FAS Research Computing - Network issues - Cluster degraded – Incident details

Experiencing partially degraded performance

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

Cluster Utilization (VPN and FASRC login required): Cannon | FASSE


Please scroll down to see details on any Incidents or maintenance notices.
Monthly maintenance occurs on the first Monday of the month (except holidays).

GETTING HELP
Documentation: https://docs.rc.fas.harvard.edu | Account Portal https://portal.rc.fas.harvard.edu
Email: rchelp@rc.fas.harvard.edu | Support Hours

The colors shown in the bars below were chosen to increase visibility for color-blind visitors.
For higher contrast, switch to light mode at the bottom of this page if the background is dark and colors are muted.

Network issues - Cluster degraded

Monitoring
Major outage
Started 3 days ago

Affected

Cannon Cluster

Degraded performance from 1:34 PM to 12:00 AM

SLURM Scheduler - Cannon

Degraded performance from 1:34 PM to 12:00 AM

Cannon Compute Cluster (Holyoke)

Degraded performance from 1:34 PM to 12:00 AM

Boston Compute Nodes

Degraded performance from 1:34 PM to 12:00 AM

GPU nodes (Holyoke)

Degraded performance from 1:34 PM to 12:00 AM

seas_compute

Degraded performance from 1:34 PM to 12:00 AM

Updates
  • Monitoring
    Monitoring

    Mounts to Holyoke Isilon (specifically /n/sw) are broken on numerous nodes across the cluster. We have a check rolling out to find these nodes so we can remediate them individually. Until remediated the cluster will be in a degraded state. Running jobs may randomly die or fail as they hit nodes that have stale mounts.

    It will be risky to run jobs for the next hour and then, after that point, the cluster will have a large number of nodes closed waiting for them to drain so we can reboot them and fix the mounts.

    At this time we are unaware of any holy-isilon problems other than the effect this had on cluster nodes/running jobs. We will update should we identify any data storage concerns.

  • Identified
    Identified

    Mounts to Holyoke Isilon (specifically /n/sw) are broken on numerous nodes across the cluster. We have a check rolling out to find these nodes so we can remediate them individually. Until remediated the cluster will be in a degraded state. Running jobs may randomly die or fail as they hit nodes that have stale mounts.

    It will be risky to run jobs for the next hour and then, after that point, the cluster will have a large number of nodes closed waiting for them to drain so we can reboot them and fix the mounts.

  • Investigating
    Investigating

    A network issue affecting storage critical to the cluster is It's causing instability. The cluster is currently in a degraded state as a result. We are looking into the problem. Updates to follow..

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