December 2017: CENTERWIDE Downtime 12/19 (noon) - 12/20
D
Dori Sajdak
started a topic
about 7 years ago
Status as of 11:45am: Downtime complete. All clusters and faculty servers are back in production. Please email ccr-help@buffalo.edu if you encounter any problems.
Status as of 9am 12/20: Power outage and facilities work overnight was successful. We are working as quickly as possible to bring clusters and servers back online. We expect the majority of services to be restored by noon. Thank you for your patience.
Date of downtime: Tuesday, December 19, 2017 at noon through Wednesday, December 20, 2017
NOTE: this was moved from the first Tuesday of the month to accommodate the academic calendar
Resources affected by downtime: EVERYTHING AT CCR except the research cloud (LakeEffect)
UB-HPC cluster (general-compute, debug, viz, largemem, and gpu partitions)
Industry cluster (compute, scavenger partitions)
PI clusters: MAE, Physics, Chemistry (all partitions)
Portals: WebMO, Remote Visualization
All additional servers in the machine room
What will be done: UB Facilities requires a full power outage to accomplish upgrades to the electrical system in the CCR machine room. We will perform standard downtime activities such as a reboot of all nodes and operating system security updates.
Dori Sajdak
Status as of 11:45am:
Downtime complete. All clusters and faculty servers are back in production. Please email ccr-help@buffalo.edu if you encounter any problems.
Status as of 9am 12/20:
Power outage and facilities work overnight was successful. We are working as quickly as possible to bring clusters and servers back online. We expect the majority of services to be restored by noon. Thank you for your patience.
Date of downtime: Tuesday, December 19, 2017 at noon through Wednesday, December 20, 2017
NOTE: this was moved from the first Tuesday of the month to accommodate the academic calendar
Resources affected by downtime: EVERYTHING AT CCR except the research cloud (LakeEffect)
UB-HPC cluster (general-compute, debug, viz, largemem, and gpu partitions)
Industry cluster (compute, scavenger partitions)
PI clusters: MAE, Physics, Chemistry (all partitions)
Portals: WebMO, Remote Visualization
All additional servers in the machine room
What will be done: UB Facilities requires a full power outage to accomplish upgrades to the electrical system in the CCR machine room. We will perform standard downtime activities such as a reboot of all nodes and operating system security updates.
Please see the downtime schedule for 2017-18 for more information
Jobs will be held in the queue while maintenance is done. If you have any questions or concerns please e-mail ccr-help_at_buffalo.edu