FAS Research Computing - История на известията

Системата е в поддръжка

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.
В поддръжка

SLURM Scheduler - Cannon - В поддръжка

Cannon Compute Cluster (Holyoke) - В поддръжка

Boston Compute Nodes - В поддръжка

GPU nodes (Holyoke) - В поддръжка

seas_compute - В поддръжка

В поддръжка

SLURM Scheduler - FASSE - В поддръжка

FASSE Compute Cluster (Holyoke) - В поддръжка

В поддръжка

Kempner Cluster CPU - В поддръжка

Kempner Cluster GPU - В поддръжка

В поддръжка

FASSE login nodes - В поддръжка

В поддръжка

Cannon Open OnDemand - В поддръжка

FASSE Open OnDemand - В поддръжка

Работи

Netscratch (Global Scratch) - Работи

Home Directory Storage - Boston - Работи

Tape - (Tier 3) - Работи

Holylabs - Работи

Isilon Storage Holyoke (Tier 1) - Работи

Holystore01 (Tier 0) - Работи

HolyLFS04 (Tier 0) - Работи

HolyLFS05 (Tier 0) - Работи

HolyLFS06 (Tier 0) - Работи

Holyoke Tier 2 NFS (new) - Работи

Holyoke Specialty Storage - Работи

holECS - Работи

Isilon Storage Boston (Tier 1) - Работи

BosLFS02 (Tier 0) - Работи

Boston Tier 2 NFS (new) - Работи

CEPH Storage Boston (Tier 2) - Работи

Boston Specialty Storage - Работи

bosECS - Работи

Samba Cluster - Работи

Globus Data Transfer - Работи

История на известията

май 2026

Cannon cluster down
  • Решен
    UTC
    Решен

    Slurm crashed on 4:30p on Friday due to a user running a large sacct query against the Slurm database. This caused the database host to run out of memory and crash the scheduler. To prevent this from reoccurring we are reducing the time range that users are permitted to query at one time to 7 days. Thus if you need to cover a month you would need to query in four 7 day increments.

    We do ask users to be judicious in their querying of the Slurm. Only ask for those fields that you require. Please also ensure any AI agents you have running limit their queries appropriately.

  • Идентифициран
    UTC
    Идентифициран

    To temporarily stabilize the situation, we have reduced the maximum query time for sacct and other Slurm commands to be 1 day. We have filed a ticket with SchedMD to further analyze the issue.

    The cluster is back up and the scheduler is accepting new jobs.

    We will continue to monitor for emergencies over the weekend, and resume in-depth troubleshooting on Monday.

  • Разследва се
    UTC
    Разследва се

    The Slurm scheduler is experiencing an error which is impacting jobs. The Cannon cluster will be inaccessible while we troubleshoot.

    We are currently investigating this incident.

Account approvals down
  • Решен
    UTC
    Решен

    This incident has been resolved. Login is again available for approvals/onboarding.

  • Разследва се
    UTC
    Разследва се

    Account approvals through Portal are not available at this time.

    We are currently investigating this incident.

FASRC VPN (Cambridge) is back up
  • Решен
    UTC
    Решен
    FASRC VPN (Cambridge) is back up. This incident was automatically resolved by Instatus monitoring.
  • Разследва се
    UTC
    Разследва се
    FASRC VPN (Cambridge) is down at the moment. This incident was automatically created by Instatus monitoring.
FASRC VPN (Cambridge) is back up
  • Решен
    UTC
    Решен
    FASRC VPN (Cambridge) is back up. This incident was automatically resolved by Instatus monitoring.
  • Разследва се
    UTC
    Разследва се
    FASRC VPN (Cambridge) is down at the moment. This incident was automatically created by Instatus monitoring.
FASRC VPN (Cambridge) is back up
  • Решен
    UTC
    Решен
    FASRC VPN (Cambridge) is back up. This incident was automatically resolved by Instatus monitoring.
  • Разследва се
    UTC
    Разследва се
    FASRC VPN (Cambridge) is down at the moment. This incident was automatically created by Instatus monitoring.

апр 2026

OpenOnDemand maintenance
  • Завършен
    април 30, 2026 в 14:00UTC
    Завършен
    април 30, 2026 в 14:00UTC
    Maintenance has completed successfully
  • В ход
    април 30, 2026 в 12:00UTC
    В ход
    април 30, 2026 в 12:00UTC
    Maintenance is now in progress
  • Планиран
    април 30, 2026 в 12:00UTC
    Планиран
    април 30, 2026 в 12:00UTC

    At 8am on Thursday April 30th we will be upgrading from Open OnDemand version 4.0.7 to 4.1.4 on both the Cannon and FASSE clusters.

    This is not expected to impact running jobs.

    This upgrade adds the Jobs->Project Manager menu item and fixes an issue that affected access to the Clusters->Shell Access menu item when using Firefox.

login.rc.fas.harvard.edu is responding normally
  • Решен
    UTC
    Решен
    \[login.rc.fas.harvard.edu\](http://login.rc.fas.harvard.edu) is responding normally. This incident was automatically resolved.
  • Разследва се
    UTC
    Разследва се
    login.rc.fas.harvard.edu is not responding normally. This incident was automatically created.
Login and OOD node access restricted due to serious security issue - No ETA
  • Решен
    UTC
    Решен

    The cluster has been rebooted and all nodes, including login and OOD, have been patched.

    The scheduler is re-opened and jobs which were preempted/requeued have priority for re-scheduling.

    Some non-standard, lab-owned nodes may still require patching. The owners of these machines may be contacted about this.

    Thank you for your patience. This is a global issue and is being addressed at centers everywhere.

  • Актуализация
    UTC
    Актуализация

    To mitigate this exploit we will need to restart -all nodes- on the cluster.

    This will begin at 1PM and run until all nodes have restarted (no ETA).

    This will mean any un-finished jobs will be terminated. There is no way to avoid this.

    We will then be validating the fix before re-opening the login. OOD nodes, and scheduler.

    Next steps and updates will be posted here.

  • Актуализация
    UTC
    Актуализация

    We are developing a plan of attack to mitigate this exploit. Please know that this is a very serious issue and so we are treating it as such. Thank you for your understanding.

    We are currently awaiting further information from the Redhat/Fedora/Rocky community but building a plan in the meantime with the information we have. More details to follow as we can share them.

    If you need to access storage (except scratch and home directories), Globus is still online and available. But again, login nodes and OOD are not available.

  • Идентифициран
    UTC
    Идентифициран

    Due to a serious in-the-wild exploit which can compromise Fedora-based Linux distributions including Rocky, which is used on the cluster, we need to restrict access. All login and OOD nodes are shut down until a fix can be put in place. Jobs running on the cluster will continue running.

    No ETA, There is not fix at this time. We will update our status page in the morning once we have more information or a fix to roll out.

    This is a serious exploit and we do not take this measure lightly. Please follow this status page for updates and eventual resolution.

holylfs06 down
  • Решен
    UTC
    Решен

    Holylfs06 is accessible again.

    This incident has been resolved.

  • Идентифициран
    UTC
    Идентифициран
    Holylfs06 storage is down. We are investigating. More details as they are known.
Website security maintenance (www.rc and docs.rc) 4-28-26 1pm
  • Завършен
    април 28, 2026 в 17:16UTC
    Завършен
    април 28, 2026 в 17:16UTC

    Website maintenance has completed successfully.

  • В ход
    април 28, 2026 в 17:00UTC
    В ход
    април 28, 2026 в 17:00UTC
    Maintenance is now in progress
  • Планиран
    април 28, 2026 в 17:00UTC
    Планиран
    април 28, 2026 в 17:00UTC

    Security updates are required for www.rc.fas.harvard.edu and docs.rc.fas.harvard.edu
    This work will take place today between 1pm and 2pm
    Both sites will be down for very short periods during the updates.

мар 2026

Scheduler is degraded
  • Решен
    UTC
    Решен

    This incident has been resolved. The scheduler is running normally.

  • Разследва се
    UTC
    Разследва се

    The scheduler is in a degraded state due to thrashing
    We are actively working to resolve this problem.

two-factor.rc.fas.harvard.edu (openauth) error
  • Решен
    UTC
    Решен

    This incident has been resolved. two-factor.rc.fas.harvard.edu is working normally again.

  • Разследва се
    UTC
    Разследва се

    We are currently investigating this incident. Requesting a new token or re-requesting your token from two-factor is not currently working.

The web front end to two-factor.rc.fas.harvard.edu is currently not allowing logins, generating new tokens is currently unavailable
  • Решен
    UTC
    Решен
    This incident has been resolved.
  • Разследва се
    UTC
    Разследва се
    We are currently investigating this incident.
Network issues - Cluster degraded
  • Решен
    UTC
    Решен

    This incident has been resolved by draining and rebooting any nodes with stuck mounts.

  • Наблюдаваме
    UTC
    Наблюдаваме

    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.

  • Идентифициран
    UTC
    Идентифициран

    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.

  • Разследва се
    UTC
    Разследва се

    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..

ColdFront is down.
  • Решен
    UTC
    Решен

    Cold front is back up. Thank you for your patience.

  • Идентифициран
    UTC
    Идентифициран

    ColdFront is down. We are working to bring it back up. The instance got replaced last night, but it had trouble configuring itself on the way up again.

мар 2026 до май 2026

Следващ

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