Hi guys,

We have a bit of a situation. Our research group at school is kind of "in between" system administrators, so I am trying to fix this problem - I don't have much experience with clusters so I am asking for your help. Our cluster was rebooted over the weekend. After that, for some reason, slave nodes suddenly began switching their ip addresses. As a result of that, they don't have a proper access to the head node and any jobs that were submitted to the slave nodes (this are heavy calculations that last for couple of days) don't get saved. Can someone point me to a reason why the nodes are switching ip's and a way to fix this? It would be greatly appreciated.