Subscribe to the feed

You may have heard that Podman V2 has a new RESTful API. This document demonstrates the API using code examples in Python and shell commands. Additional notes are included in the code comments. The provided code was written to be clear vs. production quality.

Requirements

  • You have Python >3.4 installed.
  • You have installed the Python requests library.
  • An IDE for editing Python is recommended.
  • Two terminal windows: one for running the Podman service and reviewing debugging information, the second window to run scripts.
  • Usage of curl and jq commands are demonstrated.
  • You can review connection URLs here.

Getting started

The service

For these examples, we are running the Podman service as a normal user and on an unsecured TCP/IP port number.

For production, the Podman service should use systemd's socket activation protocol. This allows Podman to support clients without additional daemons and secure the access endpoint.

The following command runs the Podman service on port 8080 without timing out. You will need to type ^C into this terminal window when you are finished with the tutorial.

# podman system service tcp:localhost:8080 --log-level=debug --time=0

In addition to the TCP socket demonstrated above, the Podman service supports running under systemd's socket activation protocol and Unix domain sockets (UDS).

[ You might also like: Sneak peek: Podman's new REST API ]

Python code

Info resource

The following shows us information about the Podman service and host:

import json
import requests


response = requests.get("http://localhost:8080/v1.40.0/libpod/info")

Deep dive

  • requests.get() calls the requests library to pass the URL to the Podman service using the GET HTTP method.
    • The requests library provides helper methods for all the popular HTTP methods.
  • http://localhost:8080 matches the Podman service invocation above.
  • /v1.40.0 denotes the API version we are using.
  • /libpod denotes that we expect the service to provide a libpod-specific return payload.
    • Not using this element causes the server to return a compatible payload.
  • /info specifies the resource we are querying.

Interesting to read, but without output, how do we know it worked?

Getting output

Append the lines below, and you can now see the version of Podman running on the host:

response.raise_for_status()


info = json.loads(response.text)
print(info.version.Version)
  • raise_for_status() raises an exception if status code is not between 200 and 399.
  • json.loads() decodes the body of the HTTP response into an object/dictionary.

When executed, the output is:

2.1.0-dev

The following works from the shell:

$ curl -s 'http://localhost:8080/v1.40.0/libpod/info' | jq .version.Version


"2.1.0-dev"

Listing containers

import json
import requests


response = requests.get("http://localhost:8080/v1.40.0/libpod/containers/json?all=true")
response.raise_for_status()


ctnrs = json.loads(response.text)
for c in ctnrs:
    print(c.Id)

json.loads() decodes the HTTP body into an array of objects/dictionaries, the program then prints each container Id. For example:

$ curl -s 'http://localhost:8080/v1.40.0/libpod/containers/json?all=true' | jq .[].Id


"81af11ef7188a826cb5883330525e44afea3ae82634980d68e4e9eefc98d6f61"

If the query parameter all=true had not been provided, then only the running containers would have been listed. The resource queries and parameters for the API are documented here.

Something useful

We've looked at a couple of examples, but how about something a little more useful? You have finished developing the next great container, and the script below will remove everything from your local storage. (If you want to save on typing clean_storage.py)

#!/usr/bin/env python
import json


import requests


# Clean up local storage by removing all containers, pods, and images.  Any error will
#   abort the process


confirm = input("Really delete all items from storage? [y/N] ")
if str(confirm).lower().strip() != 'y':
    exit(0)


# Query for all pods in storage
response = requests.get("http://localhost:8080/v1.40.0/libpod/pods/json")
response.raise_for_status()


pods = json.loads(response.text)
# Workaround for https://github.com/containers/podman/issues/7392
if pods is not None:
    for p in pods:
        # For each container: delete container and associated volumes
        response = requests.delete(f"http://localhost:8080/v1.40.0/libpod/pods/{p['Id']}?force=true")
        response.raise_for_status()
    print(f"Removed {len(pods)} pods and associated objects")
else:
    print(f"Removed 0 pods and associated objects")


# Query for all containers in storage
response = requests.get("http://localhost:8080/v1.40.0/libpod/containers/json?all=true")
response.raise_for_status()


ctnrs = json.loads(response.text)
for c in ctnrs:
    # For each container: delete container and associated volumes
    print(c.keys())
    response = requests.delete(f"http://localhost:8080/v1.40.0/libpod/containers/{c['Id']}?force=true&v=true")
    response.raise_for_status()
print(f"Removed {len(ctnrs)} containers and associated objects")


# Query for all images in storage
response = requests.get("http://localhost:8080/v1.40.0/libpod/images/json")
response.raise_for_status()


imgs = json.loads(response.text)
for i in imgs:
    # For each image: delete image and any associated containers
    response = requests.delete(f"http://localhost:8080/v1.40.0/libpod/images/{i['Id']}?force=true")
    response.raise_for_status()
print(f"Removed {len(imgs)} images and associated objects")

[ Getting started with containers? Check out this free course. Deploying containerized applications: A technical overview. ]

Summary

I hope you find this helpful. The API documentation provides you with all the resources and required methods. The input and output bodies are included as well as the status codes.

The Podman code is under heavy development and we welcome your input with issues and pull requests on the project's GitHub page. 


About the author

Experienced senior software developer/system administrator with 37 years of technical experience. Deep understanding of UNIX/Linux, TCP/IP networking, Internet and software development. Has worked for a diverse set of organizations from large government agencies (NASA, NARA) to 5 person startups. Always applying experience and knowledge to solve the technical challenges of the organization. 
Designing and implementing SAS applications (Cloud) on a RAIN architecture with innovative indexing technology for supporting the business rules and logic. This allows for both high performance and horizontal scaling. Examples: DICOM Grid, Inc, and Red Hat OpenShift.

Read full bio
UI_Icon-Red_Hat-Close-A-Black-RGB

Browse by channel

automation icon

Automation

The latest on IT automation for tech, teams, and environments

AI icon

Artificial intelligence

Updates on the platforms that free customers to run AI workloads anywhere

open hybrid cloud icon

Open hybrid cloud

Explore how we build a more flexible future with hybrid cloud

security icon

Security

The latest on how we reduce risks across environments and technologies

edge icon

Edge computing

Updates on the platforms that simplify operations at the edge

Infrastructure icon

Infrastructure

The latest on the world’s leading enterprise Linux platform

application development icon

Applications

Inside our solutions to the toughest application challenges

Original series icon

Original shows

Entertaining stories from the makers and leaders in enterprise tech