Some systems are experiencing issues

About This Site

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

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

Global

Authentication

Operational

OpenAuth

Operational

VDI (Open OnDemand)

Operational

FASRC Portal

Operational

Samba Cluster (smbip)

Operational

Software and Licensing

Operational

FASRC Ticket System

Operational

FASRC Website

Operational

SPINAL (Informatics)

Operational

MiniLIMs (informatics)

Operational

FASRC Downloads Site

Operational

Web Proxies

Operational

Holyoke MGHPCC

Cannon Cluster Compute

Operational

SLURM Scheduler

Operational

ScratchLFS02

Operational

ScratchLFS (read-only - Decommission Feb 3)

Operational

Holyoke Specialty Storage

Operational

Virtual Infrastructure - Holyoke

Operational

Network - MGHPCC Holyoke

Operational

Infiniband - MGHPCC

Operational

HolyLFS

Performance Issues

HolyLFS02

Operational

HolyLFS03

Operational

Boston

Virtual Infrastructure - Boston

Operational

Primary/Lab Storage - Boston

Operational

Home directory storage - Boston

Operational

BosLFS

Operational

BosLFS02

Operational

Compute nodes - Boston

Operational

Network - Boston

Operational

Infiniband - Boston

Operational

Cambridge

Network - Cambridge

Operational

Scheduled Maintenance

Monthly Maintenance

Date: February 3rd, 2020 Time: 7:00AM-11:00AM EST

Monthly cluster maintenance takes place on the first Monday of each month (unless it is a holiday) from 7am-11am.

This includes scratchlfs02 90-day file retention cleanup (https://www.rc.fas.harvard.edu/policy-scratch/) and rebooting of the login nodes.

Details of all the tasks we will be performing can be found at https://www.rc.fas.harvard.edu/monthly-maintenance

Past Incidents

29th January 2020

No incidents reported

28th January 2020

No incidents reported

27th January 2020

ScratchLFS02 scratchlfs02

3:16pm: scratchlfs should be operational

3:06pm: looking into performance issues on scratchlfs02

HolyLFS holylfs cycling

1/28: Holylfs instability is due to the fact that the new Cannon cluster has outstripped the ability of holylfs to keep up with jobs. We are actively working on a replacement for this system but it will not be available until later this year. In the meantime we ask that users on holylfs not run production jobs on holylfs and only use it for long term storage. Instead please use our scratch filesystems for production jobs (see upcoming maintenance email for more information on this). If you need assistance with rearchitecting your workflow please contact us.

Update 2:40pm: General high load on holylfs is causing instability. We are looking into this further

11:46am: looking into performance issues on holylfs

10:11am: holylfs has been rebooted 9:52am: holylfs is being rebooted for the next 15 minutes due to stuck jobs

26th January 2020

No incidents reported

25th January 2020

No incidents reported

For issues not shown here, please contact FASRC via
https://portal.rc.fas.harvard.edu or email rchelp@rc.fas.harvard.edu