Storage issue on login nodes

We are currently having an issue with our storage system (GPFS) on the login nodes.  We are working on it, and hope to have it resolved quickly.
As far as we know, the issue doesn't affect compute nodes, or already submitted, currently running jobs.