Important note concerning Wilson Facility support
A user running opportunistically under VO-wide projects is expected to first seek computing support from their VO’s computing coordinator before creating an incident with Wilson facility staff. Computing coordinators are most experienced with building and running application codes specific to each VO. A user running under a project account will be given higher priority in support than users that are running opportunistically under VO-wide accounts.
Creating incidents or requests in Service Now
Incidents reports and requests should be made via Service Now. Incident tickets are used to report something that is broken. There is usually some sort of error message involved in an INCident ticket. Everything else is classified as a REQuest ticket. If you submit using the wrong type, our support staff will cancel the first ticket and create one of the other type of ticket.
Create an incident or request by selecting the affected service. Services related to the Wilson facility may be found either by browsing for Wilson related items in the Scientific Computing Services catalog. The following table has links to the defined service areas. Note that you need a Fermilab services account to login to Service Now.
| Service area link | Description |
| Wilson Facility Application Support | Assists users with the installation of a collaboration or experiment specific application on the Wilson cluster. Diagnoses application failures when running jobs on the Wilson cluster. |
| Wilson Facility File System | Diagnosing performance and access related issues with the Parallel & Distributed file-systems (Lustre) or the Networked File Systems (NFS). First order diagnosis of failures related to file transfers especially to remote sites using Globus Online. |
| Wilson Facility General User Support | Creating new user accounts on the various facility clusters for a user with a valid Kerberos principal. Also assisting users with any remote access issues especially with Kerberos and Secure SHell. Creating project specific groups and modifying directory and file permissions as requested. |
| Wilson Facility Parallel and Tightly Coupled Batch Computing | Parallel and tightly coupled batch computing consists primarily of the batch queuing system for the various facility clusters. Each cluster in the Wilson facility has a dedicated login and server head node. The batch queuing system consists of the batch server (Torque) and scheduler (Maui) that run on the dedicated server head node. Users access the cluster resources by submitting jobs to the batch queuing system by logging in to the corresponding cluster login head node. The cluster compute nodes are available as a pool of resources to the users through the batch compute server. |
Announcements Mailing List
There is an hpc-announce@listserv.fnal.gov mailing list where Wilson Admin staff will post information about outages, both current and scheduled, involving the Wilson Facility. Anyone is welcome to subscribe to the list using the Subscription Form.