All systems are operational

About This Site

OFFICE HOURS: Office Hours will be held virtually until further notice. Please see


Status page for the Harvard FAS Research Computing cluster and other resources.

Please scroll down to see details on any Incidents or maintenance notices.

Monthly maintenance will occur on Monday, June 1st from 7am to 11am.

Monthly maintenance will occur on Monday, June 1st from 7am to 11am.

Please note that our office hours (Wednesdays 12pm -3pm) have moved online. Details:

The annual June power downtime at MGHPCC has been postponed until October. We will inform you of the final date and begin giving regular notices closer to the date.


  • Update security on MGHPCC server that creates two-factor tokens

    • Audience: All users
    • Impact: No impact expected as this service is redundant with Boston server, but potential to affect two-factor login
  • LNET Upgrades

    • Audience: Lustre filesystems
    • Impact: No visible impact expected.
  • BOS Globus managed endpoint

    • Audience: All Globus users
    • Impact: No impact. New endpoint will be added to serve Boston based filesystems.
  • Nvidia driver update - staggered over several days

    • Audience: All GPU users
    • Impact: None expected. Updates will occur in stages when nodes are idle or drained
  • Lustre root squash

    • Audience: Lustre filesystems
    • Impact: No impact. Noted for informational purposes.
  • Login and VDI nodes will be rebooted

    • Audience: Cluster users of VDI
    • Impact: Brief disruption to users currently logged in.
  • Renumber MGHPCC login nodes - public IPs only

    • Audience: Anyone logged into a holyoke login node
    • Impact: Will occur concurrent with login reboots
  • Scratch cleanup ( )

    • Audience: Cluster users
    • Impact: Files older than 90 days will be removed.

Thanks, FAS Research Computing

Reminder: Scratch 90-day file retention purging runs occur regularly not just during maintenance periods.

Past Incidents

19th May 2020

Github timeouts - intermittent

Some users are experiencing timeouts with Github. This is intermittent and does not apply to all. We are investigating.

  • A temporary workaround has been implemented until the routing issue is fixed upstream.

  • We have reached out to Github to update their Internet2 routing. Until then, github will be inaccessible from all machines in Holyoke (holylogin01-04, all compute nodes).

    Please use boslogin01-04 in the meantime if you need to access Github.

  • For issues not shown here, please contact FASRC via or email