FAS Research Computing - Hinweisverlauf

Our maintenance tasks should be completed between 8am-6pm.

NOTICES:

• New: Netscratch retention will be reduced to 60 days starting with the December maintenance. More information will be announced soon.

• Home directory permissions have been reset to allow only owner access per FASRC and Harvard policy. Data in home directories cannot be shared among users. Shared data should be stored in lab storage. Please reach out to rchelp@rc.fas.harvard.edu with questions or concerns.

• Python/conda/mamba/venv security issue: A security problem was found with some of the certificates that software uses to confirm it is talking to the real website. Browsers have already been fixed, but Python tools like conda and mamba keep their own copy of these certificates in two packages, ca-certificates and certifi. Please update both to version 2026.10.19 or newer. We know of no attacks against these tools, but updating is a simple precaution. For more details, see the eMudhra Security Update.

• Training: Upcoming training from FASRC and other sources can be found on our Training Calendar. at https://www.rc.fas.harvard.edu/upcoming-training/

• Status Page: You can subscribe to our status to receive notifications of maintenance, incidents, and their resolution at https://status.rc.fas.harvard.edu/ (click Get Updates for options).

MAINTENANCE TASKS

Cannon cluster will be paused during this maintenance?: YES
FASSE cluster will be paused during this maintenance?: YES

• Holyoke/MGHPCC rack 8a26 power work - 8am-6pm

  • Audience: All Cannon cluster users

  • Impact: The power distribution unit in the holy8a26 rack needs to be replaced, which necessitates us powering down the nodes in the rack. This will take place between 8am and 6pm on November 2nd. This is a partial outage for most partitions, jobs will adjust around the down nodes. For partitions suffering a full outage, please use other resources for the duration.

  • Public partitions:

    • test
      sapphire
      bigmem

  • Group or private partitions:

    • hsph
      itc_cluster

    • eddy

      janson_sapphire

      yao

      yao_alphatns

      yao_gpu

• Slurm 26.05.5 Upgrade

  • Audience: All cluster users (Cannon and FASSE)

  • Impact: Jobs will be paused and resumed after maintenance

• Login node reboots

  • Audience: All login nodes

  • Impact: Login nodes will be unavailable until after maintenance

• OOD/Open OnDemand reboots

  • Audience: All OOD users

  • Impact: OOD will be unavailable until after maintenance

• Netscratch 90-day retention cleanup

  • Audience: All netscratch users

  • Update: Due to low space on netscratch, please note that retention cleanup will be run more frequently. In addition, the retention period will be reduced to 60 days next month.

  • Impact: Files older than 90 days will be removed per our scratch policy. Please note that this cleanup can happen at any time, not just during maintenance.

Globus services will be unavailable for a period between 9:30am and 10:30am U.S. Central time on Saturday, October 24, 2026 due to planned maintenance (https://www.globus.org/blog/2026-10-24-maintenance) that includes database upgrades.

Please refer to the posted link for additional information on how services and in-progress work will be impacted during this window.