Maintenance Downtime Logs for 2026

The full 2026 maintenace schedule is available in CCR's Downtime documentation

September 2026 Downtime

Date of downtime: Tuesday, September 29, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Update and reboot of all cluster nodes

  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)

  • Infrastructure Updates

Specific Effects to CCR users:
None anticipated

August 2026 Downtime

Date of downtime: Tuesday, August 18, 2026
NOTE DATE CHANGED to accommodate the start of the fall semester

Approximate time of outage: 6am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Update and reboot of all cluster nodes

  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)

  • Infrastructure Updates

Specific Effects to CCR users:
None anticipated

July 2026 Downtime

Date of downtime: Tuesday, July 28, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Update and reboot of all cluster nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Infrastructure Updates
  • Storage mounts switching to autofs

Specific Effects to CCR users:

What changed? The /projects and /vscratch filesystems now use automounting on most CCR systems. Rather than making every project and scratch directory available all the time, directories are made available only when they are accessed.

What does this mean for you?
Directory listings such as ls /projects or ls /vscratch will no longer show all subdirectories.

Instead, you should navigate directly to your group's directory or reference it by its full path in web interfaces such as Globus. For example: 

   cd /projects/academic/[YourGroup]  
   cd /projects/rpci/[YourGroup]  
   cd /projects/industry/[YourGroup]  
   cd /vscratch/grp-[YourGroup]
Once you're in your project or scratch directory, commands such as ls, find, and other file operations will work normally.

Where can I find my group’s directory paths?
In ColdFront, each storage allocation has the full path listed both in the “description” and under the “Allocation Attributes” section on the Allocation Detail page.  To access this information, navigate to your project and click on the folder icon next to the storage allocation you’re interested in.  You can also quickly access this on the home page under the “Allocations” section by clicking on the green “Active” button next to the allocation name.  Note that for those with access to many allocations, not all will show up on the home page.

What systems does this affect?
Login nodes, compile nodes, compute nodes, Globus.  Coming soon: OnDemand

Why was this done?
This change reduces unnecessary filesystem activity and helps improve overall system performance.

June 2026 Downtime

Date of downtime: Tuesday, June 30, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Update and reboot of all cluster nodes

  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)

  • Infrastructure Updates

Specific Effects to CCR users:
None anticipated

May 2026 Downtime

Date of downtime: Tuesday, May 26, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Update and reboot of all cluster nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Update Slurm version
  • Infrastructure Updates
  • Switch from directory quotas to user & group quotas on home directories
  • Deployment of Jobstats
  • Replace outdated RStudio Server app with "RStudio Server - Containers" app. This app allows users to select how they launch RStudio. Options include the CCR provided R 4.2 software module, a set of RStudio containers provided by CCR, or your own container with RStudio installed.
  • Slurm dashboard decommissioned. For security reasons, this tool will not be provided by CCR at this time

Specific Effects to CCR users:
Though issues are not anticipated, the switch from directory to user quotas on the /user filesystem may result in quota errors. Please report any problems to CCR Help

April 2026 Downtime

Date of downtime: Tuesday, April 28, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Reboot of all cluster nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Infrastructure Updates

Specific Effects to CCR users:
ColdFront was updated to a new version and there are minor display differences that users may notice. If any of the functionality doesn't work as expected, please submit a report to CCR Help.

March 2026 Downtime

Date of downtime: Tuesday, March 31, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Reboot of all cluster nodes; new operating system image installed on nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Infrastructure Updates

Specific Effects to CCR users:

  • Fix for OnDemand to resolve issue with cluster (shell) app and Firefox version 148+

February 2026 Downtime

Date of downtime: Tuesday, February 24, 2026

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Reboot of all cluster nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Infrastructure Updates

Specific Effects to CCR users:
None

January 2026 Downtime

Date of downtime: Tuesday, January 20, 2026
NOTE: This is a change from the normal last Tuesday of the month schedule to accomodate the start of the spring semester.

Approximate time of outage: 7am-5pm

Resources affected by downtime:

  • UB-HPC cluster (all partitions)
  • Faculty cluster (all partitions)
  • Portals: OnDemand, ColdFront (temporarily), IDM (temporarily)

What will be done:

  • Reboot of all cluster nodes
  • Updates of front-end login nodes (login1, login2, login-future - formerly vortex-future)
  • Infrastructure Updates

Specific Effects to CCR users:
None