You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 588d779
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: docs/marketplace-docs/guides/apache-airflow/index.md
+23-26Lines changed: 23 additions & 26 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -2,6 +2,7 @@
2
2
title: "Deploying Apache Airflow"
3
3
description: "Deploy Apache Airflow on a Linode Compute Instance. This provides you with an open-source workflow management platform for data engineering pipelines."
1. Log into your new Compute Instance through [LISH](https://techdocs.akamai.com/cloud-computing/docs/access-your-system-console-using-lish) or [SSH](https://techdocs.akamai.com/cloud-computing/docs/set-up-and-secure-a-compute-instance#connect-to-the-instance) using the root user and the password you entered when creating the instance.
51
+
Once the app is deployed, you need to obtain the credentials from the server:
51
52
52
-
1.If you log in on the same day you deployed the app, a message is displayed with a link to your Apache Airflow GUI and the username and password to use.
53
+
1.Log in to your new Compute Instance using one of the methods below:
53
54
54
-
```output
55
-
The installation of Apache Airflow is now complete, and the application is running in standalone mode.
56
-
#
57
-
You can log into the Airflow GUI at 192-0-2-170.ip.linodeusercontent.com
58
-
With the credentials:
59
-
Username: admin
60
-
Password: pNbgWNDhYcRNvgRq
61
-
#
62
-
Standalone mode is not recommended for production.
63
-
```
55
+
-**Lish Console**: Log in to Cloud Manager, click the **Linodes** link in the left menu, and select the Compute Instance you just deployed. Click **Launch LISH Console**. Log in as the `root` user. To learn more, see [Access your system console using Lish](https://techdocs.akamai.com/cloud-computing/docs/access-your-system-console-using-lish).
56
+
-**SSH**: Log in to your Compute Instance over SSH using the `root` user. To learn how, see [Connect to the Linode](https://techdocs.akamai.com/cloud-computing/docs/set-up-and-secure-a-compute-instance#connect-to-the-linode).
64
57
65
-
If you don't see this message, you can output it using the command below:
58
+
2. Run the following command to access the contents of the `.credentials` file:
66
59
67
60
```command
68
-
cat /etc/motd
61
+
cat /home/$USERNAME/.credentials
69
62
```
70
63
71
-
1. Open your web browser and navigate to the Airflow URL provided in the last step. This should be the custom domain you entered during deployment or your Compute Instance's rDNS domain (such as `192-0-2-1.ip.linodeusercontent.com`). See the [Managing IP Addresses](https://techdocs.akamai.com/cloud-computing/docs/managing-ip-addresses-on-a-compute-instance) guide for information on viewing the rDNS.
64
+
This returns the admin password and other details that were automatically generated when the instance was deployed. Save them securely. After your credentials are saved, you can safely delete the file.
65
+
66
+
## Getting Started After Deployment
67
+
68
+
Once you have the credentials, you can access your Apache Airflow instance. Open a browser, and navigate to your Linode domain entered during deployment or the rDNS domain `https://203-0-113-0.ip.linodeusercontent.com`.
72
69
73
-
1. Within the Airflow login prompt that appears, enter the credentials provided in a previous step and sign in.
70
+
Within the Airflow login prompt that appears, enter the credentials provided inthe `.credentials` file, and sign in.
74
71
75
-
72
+
76
73
77
-
1. Once you are signed in, the Airflow dashboard appears. From here, you can view the DAGs (Directed Acyclic Graphs) and access all other areas of the dashboard.
74
+
The Airflow dashboard appears once you're signed in. From here, you can view the DAGs (Directed Acyclic Graphs) and access all other areas of the dashboard.
78
75
79
-
76
+
80
77
81
-
1. You can now start using Apache Airflow. If you are unfamiliar with it, consider reading through the official documentation or Linode's own guides:
78
+
You can now start using Apache Airflow. If you are unfamiliar with it, consider reading through the official documentation or our provided guide:
- [Create Connections and Variables in Apache Airflow](/cloud/guides/apache-airflow-tutorial-creating-connections-and-variables/)
86
83
87
84
{{< note type="warning">}}
88
-
The Akamai Quick Deploy App app deploys Apache Airflow in standalone mode, suitable for development, testing, and initial configurations. Standalone mode is not recommended for [production deployments](https://airflow.apache.org/docs/apache-airflow/stable/production-deployment.html).
85
+
This Akamai Quick Deploy App deploys Apache Airflow in standalone mode, suitable for development, testing, and initial configurations. Standalone mode is not recommended for [production deployments](https://airflow.apache.org/docs/apache-airflow/stable/production-deployment.html).
CrewAI is an open source framework for building and orchestrating AI agents that work together as a team to accomplish complex tasks. Instead of relying on a single large language model prompt, CrewAI allows developers to create specialized agents with distinct roles, responsibilities, and goals. These agents can collaborate, delegate work, share information, and execute multi-step workflows autonomously.
Once the deployment is complete, the `crewai` binary is installed on the instance. This allows you to create the first group of agents.
47
+
48
+
1. To get started, create an example crew called `research`.
49
+
50
+
```command
51
+
crewai create crew research
52
+
```
53
+
54
+
This command creates a project scaffold with everything that you need to run your agents. To understand more about the project structure, see the [Crew AI documentation](https://docs.crewai.com/en/installation#creating-a-crewai-project).
55
+
56
+
2. Install the `research` crew. This installs the default crew of agents that comes with CrewAI.
57
+
58
+
```command
59
+
cd research
60
+
crewai install --prerelease=allow
61
+
```
62
+
63
+
3. To run the agents, use the following command.
64
+
65
+
```command
66
+
crewai run
67
+
```
68
+
69
+
Once the command is executed, all of the agents and tasks start running. On execution, you can see the agent's reasoning and how they perform tasks assigned to them.
70
+
71
+
The default crew for CrewAI allows a group of agents to research AI trends for the year. A Markdown file is created, highlighting the research in a bulleted list.
[Hermes Agent](https://hermes-agent.nousresearch.com) is an open-source AI agent platform designed for autonomous task execution through a persistent gateway service. The gateway connects communication channels, tools, and AI models, allowing the agent to receive instructions, perform actions, and automate multi-step workflows. Administrators configure and manage the system through a CLI setup wizard and a local web dashboard. Our Quick Deploy App allows you to connect to the Hermes Agent dashboard via a secure HTTPS endpoint protected by HTPASSWD.
20
+
21
+
This Quick Deploy App creates a Hermes Agent user on the system called `hermes`.
Once the deployment is complete, `hermes` is installed on the instance but is not running yet. Before you can start using Hermes Agent, you need to go through the setup wizard. This Quick Deploy App triggers the onboarding for you when you log in as root.
53
+
54
+
1. Log in to the instance:
55
+
56
+
If you disabled root login to the server during the setup of the Hermes Agent app, log in to the server as a sudo user.
57
+
58
+
```command
59
+
ssh admin@YOUR_INSTANCE_IP
60
+
```
61
+
62
+
Replace `YOUR_INSTANCE_IP` with the IP address of your Linode instance and `admin` with the sudo user you created.
63
+
64
+
2. Escalate privileges to root:
65
+
66
+
Once you've logged in, you can see the [motd](https://man7.org/linux/man-pages/man5/motd.5.html):
Copy the sudo password from the `~/.credentials.txt` file and enter the following command in the terminal:
77
+
78
+
```command
79
+
sudo su -
80
+
```
81
+
82
+
When prompted for the password, paste the sudo password you copied from the `~/.credentials.txt` file. When you log in as **root**, the message about the setup wizard appears. If you are ready to perform the onboarding, enter `y` and it takes you to Hermes Agent's & Hermes Gateway setup wizards where you can complete the configuration.
83
+
84
+
Once onboarding is complete, the setup script is removed.
85
+
86
+
### Confirm Hermes and Gateway Statuses
87
+
88
+
At this point, you've configured Hermes Agent on the server. To verify the Hermes and Hermes gateway are running, you need to become the `hermes` user. Enter the following in the terminal as the **root** user:
89
+
90
+
```command
91
+
su - hermes
92
+
```
93
+
94
+
To view the hermes and gateway statuses, enter the following as the **hermes** user:
[Langflow](https://www.langflow.org/) is an open source, Python-based framework for building AI applications. It supports key AI functionality, such as agents and the Model Context Protocol (MCP), and it is not tied to any specific large language model (LLM) or vector store. Its visual editor simplifies prototyping application workflows, letting developers quickly turn ideas into working solutions.
**Estimated deployment time:** Langflow should be fully installed within 5-10 minutes after the Compute Instance has finished provisioning.
30
+
{{< /note >}}
31
+
32
+
## Configuration Options
33
+
34
+
-**Supported distributions:** Ubuntu 24.04 LTS
35
+
-**Recommended plan:** We recommend a plan with at least 4 GB of memory, such as the 4 GB Shared CPU plan. Langflow runs both the application and its PostgreSQL database as containers on the instance, so higher-tier plans improve responsiveness when building and running larger flows.
36
+
37
+
### Langflow Options
38
+
39
+
-**Email address***(required)*: Enter the email address to use for generating the SSL certificates.
1. Open your web browser and navigate to `https://[domain]`, where *[domain]* is the custom domain you entered during deployment or your Compute Instance's rDNS domain, for example, `192-0-2-1.ip.linodeusercontent.com`. To learn more about viewing IP addresses and rDNS, see the [Manage IP addresses](https://techdocs.akamai.com/cloud-computing/docs/managing-ip-addresses-on-a-compute-instance) guide.
52
+
53
+
54
+
55
+
2. Use the following credentials to log in:
56
+
-**Username:***admin*
57
+
-**Password:** Enter the Langflow admin password stored in the credentials file on your server. To obtain it, log in to your Compute Instance via [SSH](https://techdocs.akamai.com/cloud-computing/docs/set-up-and-secure-a-compute-instance#connect-to-the-linode) or [Lish](https://techdocs.akamai.com/cloud-computing/docs/access-your-system-console-using-lish) and run:
58
+
59
+
```command
60
+
cat /home/$USER/.credentials
61
+
```
62
+
63
+
The credentials file also contains your limited sudo user's password and the PostgreSQL database credentials.
64
+
65
+
3. Once you are logged in to Langflow, go to the canvas.
66
+
4. Drag components from the left panel onto the board and connect them to build a flow. Running a flow that calls an LLM or embeddings provider requires you to add your own provider API key (for example, OpenAI) in the relevant component or under **Settings > Global Variables**. To learn more, see:
0 commit comments