I am trying to run some test jobs on Europa, but I am running into a lot of problems.
Logs aren’t being written for jobs I’m submitting and it also seems like the network is experiencing major problems as well. I’m unable to do any testing on Europa whatsoever since I can’t even ssh into nodes. The screen just gets stuck trying to connect.
This is making my tasks impossible to debug. Is someone working on this? I have a big deadline just before Christmas and I have thousands of jobs that need to be run.
You are correct that Europa is current suffering under some extreme loads. We have several people “fighting” trying to finish up projects before the end of the semester. As we’ve discussed, Europa doesn’t have a parallel file system yet. The good news is that we will have one coming in the Spring but that doesn’t address your immediate needs.
Can we set up a meeting for Monday and discuss your workloads and workflows? I might be able to get you access to other systems that can help you finish up in time. Send me an email directly to csim@utdallas.edu or ping me in Teams and we can set up a meeting.