# Data Portal

Researchers utilizing Monsoon's compute power now have the ability to share project data with colleagues via a web browser. Any individual files or directories in your `/projects/<project_name>/` folder can be shared. No file is too large, or too small.

*A long-term storage space is required for this service to work. To obtain a long-term storage space, please request one on the [long-term storage request](<https://in.nau.edu/hpc/request-storage/>).*

## Publishing data

To share data using the data portal, perform the following steps while [SSH'd into Monsoon](<https://rcdata.nau.edu/hpcpub/docs-test/connecting-to-monsoon/index.html>):

1. Change directory to your project directory. The publish script must be ran there. For example, if your project is called `hpc_project`, you'd run:

`cd /projects/hpc_project`

1. Run the `publish_data` script. When provided with no files to publish, it will print some usage information. If one or more files and/or directories are specified, those files and/or directories will be published to the web.For example, if you wanted to publish the following files and directories:

```text
- /projects/hpc_project/january_results
- /projects/hpc_project/computational_data/dataset1.csv
- /projects/hpc_project/computational_data/dataset2.txt
```

You would run:

```bash
publish_data january_results computational_data/dataset1.csv computational_data/dataset2.txt
```

1. Your data can now be accessed at `https://rcdata.nau.edu/<project_name>`. Using the previous examples, the URL would be `https://rcdata.nau.edu/hpc_project`.

## Unpublishing data

If you wish to unpublish files, the `unpublish_data` script works similarly:

1. Change directory to your project directory.  
   For example, if your project is called `hpc_project`, you'd run:  
   `cd /projects/hpc_project`
2. Run the `unpublish_data` script. For example, to unpublish the following files:

```text
- dataset1.csv 
- january_results/bad_results.out
```

you would run:

```bash
unpublish_data january_results/bad_results.out computational_data/dataset1.csv
```

if you wanted to unpublish all of the `january_results` directory, you'd run:

```bash
unpublish_data january_results
```

Once done, the data will still be in your project directory, but will no longer appear on the web.

## Optionally securing your data

To secure your data so only users with authorized credentials can access your published data, use the `secure_published_data` script to secure your published site.

```bash
secure_published_data /projects/hpc_project
```

The script will create a file under `/projects/hpc_project/public_data/` called `.htaccess` and will automatically fill it with the following contents:

```text
AuthType Basic
AuthName "Restricted Area"
AuthUserFile "/projects/hpc_project/.htpasswd"
require valid-user
```

The `secure_published_data` script will also ask you if you would like to add your first user. To add your first user, navigate to your projects folder and run the following command:

`htpasswd -cB .htpasswd <username>`

To add subsequent users, navigate to your projects folder and run the following command:

`htpasswd -B .htpasswd <username>`

To change the password of a user, you can re-run the command above for the target user.

## FAQs

**What happens when I delete files in my project area that have been published?**

The files remain published.

**What happens when I manually delete files from the `public_data` area?**

The files are unpublished effectively. The files that were published still remain in your project area.