August 2020: Maintenance Downtime (MOVED to 9/1/20)
D
Dori Sajdak
started a topic
over 3 years ago
Date of downtime: Tuesday, September 1, 2020 NOTE: The July & August downtimes have been moved back one week
Approximate time of outage: 7am-5pm
Resources affected by downtime:
UB-HPC cluster (general-compute, debug, viz, largemem, and gpu partitions)
Industry cluster (compute, scavenger partitions)
Faculty cluster (all partitions)
Portals: WebMO, OnDemand
What will be done:
Operating system updates on front end servers and OnDemand
Reboots of all cluster nodes and front end servers
New scratch storage, Panasas, moved to production - More details
Slurm environment variable $GLOBAL_SCRATCH changing to /panasas/scratch. Users will need to to append their new group scratch directory name. See link above.
New Slurm server moved to production
Enable pam_slurm_adopt module to restrict users' access to only nodes their jobs are running on - Technical Details
OnDemand software update to version 1.8 - Release Notes One highlight: Copy and pasting now works out of the box in Chrome. An issue with copy and pasting large clipboard buffers into noVNC applications on Chrome was fixed in OnDemand version 1.8. Interactive Apps menu is getting reorganized. StarCCM app now called "OpenGL DISABLED" and desktops on visualization nodes are under separate category.
Dori Sajdak
Date of downtime: Tuesday, September 1, 2020 NOTE: The July & August downtimes have been moved back one week
Approximate time of outage: 7am-5pm
Resources affected by downtime:
UB-HPC cluster (general-compute, debug, viz, largemem, and gpu partitions)
Industry cluster (compute, scavenger partitions)
Faculty cluster (all partitions)
Portals: WebMO, OnDemand
What will be done:
One highlight: Copy and pasting now works out of the box in Chrome. An issue with copy and pasting large clipboard buffers into noVNC applications on Chrome was fixed in OnDemand version 1.8.
Interactive Apps menu is getting reorganized. StarCCM app now called "OpenGL DISABLED" and desktops on visualization nodes are under separate category.
Jobs will be held in queue during the maintenance downtime and will run after the updates are complete.
If you have any questions or concerns please e-mail ccr-help_at_buffalo.edu