FAS Research Computing - Histórico de avisos

Holyoke Tier 2 NFS Enfrentando Interrupção Parcial

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.
Operacional

SLURM Scheduler - Cannon - Operacional

Cannon Compute Cluster (Holyoke) - Operacional

Boston Compute Nodes - Operacional

GPU nodes (Holyoke) - Operacional

seas_compute - Operacional

Operacional

SLURM Scheduler - FASSE - Operacional

FASSE Compute Cluster (Holyoke) - Operacional

Operacional

Kempner Cluster CPU - Operacional

Kempner Cluster GPU - Operacional

Operacional

FASSE login nodes - Operacional

Operacional

Cannon Open OnDemand - Operacional

FASSE Open OnDemand - Operacional

Parcialmente indisponível

Netscratch (Global Scratch) - Operacional

Home Directory Storage - Boston - Operacional

Tape - (Tier 3) - Operacional

Holylabs - Operacional

Isilon Storage Holyoke (Tier 1) - Operacional

Holystore01 (Tier 0) - Operacional

HolyLFS04 (Tier 0) - Operacional

HolyLFS05 (Tier 0) - Operacional

HolyLFS06 (Tier 0) - Operacional

Holyoke Tier 2 NFS - Parcialmente indisponível

Holyoke Specialty Storage - Operacional

holECS - Operacional

Isilon Storage Boston (Tier 1) - Operacional

BosLFS02 (Tier 0) - Operacional

Boston Tier 2 NFS - Operacional

CEPH Storage Boston (Tier 2) - Operacional

Boston Specialty Storage - Operacional

bosECS - Operacional

Samba Cluster - Operacional

Globus Data Transfer - Operacional

Histórico de avisos

mar 2023

Portal security/SSL error
  • Resolvido
    UTC
    Resolvido

    The SSL cert issue has been fixed. Portal is accessible again. You may need to force-reload the page if you browser still complains (hold dorn Ctrl or Cmd while clicking reload).

  • Investigando
    UTC
    Investigando

    Users are unable to access Portal.rc.fas.harvard.edu
    We are currently investigating this problem.

Monthly Maintenance March 6th, 2023 7am-11am
  • Concluído
    março 06, 2023 em 17:06UTC
    Concluído
    março 06, 2023 em 17:06UTC

    Maintenance has completed successfully at 12:00PM.

  • Em curso
    março 06, 2023 em 16:30UTC
    Em curso
    março 06, 2023 em 16:30UTC

    Maintenance is still in progress as of 11:30AM as one of our vendors needs to complete some hardware work.

    Access to the Slurm scheduler and all compute are still paused, but access to storage and other services has been restored.

    We appreciate your patience.

  • Concluído
    março 06, 2023 em 16:00UTC
    Concluído
    março 06, 2023 em 16:00UTC

    Maintenance has completed successfully

  • Em curso
    março 06, 2023 em 12:00UTC
    Em curso
    março 06, 2023 em 12:00UTC

    Maintenance is now in progress

  • Ainda não começou
    março 06, 2023 em 12:00UTC
    Ainda não começou
    março 06, 2023 em 12:00UTC

    NOTICES

    The annual MGHPCC power downtime will take place June 5th-8th, 2023
    Calendar Event: https://www.rc.fas.harvard.edu/events/mghpcc-power-shutdown-2023/
    Blog Post: https://www.rc.fas.harvard.edu/blog/2023-downtime/

    GENERAL MAINTENANCE

    • NOTE: All jobs will be paused during maintenance to reduce heat load and allow data center cooling maintenance to take place.

    • Login node updates and reboots, VDI reboots
      Audience: VDI/OpenOnDemand users
      Impact: VDI will be unavailable during this and the above Slurm upgrade

    • RCSMB (samba) Boston network changes
      Audience: RCSMB shares mounted out of Boston
      Impact: Could cause brief share disconnects during updates

    • UPDATE: Nexus control plane supervisor switchover - ETA 5 minutes, short network disconnect while restarting

    • Login node updates/reboot and VDI node reboots
      Audience: Anyone logged into a a login node or VDI/OOD node
      Impact: Login and VDI/OOD nodes will be unavailable while updating and rebooting

    • Scratch cleanup ( https://docs.rc.fas.harvard.edu/kb/policy-scratch/ )
      Audience: Cluster users
      Impact: Files older than 90 days will be removed.
      Reminder: Scratch 90-day file retention purging runs occur regularly not just during maintenance periods.

    SECURITY UPDATES
    HUIT and the CIO Council have set a goal of reducing risk across all schools within the University. All schools are looking to reduce their outstanding vulnerability count 75% by June 2023. These numbers are based on HUIT security scans of our infrastructure.
    We at FAS Research Computing are responsible for thousands of physical and virtual machines. To make progress in reducing our total open vulnerabilty count, we’re going to update internal and user facing systems as part of scheduled monthly maintenance windows and on a rolling basis outside of these windows. This will generally mean running OS and security updates as needed and rebooting these nodes when required.

    This month, these hosts will get updates as part of our scheduled maintenance:
    boslogin01 - boslogin04
    holylogin01 - holylogin04
    holydtn01 - holydtn04
    xdmod4.rc.fas.harvard.edu
    rchelp.rc.fas.harvard.edu (our ticket system)
    rcsmtp.rc.fas.harvard.edu (our mail system)

    Thanks!
    FAS Research Computing
    Department and Service Catalog: https://www.rc.fas.harvard.edu/
    Documentation: https://docs.rc.fas.harvard.edu/
    Status Page: https://status.rc.fas.harvard.edu/

Holylfs04 accessibility
  • Resolvido
    UTC
    Resolvido

    Hardware maintenance was completed.

  • Identificado
    UTC
    Identificado

    Due to an immediate need, we are doing some work on holylfs04 right now to replace some batteries in the controllers. Files that are on the impacted controllers will be inaccessible until they are back on line. The work should be completed later this afternoon.

fev 2023

VDI unresponsive
  • Resolvido
    UTC
    Resolvido

    Normal VDI operation has resumed.

  • Investigando
    UTC
    Investigando

    VDI is unresponsive or results in an error. We are currently investigating this issue.

Monthly Maintenance Feb. 6th, 2023 7am-11am
  • Concluído
    fevereiro 06, 2023 em 16:00UTC
    Concluído
    fevereiro 06, 2023 em 16:00UTC

    Maintenance has completed successfully

  • Em curso
    fevereiro 06, 2023 em 12:00UTC
    Em curso
    fevereiro 06, 2023 em 12:00UTC

    Maintenance is now in progress

  • Ainda não começou
    fevereiro 06, 2023 em 12:00UTC
    Ainda não começou
    fevereiro 06, 2023 em 12:00UTC

    NOTICES

    GPU PARTITIONS
    The gpu_test partition is back in service. Job limits are now 64 cores, 8 GPU's, and 750G of RAM. Users can run up to 2 jobs.

    HOLIDAY NOTICE
    February 20th is a university holiday (Presidents' Day)

    GENERAL MAINTENANCE

    • OnDemand Version upgrade to 2.0.29
      Audience: VDI/OpenOnDemand users
      Impact: VDI will be unavailable during this and the above Slurm upgrade

    • Domain controller updates
      Audience: All cluster
      Impact: Could briefly impact some older systems, otherwise no impact expected

    • Login node and VDI node reboots and firmware updates
      Audience: Anyone logged into a a login node or VDI/OOD node
      Impact: Login and VDI/OOD nodes will be unavailable while updating and rebooting

    • Scratch cleanup ( https://docs.rc.fas.harvard.edu/kb/policy-scratch/ )
      Audience: Cluster users
      Impact: Files older than 90 days will be removed.

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

    Thanks!
    FAS Research Computing
    Department and Service Catalog: https://www.rc.fas.harvard.edu/
    Documentation: https://docs.rc.fas.harvard.edu/
    Status Page: https://status.rc.fas.harvard.edu/

jan 2023

VM and Ceph infrastructure issues - also affects FASRC ticket system
  • Atualizar
    UTC
    Atualizar

    The Ceph/VM cluster issues have subsided and performance should be back to normal.

  • Resolvido
    UTC
    Resolvido

    The virtual infrastructure is responding normally again.
    This is a known issue that we are working to eliminate, but which requires new infrastructure (and time and resources). Thanks for your patience.

  • Identificado
    UTC
    Identificado

    An issue with our virtual infrastructure is affecting virtual machines and services which rely on that including Ceph storage, VDI, the FASRC ticket system, and hosted virtual machines.

    Updates will be posted here.

Monthly Maintenance Jan. 9th, 2023 7am-11am
  • Concluído
    janeiro 10, 2023 em 04:00UTC
    Concluído
    janeiro 10, 2023 em 04:00UTC

    Maintenance has completed successfully

  • Em curso
    janeiro 10, 2023 em 00:00UTC
    Em curso
    janeiro 10, 2023 em 00:00UTC

    Maintenance is now in progress

  • Ainda não começou
    janeiro 10, 2023 em 00:00UTC
    Ainda não começou
    janeiro 10, 2023 em 00:00UTC

    NOTICES

    GPU PARTITIONS
    The gputest partition is back in service. Job limits are now 64 cores, 8 GPU's, and 750G of RAM. Users can run up to 2 jobs.

    GLOBUS PERSONAL CLIENT - 3.1 Client Deprecated
    If you are using the Globus Connect Personal client on your machine, please ensure you have updated and are running version 3.2 or greater. Version 3.1 and below are deprecated and will not work as of December 17th, 2022. https://docs.globus.org/ca-update-2022/#globus    connect_personal

    HOLIDAY NOTICE
    January 16th is a university holiday (MLK Day)

    GENERAL MAINTENANCE

    * Slurm upgrade
    Audience: Cluster users
    Impact: Jobs will be paused during upgrade

    * OnDemand Version upgrade to 2.0.29
    Audience: VDI/OpenOnDemand users
    Impact: VDI will be unavailable during this and the above Slurm upgrade

    * Domain controller updates
    Audience: All cluster
    Impact: Could briefly impact some older systems, otherwise no impact expected

    * Login node and VDI node reboots and firmware updates
    Audience: Anyone logged into a a login node or VDI/OOD node
    Impact: Login and VDI/OOD nodes will be unavailable while updating and rebooting

    * Scratch cleanup ( https://docs.rc.fas.harvard.edu/kb/policy-scratch/ )
    Audience: Cluster users
    Impact: Files older than 90 days will be removed.

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

    Thanks!
    FAS Research Computing
    Department and Service Catalog: https://www.rc.fas.harvard.edu/
    Documentation: https://docs.rc.fas.harvard.edu/
    Status Page: https://status.rc.fas.harvard.edu/

jan 2023 para mar 2023

Seguinte

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