Basic usage of Allas

Accessing Allas via the Puhti web interface (optional)

Go to the Puhti web interface: https://www.puhti.csc.fi.
Login with your CSC user account (or Haka).
Select Cloud storage configuration from the Pinned Apps view on the dashboard.
Enter your CSC password to authenticate, then create an Allas S3 connection for the project you want to use for this exercise.
- You may skip the previous step if you’ve already configured a connection.
💡 Now you’ll be able to browse your buckets and objects in Allas using the Files app!
From the Files dropdown menu in the top navigation bar, select s3allas-project_<id> where <id> is the number of your project (e.g. 2001234).
Create a new bucket by pressing the New Directory button.
- Name it as <id>_<username>, in which <id> is again the number of your project and <username> is your CSC username. Note that you cannot use a bucket name that already exists!
Open the created bucket by clicking it.
Upload one file from your computer into the bucket (any file should do, but prefer a file that you can open in Puhti, e.g. a text file).

💭 During the exercises, you can use this web interface to get another view to your buckets and objects in Allas.

Accessing Allas from Puhti command-line

Preparations (if not done already)

Login to puhti.csc.fi (open a login node shell if using the web interface)
In Puhti, check your environment with the command
```
csc-workspaces
```

Move to the scratch directory of your project:

cd /scratch/<project>   # replace <project> with your CSC project, e.g. project_2001234

Create your own subdirectory named as your username (tip! your username is automatically stored in the environment variable $USER):
```
mkdir -p $USER
cd $USER
```

Connecting to Allas

Open a connection to Allas with these commands:
```
module load allas
allas-conf 
```
💡 It might take a while to run module load allas
If you have several projects with access to Allas available, select the one where you just created a bucket using the Puhti web interface

Study what you have in Allas with a-commands and with rclone:

With a-commands:

a-list
a-list <id>_$USER             # replace <id> with your CSC project number, e.g. 2001234
a-info <id>_$USER/<filename>  # replace <id> with your CSC project number, e.g. 2001234, and <filename> with the file you uploaded

With rclone:

rclone lsd allas:
rclone ls allas:<id>_$USER
rclone lsl allas:<id>_$USER
rclone lsf allas:<id>_$USER
rclone cat allas:<id>_$USER/<filename>

Download to Puhti the file that you just uploaded from your local computer to Allas. This can be done in two ways:

With a-commands:

a-get <id>_$USER/<filename>     # replace <id> with your CSC project number, e.g. 2001234, and <filename> with the file you uploaded

With rclone:

rclone copy allas:<id>_$USER/<filename> ./    # replace <id> with your CSC project number, e.g. 2001234, and <filename> with the file you uploaded

Open, edit and rename the file so that you can distinguish it from the original one
Upload the new file to Allas:
- With a-commands:
```
a-put -b <id>_$USER <newfilename>   # replace <id> and <newfilename> accordingly
```
  💭 Try running a-put -h to understand the command-line switch above and to find more information on options.
  
  💬 With larger files it is good to include the option -c to enable zstdmt compression of the files.
- With rclone:
```
rclone copy <newfilename> allas:<id>_$USER/   # replace <newfilename> and <id> accordingly
```

Check that the file in Puhti indeed has a counterpart in Allas:

a-check -b <id>_$USER <newfilename>   # replace <id> and <newfilename>

Locate the files you just uploaded to Allas in the Puhti web interface (search for the bucket name)

Clean up

Delete the local file from Puhti:

rm <filename>             # replace <filename>

Whenever you need your data again, you can download it from Allas

💭 If you can’t find your file but remember the name, try a-find. Use a-find -h for help.

Extra: publish a file to the internet

💬 The a-commands include basic tools for publishing files to the internet. You might notice that the course slides use one of these! 🤓

‼️ Note: Using these commands makes your entire bucket public! Do not engage if you don’t want that to happen. All files that you a-put in the bucket later will also be accessible from internet, since the bucket is accessible.

Option 1: `a-publish`

Select a file that has an appropriate content and publish it with the command:
```
a-publish -b <id>_$USER <filename>   # replace <id> and <filename>
```
The command outputs a URL (public link). Copy it to your browser or send it to your friends 😎

Option 2: `a-flip`

💬 a-flip is meant for files that need to be published only temporarily, for example for a one-time share.

Select a file that has an appropriate content and publish it with the command:
```
a-flip <filename>         # replace <filename>
```
The command outputs an URL (public link). Copy it to your browser or send it to your friends 😎

‼️ Note: a-flip takes just the file name, not the bucket name like many of the previous commands.