AiiDA CLI Cheatsheet

Here we will list some useful commands for working with AiiDA. Have a look here to explore all available AiiDA command line subcommands.

Monitoring Submitted Processes

List all submitted processes:

verdi process list -a

Delete a process node, identified by its primary key value <PK>:

verdi node delete <PK>

Delete multiple process nodes between primary key values <PK_1> and <PK_n>:

verdi node delete {<PK_1>..<PK_n>}

View all the nodes associated with a process node:

verdi node show <PK>

View attributes of a process node (such as retrieved files and find the path on disk where outputs are stored temporarily):

verdi node attributes <PK>

From the node attributes output dictionary, you can find where the input and output files are temporarily stored for a process in the “remote_workdir” value.

Visualise Data Provenance

Visualise your submitted jobs as a provenance graph outputted in a .pdf file. Select the latest <PK> to include all previous nodes in the graph:

verdi node graph generate <PK>

An example provenance graph for the first eight steps of the lysozyme tutorial, will look something like this:

../_images/53.dot.png

Debugging

Any aiida errors are logged in .aiida/daemon/log/.

If any changes to the plugin code are made, after an update for example, restart the daemon if it is already running to implement the code changes:

verdi daemon restart --reset

Sharing Data

When you are ready to share data, the AiiDA database and accompanying files inputted and outputted in each process can be archived into a single file:

verdi archive create --all archive_name.aiida

where the --all flag saves all the data in the AiiDA profile. To import an existing AiiDA archive file to a loaded profile:

verdi archive import archive_name.aiida