json.loads( '{"computer_group": {"name": "newer name", "is_smart": "false", "site": {"id": "-1", "name": "None"}, "criteria": {"size": "0"}, "computers": {"size": "0"}}}' ))

Deleting data

This deletes the 2 groups we just created.

api.delete("computergroups/name/test")
api.delete(f"computergroups/id/{newgroup[0]['id']}")

Updating policies en masse

This is where the real power of this utility comes in.

The following example searches all policies for the custom trigger "update_later" and replaces it with "update_now".

#!/usr/bin/env python3

import jamf

api = jamf.API()
all_policies = api.get('policies')['policies']['policy']
for policy_hook in all_policies:
    policy = api.get(f"policies/id/{policy_hook['id']}")
    name = policy['policy']['general']['name']
    custom_trigger = policy['policy']['general']['trigger_other']
    print(f"Working on {name}")
    if (custom_trigger == "update_later"):
        policy['policy']['general']['trigger_other'] = "update_now"
        api.put(f"policies/id/{policy_hook['id']}", policy)
        print(f"Changed custom trigger from {custom_trigger} to update_now")

The next example prints out the code you'd need to enter into a python3 repl to set the custom_triggers. Save the output of this script to a file, then edit the file with the custom triggers you want for each item. Delete the items you don't want to change.

#!/usr/bin/env python3

import jamf

api = jamf.API()
all_policies = api.get('policies')['policies']['policy']

print("""#!/usr/bin/env python3

import jamf

api = jamf.API()
""")

for policy_hook in all_policies:
    policy = api.get(f"policies/id/{policy_hook['id']}")
    custom_trigger = policy['policy']['general']['trigger_other']
    print(f"print(f\"Working on {policy['policy']['general']['name']}\")\n"
          f"policy = api.get(\"policies/id/{policy_hook['id']}\")\n"
          f"policy['policy']['general']['trigger_other'] = "
          f"\"{policy['policy']['general']['trigger_other']}\"\n"
          f"api.put(\"policies/id/{policy_hook['id']}\", policy)\n\n")

Save the script as "custom_triggers_1.py" Run it like this.

./custom_triggers_1.py > custom_triggers_2.py
chmod 755 custom_triggers_2.py

Then edit custom_triggers_2.py with the custom triggers you want (and remove what you don't want to modify). Then run custom_triggers_2.py.

Categories

from jamf.category import Categories
allcategories = Categories()
allcategories.names()
allcategories.ids()
allcategories.categoryWithName("Utilities")
allcategories.categoryWithId(141)

for item in allcategories:
    repr(item)

category = Categories().find("Utilities")
repr(category)
category = Categories().find(141)
repr(category)

Running Tests

The following doesn't work as of 2020/12.

cd python-jamf

# runs all tests
python3 -m unittest discover -v

# run tests individually
python3 -m python-jamf.tests.test_api
python3 -m jamf.tests.test_config
python3 -m jamf.tests.test_convert
python3 -m jamf.tests.test_package

If you see an error that says something like SyntaxError: invalid syntax, check to see if you're using python3.

Contributers

• Sam Forester
• James Reynolds
• Topher Nadauld
• Tony Williams