FAS Research Computing - Historique des avis

BosLFS02 (Tier 0) connaît une panne majeure

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.
Opérationnel

SLURM Scheduler - Cannon - Opérationnel

Cannon Compute Cluster (Holyoke) - Opérationnel

Boston Compute Nodes - Opérationnel

GPU nodes (Holyoke) - Opérationnel

seas_compute - Opérationnel

Opérationnel

SLURM Scheduler - FASSE - Opérationnel

FASSE Compute Cluster (Holyoke) - Opérationnel

Opérationnel

Kempner Cluster CPU - Opérationnel

Kempner Cluster GPU - Opérationnel

Opérationnel

FASSE login nodes - Opérationnel

Opérationnel

Cannon Open OnDemand - Opérationnel

FASSE Open OnDemand - Opérationnel

Panne majeure

Netscratch (Global Scratch) - Opérationnel

Home Directory Storage - Boston - Opérationnel

Tape - (Tier 3) - Opérationnel

Holylabs - Opérationnel

Isilon Storage Holyoke (Tier 1) - Opérationnel

Holystore01 (Tier 0) - Opérationnel

HolyLFS04 (Tier 0) - Opérationnel

HolyLFS05 (Tier 0) - Opérationnel

HolyLFS06 (Tier 0) - Opérationnel

Holyoke Tier 2 NFS - Opérationnel

Holyoke Specialty Storage - Opérationnel

holECS - Opérationnel

Isilon Storage Boston (Tier 1) - Opérationnel

BosLFS02 (Tier 0) - Panne majeure

Boston Tier 2 NFS - Opérationnel

CEPH Storage Boston (Tier 2) - Opérationnel

Boston Specialty Storage - Opérationnel

bosECS - Opérationnel

Samba Cluster - Opérationnel

Globus Data Transfer - Opérationnel

Historique des avis

Voir l’état actuel

oct. 2023

VPN warning
Résolu

  • A VPN certificate for vpn.rc.fas.harvard.edu expired, which led to "Untrusted Server Blocked" messages when attempting to connect. A new certificate has already been added, and you should no longer be getting any errors when connecting to the VPN.

    This incident has been resolved.

    UTC
Major power event at MGHPCC (Holyoke) data center
Résolu40 heures 58 minutes
  • RésoluUTC

    FASSE login and OOD have been returned to service.

  • SurveilléUTC

    Most resources are once again available. The Cannon (including Kempner), FASSE, and Academic cluster are open for jobs. Please note that FASSE login and OpenOnDemand (OOD) nodes are not yet available. ETA Monday morning.

    Thanks for your patience through this unexpected event.

  • Mettre à jourUTC

    Power-up is progressing with, so far, only minor issues which we are addressing according to their impact on returning the cluster to service.

    Expect some remaining effects on less-essential services into tomorrow.

    Please note that login nodes will remain down until we return the cluster and scheduler to service.

  • Mettre à jourUTC

    MGHPCC has isolated the cause of the generator failure and will continue to look into the grid failure.

    At this time they will begin re-energizing the facility. Once that is complete and we have confirmed the networking is stable we can begin powering up our resources.

    Please bear with us as this is a long process given the number of systems we maintain and it must be done in stages. Watch this page for updates.

  • IdentifiéUTC

    With an abundance of caution FASRC and other MGHPCC occupants will not attempt to rush to restoration but will wait until the facility has restored primary power and confirmed stable operation before attempting to resume normal operations.

    As such, we expect to begin restoring FASRC services tomorrow (Sunday). Since all Holyoke services and resources are down, this is a lengthy process similar to the startup process after the annual power-down.

    Updates will be posted here. Please consider subscribing to our status page (see 'Get Updates' up top).

  • En cours d'investigationUTC

    There has been a major power event at MGHPCC, our Holyoke data center.
    We are awaiting further details

    This likely affects all holyoke resources including the cluster and storage housed in holyoke.

    More details as we learn them.

Slurm - critical security patch
RésoluPanne majeure1 heure 23 minutes
  • RésoluUTC

    The security patch has been applied, and all clusters are accepting jobs at this time.

  • En cours d'investigationUTC

    SchedMD (the maintainers of Slurm) have discovered a critical security flaw in Slurm. Due to the nature and severity of the issue, we will be immediately applying this patch.

    Cannon and FASSE schedulers will remain down for the duration of the patching. All running jobs will be paused, and new jobs will not be accepted until the scheduler is back up.

    ETA is expected to be approximately one hour.

Coldfront Offline
Résolu17 minutes
  • RésoluUTC

    This incident has been resolved.

  • En cours d'investigationUTC

    We are currently investigating this incident.

OCT2
FASRC monthly maintenance Monday October 2nd, 2023 7am-11am
TerminéMaintenance4 heures
  • Terminéoctobre 02, 2023 à 15:00UTC

    Maintenance has completed successfully

  • En coursoctobre 02, 2023 à 11:00UTC

    Maintenance is now in progress

  • Planifiéeoctobre 02, 2023 à 11:00UTC

    FASRC monthly maintenance will take place Monday October 2nd, 2023 from 7am-11am

    NOTICES

    New training sessions are available. Topics include New User Training, Getting Started on FASRC with CLI, Getting Started on FASRC with OpenOnDemand, GPU Computing,  Parallel Job Workflows, and Singularity. To see current and uture training sessions, see our calendar at: https://www.rc.fas.harvard.edu/upcoming-training/

    MAINTENANCE TASKS

    Cannon cluster will be paused during this maintenance?: Yes
    FASSE cluster will be paused during this maintenance?: No

    Cannon UFM updates
    -- Audience: Cluster users
    -- Impact: The cluster will be paused while this update takes place.

    Login node and OOD/VDI reboots
    -- Audience: Anyone logged into a login node or VDI/OOD node
    -- Impact: Login and VDI/OOD nodes will rebooted during this maintenance window  

    Scratch cleanup ( https://docs.rc.fas.harvard.edu/kb/policy-scratch/ )
    -- Audience: Cluster users
    -- Impact: Files older than 90 days will be removed. Please note that retention cleanup can run at any time, not just during the maintenance window.

    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/

sept. 2023

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées1 heure 36 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées6 heures 17 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées4 heures 4 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées7 heures 41 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées2 heures 3 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

août 2023

holylabs inaccessible - cluster paused
Résolu22 heures 2 minutes
  • RésoluUTC

    We have reverted the routing systems involved in causing this issue to their previous version/state. We will continue to investigate why this issue occurred and what will allow us to upgrade these systems at a later date.

  • En cours d'investigationUTC

    The issue with holylabs (and potentially other lustre filesystems) has recurred. This may have effects on jobs and any process using this and potentially other lustre filesystems.

    No ETA at this time.

  • RésoluUTC

    We have restored access to holylabs and the cluster/jobs are no longer paused.

    We have identified a root cause which we will be working to remediate to prevent this issue in future.

  • IdentifiéUTC

    The scheduler and all jobs have been paused in order to reduce the load on holylabs.

    We are continuing to work on a fix for this incident.

  • En cours d'investigationUTC

    The holylabs filesystem is currently down due to high load.

    OOD, software, and modules are all functional but if your workflow uses holylabs for storage, scripts, or jobs it may hang or fail. Our engineers are investigating this issue further.

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées11 heures 57 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

holyscratch01 performance
Résolu71 heures 8 minutes
  • RésoluUTC

    This incident has been resolved.

  • En cours d'investigationUTC

    holyscratch01 is in a degraded state currently. A group of improperly architected jobs are hammering the filesystem which is impeding access for other users. We are in the process of identifying and stopping these jobs.

    Until then, we recommend starting an interactive session and working from there as those will have the lowest impact. However, performance will still be slow until we are able to stop the problematic jobs.

AOÛT7
FASRC Monthly maintenance August 7, 2023 7am-1pm *NOTE EXTENDED TIME*
TerminéMaintenance2 heures 42 minutes
  • Terminéaoût 07, 2023 à 13:42UTC

    Due to a vendor error we were unable to complete holyscratch01 disk shelf replacement. We will work with the vendor to reschedule.

    All other maintenance tasks have completed.

  • En coursaoût 07, 2023 à 11:00UTC

    Maintenance is now in progress

  • Planifiéeaoût 07, 2023 à 11:00UTC

    August maintenance will run August 7, 2023 from 7am-1pm.

    Please note the extended timeframe.
    See tasks section below for explanation.

    NOTICES

    • CentOS 7 Support EOL: We will be dropping support for CentOS 7 support in September. If your machine or VM is CentOS 7 and connects with Slurm please contact FASRC to discuss options.

    • Test Partition Changes: We are changing test partitions based on changing needs and increasing max time to 12hrs instead of 8 hrs. A reminder that this partition is not for running jobs.

    MAINTENANCE TASKS

    • holyscratch01 Disk Shelf Replacement  All Jobs Will Be Paused
      -- Audience : All cluster and scratch users - Cannon and FASSE
      -- Impact:  Hardware issues with holyscratch01 necessitate the replacement of one of the disk shelves. As a result all jobs and scratch will need to be paused for the duration.
      -- ETA: This swap is expected to take 3-4 hours, but pausing the cluster, vendor interactions, and allowing a margin for over-run requires that we extend maintenance by 2 hours  (7am-1pm)

    • Login node and OOD/VDI reboots
      -- Audience: Anyone logged into a a login node or VDI/OOD node
      -- Impact: Login and VDI/OOD nodes will rebooted during this maintenance window  

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

    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/

Ceph instability - Affects Boston VMs (Virtual Machines) and Tier2 Ceph shares
RésoluPerformances dégradées48 heures 9 minutes
  • RésoluUTC

    The Ceph instability has been resolved. Caeph Tier2 shares, VDI, and VMs should be back to their normal state.

    If your VM, /net/fs-[labname] share, or VDI session is still impacted, please contact rchelp@rc.fas.harvard.edu

  • IdentifiéUTC

    The infrastructure behind Tier2 Ceph shares and VMs is unstable.
    This also affects VDI/OOD which relies on virtual machines.

    /net/fs-[labname] shares, new OOD/VDI sessions, and VMs are affected and may will be inaccessible until this is resolved.

    Thanks for your patience.

Précédent

août 2023 à oct. 2023

Suivant

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