Install XFC client

XFC client install  

In order to initiate and manage your usage of the XFC (transfer cache) service, you need to use the XFC client.

Once set up, you don’t need to use this client to read/write data to XFC storage itself, but you can use it to interrogate the system to find out what data you have stored and how much of your quota(s) you are using.

The following steps should be used to create a python virtual environment and pip to install the xfc client:

NOTE: do these steps on one of the sci (not xfer) servers:

1. Log into one of the sci machines: e.g. sci-vm-01.jasmin.ac.uk (1 - 8 is available)
2. The xfc client can be used with Python 3 using jaspy.

module load jaspy

Setup a virtual environment in your home directory:

python -m venv ~/xfc_venv

3. The following steps for Python 2 and Python 3 are now the same.
4. Activate the virtual environment:

source ~/xfc_venv/bin/activate

5. Download the client software using git to your home directory:

git clone https://github.com/cedadev/xfc_client.git

6. pip install the xfc client

pip install -e ~/xfc_client

7. Use xfc on the command line

xfc -h

Users who already have a Python virtual environment can skip step 2 and install into an existing virtual environment.

The recommendation for users using NLA, XFC and JDMA is to create a single virtual environment to install all 3 client applications into.

The xfc client can be used without activating the python virtualenv by adding the path to the xfc client to the $PATH$ environment variable:

export PATH="$PATH:~/xfc_venv/bin" `
echo 'export PATH="$PATH:~/xfc_venv/bin"' >> "$HOME/.bashrc"

The xfc client can now be used by invoking xfc from the command line without activating the virtualenv. Python 3 users load jaspy using the command module load jaspy

For instructions on how to use the xfc client, see the article JASMIN Transfer Cache (XFC).