In here I will show you how to find out what your home directory looks like on the slave nodes on the cluster. In particular, I will show you a script that you can copy to your home directory on the master node and then use to find out which files are present on your home directory on each of the slave nodes.
The script checks your home directory on each of the available nodes and lists the files contained there. As a side benefit, you will also see which nodes are unavailable.
You can decide to restrict the search to just a subset of the nodes (which you can specify). For example, you may decide that you only want to work with nodes in a particular range (in an attempt to stay away from other users).
Simply enter the following command from your home directory on
the master1 node:
rls-nodes
Here is an (edited) example of the output:
[avila@master avila]$ rls-nodes node3: integrate node4: integrate node5: integrate node6: integrate node7: integrate node8: *UNAVAILABLE* node9: integrate node11: integrate ... node52: integrate node53: *UNAVAILABLE* node54: integrate node55: integrate node56: integrate node57: integrate node58: integrate
You should see those files that you have currently in your home directory on each of the nodes of the cluster. Unless you are running a parallel program on the cluster, there should be nothing in your home directory.
Remove files with the command (for example suppose you want to
remove a file called integrate):
rcom-nodes rm /home/cs159xx/integrate
where cs159xx is your particular login name.