Pegasus doesn't capture the execution site information until the user job actually runs. This means that if the job fails in the pegasus lite script (e.g. because stage in file transfers fail) it's impossible to figure out from dashboard where the job is failing. It would be helpful if this information could appear in dashboard (Or at least printed out in at the top of the pegasus lite stderr so it's easy to figure out the failing site. This is done if the job is a glide in, but not necessarily otherwise.)