RHCSA Rapid Track

Before you can run your application in a container, you must create a container image.

With Podman, you fetch container images from image registries by using the podman pull command. For example, the following command fetches a containerized version of Red Hat Enterprise Linux 9 from the Red Hat Registry.

[user@host ~]$ podman pull registry.redhat.io/rhel9/rhel-guest-image:9.4
Trying to pull registry.redhat.io/rhel9/rhel-guest-image:9.4...
Getting image source signatures
...output omitted...
Writing manifest to image destination
Storing signatures
b85986059f7663c1b89431f74cdcb783f6540823e4b85c334d271f2f2d8e06d6

The containerized RHEL 9 image is referenced in the form NAME:VERSION. In the following example, you fetch the 9.4 version of the registry.redhat.io/rhel9/rhel-guest-image image.

After you run the pull command, the image is stored locally in your system. You can list the images in your system by using the podman images command.

[user@host ~]$ podman images
REPOSITORY                        TAG         IMAGE ID      CREATED       SIZE
registry.redhat.io/rhel9/rhel-guest-image     9.4         52617ef413bd  2 weeks ago   216 MB

A container is an isolated runtime environment where applications are executed as isolated processes. The isolation of the runtime environment ensures that they do not interrupt other containers or system processes.

A container image contains a packaged version of your application, with all the dependencies that are necessary for the application to run. Images can exist without containers, but containers are dependent on images because containers use container images to build a runtime environment to execute applications.

With the image stored in your local system, you can use the podman run command to create a new container that uses the image. The RHEL image from previous examples accepts Bash commands as an argument. These commands are provided as an argument and are executed within a RHEL container.

[user@host ~]$ podman run registry.redhat.io/rhel9/rhel-guest-image:9.4 \
echo 'Red Hat'
Red Hat

In the previous example, the echo 'Red Hat' command is provided as an argument to the podman run command. Podman executes the echo command inside the RHEL container and displays the output of the command.

When the container finishes executing the echo command, the container stops because no other process keeps it running. You can list the running containers by using the podman ps command.

[user@host ~]$ podman ps
CONTAINER ID  IMAGE       COMMAND     CREATED     STATUS      PORTS       NAMES

By default, the podman ps command lists the following details for your containers.

However, stopping a container is not the same as removing a container. Although the container is stopped, Podman does not remove it. You can list all containers (running and stopped) by adding the --all flag to the podman ps command.

[user@host ~]$ podman ps --all
CONTAINER ID  IMAGE                              COMMAND       CREATED       STATUS                   PORTS       NAMES
20236410bcef  registry.redhat.io/rhel9/rhel-guest-image:9.4  echo Red Hat  1 second ago  Exited (0) 1 second ago              hungry_mclaren

You can also automatically remove a container when it exits by adding the --rm option to the podman run command.

[user@host ~]$ podman run --rm registry.redhat.io/rhel9/rhel-guest-image:9.4\
echo 'Red Hat'
Red Hat
[user@host ~]$ podman ps --all
CONTAINER ID  IMAGE                              COMMAND       CREATED       STATUS                   PORTS       NAMES

If you do not provide a name for the container during its creation, then Podman generates a random string name for the container. It is important to define a unique name to facilitate the identification of your containers when managing their lifecycle.

You can assign a name to your containers by adding the --name flag to the podman run command.

[user@host ~]$ podman run --name podman_rhel9 \
registry.redhat.io/rhel9/rhel-guest-image:9.4 echo 'Red Hat'
Red Hat

Podman can identify the containers either by the Universal Unique Identifier (UUID) short identifier, which is composed of twelve alphanumeric characters, or by the UUID long identifier, which is composed of 64 alphanumeric characters, as shown in the example.

[user@host ~]$ podman ps --all
CONTAINER ID  IMAGE                                             COMMAND       CREATED       STATUS                   PORTS       NAMES
20236410bcef   registry.redhat.io/rhel9/rhel-guest-image:9.4  echo Red Hat  1 second ago  Exited (0) 1 second ago              podman_rhel9

In this example, 20236410bcef is the container's ID, or UUID short identifier. Additionally, podman_rhel9 is listed as the container's name.

If you want to retrieve the UUID long container ID, then you can add the --format=json flag to the podman ps --all command.

[user@host ~]$ podman ps --all --format=json
{
    "AutoRemove": false,
    "Command": [
      "echo",
      "Red Hat"
    ],
...output omitted...
    "Id": "2024...b0b2", 
    "Image": "registry.redhat.io/rhel9/rhel-guest-image:9.4",
...output omitted...

You can retrieve information about a container in either the JSON format or as a Go template.

Many applications, such as web servers or databases, run indefinitely to wait for connections. Therefore, the containers for these applications must run indefinitely. At the same time, it is usually necessary for these applications to be accessed externally through a network protocol.

You can use the -p option to map a port in your local machine to a port inside the container. This way, the traffic in your local port is forwarded to the port inside the container, which enables you to access the application from your computer.

The following example creates a new container that runs an Apache HTTP server by mapping the 8080 port in your local machine to the 8080 port inside the container.

[user@host ~]$ podman run -p 8080:8080 \
 registry.access.redhat.com/ubi9/httpd-24:latest
...output omitted...
[Thu Jun 18 12:58:57.048491 2024] [ssl:warn] [pid 1:tid 140259.33613248] AH01909: 10.0.2.100:8443:0 server certificate does NOT include an ID which matches the server name
[Thu Jun 18 12:58:57.048899 2024] [:notice] [pid 1:tid 140259.33613248] ModSecurity for Apache/2.9.2 (http://www.modsecurity.org/) configured.
...output omitted...
[Thu Jun 18 12:58:57.136272 2024] [mpm_event:notice] [pid 1:tid 140259.33613248] AH00489: Apache/2.4.37 (Red Hat Enterprise Linux) OpenSSL/1.1.1k configured -- resuming normal operations
[Thu Jun 18 12:58:57.136332 2024] [core:notice] [pid 1:tid 140259.33613248] AH00094: Command line: 'httpd -D FOREGROUND'

You can access the HTTP server at localhost:8080.

If you want the container to run in the background, to avoid the terminal being blocked, then you can use the -d option.

[user@host ~]$ podman run -d -p 8080:8080 \
 registry.access.redhat.com/ubi9/httpd-24:latest
b7eb467781106e4f416ba79cede91152239bfc74f6a570c6d70baa4c64fa636a

Environment variables are variables used in your applications that are set outside of the program. The operating system or the environment where the application runs provides the value of the variable. You can access the environment variable in your application at runtime.

Environment variables are a useful and safe way of injecting environment-specific configuration values into your application. For example, your application might use a database hostname that is different for each application environment, such as the database.local, database.stage, or database.test hostnames.

You can pass environment variables to a container by using the -e option. In the following example, an environment variable called NAME with the value Red Hat is passed. Then, the environment variable is printed by using the printenv command inside the container.

[user@host ~]$ podman run -e NAME='Red Hat' \
registry.redhat.io/rhel9/rhel-guest-image:9.4 printenv NAME
Red Hat