FAS Research Computing - Historique des incidents

Holyoke Tier 2 NFS connaît une panne partielle

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 partielle

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 - Panne partielle

Holyoke Specialty Storage - Opérationnel

holECS - Opérationnel

Isilon Storage Boston (Tier 1) - Opérationnel

BosLFS02 (Tier 0) - Opérationnel

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 incidents

sept. 2022

MGHPCC network maintenance Tuesday Sept 27, 2022 6-8 PM
  • Terminé
    septembre 28, 2022 à 00:00UTC
    Terminé
    septembre 28, 2022 à 00:00UTC

    Maintenance has completed successfully

  • En cours
    septembre 27, 2022 à 22:00UTC
    En cours
    septembre 27, 2022 à 22:00UTC

    Maintenance is now in progress

  • Pas encore commencé
    septembre 27, 2022 à 22:00UTC
    Pas encore commencé
    septembre 27, 2022 à 22:00UTC

    VPN UPDATE This maintenance has been extended to accommodate an VPN update from 6-7PM. The FASRC VPN will deprecate older, less-secure ciphers. VPN connections could drop during this update.

    FIBRE UPDATE As part of the 100Gb link upgrade, networking maintenance will occur at the MGHPCC/Holyoke data center Tuesday Sept 27, 200 starting at 7PM. We are setting aside one hour for this but, barring any issues, the updates should not take that long.

    Any interruption should be very brief. However, as always the possibility does exist for knock-on issues. We will post any updates here and FASRC staff will be monitoring for any related issues.

NESE tape hardware issues
  • Résolu
    UTC
    Résolu

    This issue is resolved.

    From the NESE team:

    The NESE tape system is up and running and functional now. The tape robot had a frame corruption, parts were replaced and it needed to re-discover hardware in each frame which takes a long time. Also there was an issue where most of the fiber switch ports needed to be manually enabled due work that was performed on the robot. Sorry for this extended downtime.

  • Identifié
    UTC
    Identifié

    The NESE tape robot and tape system is currently not functional. IBM is onsite tech (Saturday). Parts are required and ordered, but no immediate ETA.

    The Spectrum Scale disk cache is operational. Data can be written to disk cache, but no data can be sent to tape or retrieved from tape at this time.

    Updates as we have them.

gpu_test queue slow - note on use
  • Résolu
    UTC
    Résolu

    This issue has become one of supply and availability. As such, we are changing this issue to Maintenance as there is nothing further we can do until the vendor replaces the hardware. Please continue to use other GPU resources. We will notify the community once we have new hardware to replenish the gpu_test queue.

    Please see Running Jobs for other available resources.

  • Mettre à jour
    UTC
    Mettre à jour

    Difficulty obtaining replacement mainboards has this issue stalled. More information when we have it.

  • Mettre à jour
    UTC
    Mettre à jour

    Correction to first part of previous update:

    Currently only 2 of the nodes in gpu_test are working and the single remoteviz node is down. These will require help from the vendor to revive. We are also investigating with them the root cause for multiple node failures. No ETA

  • Mettre à jour
    UTC
    Mettre à jour

    Currently only 2 of the nodes in gpu_test are working and the single remoteviz node is down. These will require help from the vendor to revive. We are also investigating with them the root cause for multiple node failures. No ETA

    As a result of this, we are pre-emptively modifying the QoS for gpu_test now. We had planned to make a change on Thursday after announcing it in tomorrow's maintenance email, but this forces our hand.

    QoS changes for gpu_test effective now: Limited to 1 job per user. That job is limited to a maximum 16 cores and 90GB memory.

    See Running Jobs for a list of available partitions, including the gpu partition.

  • Identifié
    UTC
    Identifié

    Please note that this issue also affects the remoteviz partition.

    Updates as we have them.

  • Détecté
    UTC
    Détecté

    Several hosts in the gpu_test queue have become unresponsive and will require a physical visit to reset. Staff are en route to the data center.

    On a related note, please do not use gputest as a workaround for the regular gpuqueues.
    This is unfair to other users and ties up the gpu_test partition. This partition is not for general job use.

    We will be addressing this issue later this week by reducing the number of jobs allowed per user. This will be noted in tomorrow's maintenance email and then implemented Thursday.

Websites & Tools - SPINAL (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Resolved. This automated notifier is being disabled due to too-frequent false alarms.

  • Détecté
    UTC
    Détecté

    Websites & Tools - SPINAL (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

Websites & Tools - SPINAL (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - SPINAL (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - SPINAL (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

août 2022

Websites & Tools - MiniLIMs (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - MiniLIMs (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - MiniLIMs (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

Websites & Tools - Portal (portal.rc) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - Portal (portal.rc) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - Portal (portal.rc) cannot be accessed at the moment. This incident was created by an automated monitoring service.

NESE tape (Tier 3) upgrades
  • Terminé
    août 26, 2022 à 20:43UTC
    Terminé
    août 26, 2022 à 20:43UTC

    Maintenance has completed successfully.

  • En cours
    août 22, 2022 à 10:00UTC
    En cours
    août 22, 2022 à 10:00UTC

    Maintenance is now in progress

  • Pas encore commencé
    août 22, 2022 à 10:00UTC
    Pas encore commencé
    août 22, 2022 à 10:00UTC

    Our tier 3 tape system is part of and run by NESE (the NorthEast Storage Exchange).

    We have been informed that they will be performing a significant upgrade of their Spectrum Scale archive system starting August 15th. This is a multi-day upgrade and will take approximately 3 days (potential for longer). Tier 3 tape allocations will be unavailable during this upgrade.

    NESE has informed us that this maintenance will be deferred to next week (9/22/22).

Websites & Tools - MiniLIMs (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - MiniLIMs (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - MiniLIMs (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

holyscratch01 issues
  • Résolu
    UTC
    Résolu

    The offending jobs have been cancelled. Holyscratch01 should be usable again now.

  • Détecté
    UTC
    Détecté

    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.

juil. 2022

Websites & Tools - SPINAL (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - SPINAL (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - SPINAL (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

Websites & Tools - MiniLIMs (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - MiniLIMs (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - MiniLIMs (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

NESE tape (Tier 3) hardware maintenance/install
  • Terminé
    juillet 21, 2022 à 22:00UTC
    Terminé
    juillet 21, 2022 à 22:00UTC

    Maintenance has completed successfully

  • En cours
    juillet 21, 2022 à 10:00UTC
    En cours
    juillet 21, 2022 à 10:00UTC

    Maintenance is now in progress

  • Pas encore commencé
    juillet 21, 2022 à 10:00UTC
    Pas encore commencé
    juillet 21, 2022 à 10:00UTC

    Our tier 3 tape system is part of and run by NESE (the NorthEast Storage Exchange).

    We have been informed that they will be installing additional tape drives to the system on July 21st and this will require a whole day downtime to accomplish. Tier 3 tape archives will not be available to our users on that date.

Websites & Tools - MiniLIMs (FAS Informatics) isn't accessible at the moment (automated)
  • Résolu
    UTC
    Résolu

    Websites & Tools - MiniLIMs (FAS Informatics) is now operational! This update was created by an automated monitoring service.

  • Détecté
    UTC
    Détecté

    Websites & Tools - MiniLIMs (FAS Informatics) cannot be accessed at the moment. This incident was created by an automated monitoring service.

holylfs02 performance issues
  • Résolu
    UTC
    Résolu

    The unrepairable volume on holylfs02 is isolated to two labs and they have been informed of next steps. This issue does not affect other areas on holylfs02, so we are closing this incident.

  • Mettre à jour
    UTC
    Mettre à jour

    Recently we noticed an uptick in bad blocks on a RAID6 disk volume that is part of the entire filesystem. Generally speaking, the operating system will vector these out so no data is written there. During this period, we had to replace a number of drives due to failures; they are part of RAID 6 multi-disk set with dual parity and will rebuild with minimal impact on performance.

    What we believe happened is that, during the rebuild process, bad data was copied to the replacement disks and the filesystem got corrupted. One of the staff ran a read-only, non-destructive repair on the volume in question and noted quite a few errors.

    So far, we have 1) contacted the vendor (who gave us a command to clear additional bad blocks)
    2) NOT run the actual repair command to "fix" the filesystem (which would delete data) 3) contacted the vendor to see if they had any partners that might be able to assist. We are planning to meet with one of those partners early next week, but are not confident they will have a solution. At this point (and depending on how that meeting goes), the next step would be to run the repair, note the extent of loss, and attempt to get the volume remounted.

    Updates to follow when we have more information.

    Note that these shelves are covered under warranty support for another year. We see no need at present to replace the 3PB allocation with new hardware (this is planned to take place Q3-4 FY23)

  • Identifié
    UTC
    Identifié

    Users may experience issues connecting to holylfs02, or slow samba performance.

    This is due to a hardware issue with holylfs02, and we have opened a ticket with the vendor. Updates to follow.

    No ETA.

juil. 2022 à sept. 2022

Suivant

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