We will be upgrading the HPC/IC2 GPFS servers (hpcgpfs01) beginning @ 9:30AM. This should be a transparent upgrade to the RHEL and GPFS software. If there are any issues, please open an RT ticket for these to be addressed.

A change in the configuration of the storage network in the SCDF legacy data center (Bldg 515/CDCE) will result in a brief disruption in connectivity to storage systems located in the legacy data center from other systems at the SCDF. This configuration change is scheduled for 9:30AM EST on Tuesday February 17.

The SCDF firewall, separating the SCDF facility from the general internet and the BNL campus network, will be upgraded in a two steps.

Step one - Feb 17 at 6:00AM EST - This step is expected to be transparent as existing network connections through the firewall are expected to remain intact.

Step two - Feb 18 at 6:00AM EST - This step will break existing network connections through the firewall. Users will need to re-establish network connections if they break.

After discussing with Atlas Experiment it's no longer using the AFS file servers. We are planning to retire/shutdown all Alas AFS service on 2/2/2026. If there are any files still being accessed, please move them to a new location.

Systems have been operating stably since the cutover to site chilled water.
SCDF is closely monitoring and confirming functionality of major systems one by one.

We will continue to run in the current state, with reduced capacities for some systems,
until we we have more info from F&O and are recommended to return all systems to normal operations.

We believe all major systems are otherwise functioning; please report any issues.

A major leak in one of the cooling tower loops triggered a temperature rise in a number of compute racks (main impact appears limited to low density compute rows 103 and 105, impacting sPHENIX, and to a smaller extent the shared pool and ATLAS Tier1), before the backup BNL site chilled water feed kicked in.

It's currently believed all systems should be functioning, albeit sPHENIX, spool and ATLAS T1 are currently running with reduced capacities. Please report any issues.

Comanage Registry is undergoing an upgrade and will be unavailable for Federated Logins for 15 min at 6 am Central Time on Friday January 16th. This will affect a variety of services but will be transparent if you are already logged into those services.

Globus Access (https://app.globus.org/) will be down for maintenance from 11AM-3PM today 01/06/26.

The BNLBox update has been completed, if there are any issues please submit tickets to the RT Storage Management queue at RT-RACF-StorageManagement@bnl.gov

BNLBox will be unavailable starting at 11:00 AM EST on December 30, 2025, due to scheduled maintenance. During this time, the service will be inaccessible. Users should complete any work before the maintenance window begins. Normal operations are expected to resume once maintenance is complete, on or before 5:00 PM EST on December 30, 2025. Users may resume usage at that time.

Thank you for your patience during this scheduled downtime.