FAS Research Computing - Hinweisverlauf

Alle Systeme betriebsbereit

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.
Betriebsbereit

SLURM Scheduler - Cannon - Betriebsbereit

Cannon Compute Cluster (Holyoke) - Betriebsbereit

Boston Compute Nodes - Betriebsbereit

GPU nodes (Holyoke) - Betriebsbereit

seas_compute - Betriebsbereit

Betriebsbereit

SLURM Scheduler - FASSE - Betriebsbereit

FASSE Compute Cluster (Holyoke) - Betriebsbereit

Betriebsbereit

Kempner Cluster CPU - Betriebsbereit

Kempner Cluster GPU - Betriebsbereit

Betriebsbereit

FASSE login nodes - Betriebsbereit

Betriebsbereit

Cannon Open OnDemand - Betriebsbereit

FASSE Open OnDemand - Betriebsbereit

Betriebsbereit

Netscratch (Global Scratch) - Betriebsbereit

Home Directory Storage - Boston - Betriebsbereit

Tape - (Tier 3) - Betriebsbereit

Holylabs - Betriebsbereit

Isilon Storage Holyoke (Tier 1) - Betriebsbereit

Holystore01 (Tier 0) - Betriebsbereit

HolyLFS04 (Tier 0) - Betriebsbereit

HolyLFS05 (Tier 0) - Betriebsbereit

HolyLFS06 (Tier 0) - Betriebsbereit

Holyoke Tier 2 NFS - Betriebsbereit

Holyoke Specialty Storage - Betriebsbereit

holECS - Betriebsbereit

Isilon Storage Boston (Tier 1) - Betriebsbereit

BosLFS02 (Tier 0) - Betriebsbereit

Boston Tier 2 NFS - Betriebsbereit

CEPH Storage Boston (Tier 2) - Betriebsbereit

Boston Specialty Storage - Betriebsbereit

bosECS - Betriebsbereit

Samba Cluster - Betriebsbereit

Globus Data Transfer - Betriebsbereit

Hinweisverlauf

Aktuellen Status anzeigen

Aug. 2024

holylfs04 down
GelöstTeilausfall23 Minuten
  • GelöstUTC
    holylfs04 is back up
  • AnalyseUTC
    Holylfs04 needs to be rebooted to address performance issues.
Cannon cluster degraded
GelöstBeeinträchtigte Leistung4 Stunden 57 Minuten
  • GelöstUTC
    The power issue has been resolved. All nodes that were drained are now being re-opened.
  • AnalyseUTC

    Due to temporary power availability issues in MGHPCC, half the nodes in 8A have been set to drain. This affects all partitions. Jobs will take longer to schedule but will not be terminated.

    We will reopen these nodes once the power issue has been resolved at the datacenter. No ETA at this time

boslfs02 degraded
GelöstBeeinträchtigte Leistung8 Minuten
  • GelöstUTC
    boslfs02 is back up
  • AnalyseUTC
    boslfs02 needs to be rebalanced. Access may be inconsistent
AUG24
Starfish upgrade
AbgeschlossenWartung84 Stunden
  • AbgeschlossenAugust 27, 2024 um 12:00UTC

    Starfish is back up

  • UpdateAugust 26, 2024 um 14:35UTC

    Starfish maintenance is still ongoing, no ETA at this time.

  • In BearbeitungAugust 24, 2024 um 00:00UTC
    Maintenance is now in progress
  • GeplantAugust 24, 2024 um 00:00UTC

    The Starfish Zones Dashboard will be undergoing a few upgrades and maintenance this weekend from Friday, August 23rd at 8AM until Monday, August 26th at 8AM. The dashboard will not be accessible during this time. Further details will be provided, if needed. Please email rchelp@rc.fas.harvard.edu if you have any questions or concerns.

Rebalancing boslfs02
GelöstBeeinträchtigte Leistung11 Minuten
  • GelöstUTC
    This incident has been resolved.
  • ÜberwachungUTC

    We are working on rebalancing boslfs02, there may be a short period of degraded performance

Juli 2024

JUL30
Coldfront maintenance 7/30/24 noon-12:30
AbgeschlossenWartung58 Minuten
  • AbgeschlossenJuli 30, 2024 um 16:58UTC

    Maintenance completed.

  • GeplantJuli 30, 2024 um 16:00UTC
    Coldfront will be unavailable Tuesday July 30th from 12:00-12:30 for crucial maintenance.
boslfs02 down
GelöstBeeinträchtigte Leistung20 Minuten
  • GelöstUTC
    boslfs02 is back up
  • IdentifizierungUTC
    boslfs02 is currently experiencing degraded performance. We are actively working on this issue.
Authentication issues - Related to global Crowdstrike incident
GelöstTeilausfall75 Stunden 59 Minuten
  • GelöstUTC

    All Crowdstrike-related resources are back up and operational.

  • UpdateUTC
    For FASRC resources affected by the Crowdstrike issue, most are back in full services. A few remaining issues involving the following may not be resolved until Monda: - waywiser2 - proteomics2 - tmsdb3 - lic3
  • UpdateUTC

    Please see HUIT Status (harvard.edu) for additional information on the global issue caused by Crowdstrike security which Harvard relies on. This is an ongoing issue university-wide.

    The systems that continue to be affected at FASRC are minimal, but some Windows-based systems managed by or connected to FASRC may still be affected.

  • ÜberwachungUTC

    Authentication is back up and running. Windows machines are still in a bad state and will need remedial work to get them back in service.

  • IdentifizierungUTC

    Authentication is back up and running. Windows machines are still in a bad state and will need remedial work to get them back in service.

  • AnalyseUTC

    Authentication is back up and running. Windows machines are still in a bad state and will need remedial work to get them back in service.

JUL18
boslogin01 rebooting
GelöstWartung1 Stunde 8 Minuten
  • GelöstUTC
    boslogin01 is rebooted and ready for use.
  • IdentifizierungUTC
    boslogin01 will be rebooted at 11:05AM. Due to stuck mounts it is currently not working as expected. Please save any necessary work before the reboot.
Infiniband networking down
Gelöst13 Minuten
  • UpdateUTC
    This incident has been resolved.
  • GelöstUTC
    Infiniband has been restored
  • IdentifizierungUTC
    The Infiniband fabric in MGHPCC that connects nodes with high-speed fiber is down. We are investigating. This _will_ cause performance issues and jobs may be stalled. Updates as we learn more.

Juni 2024

Holylabs down
GelöstSchwerer Ausfall11 Minuten
  • GelöstUTC

    Holylabs has finished rebooting and should be accessible. This incident has been resolved.

  • AnalyseUTC

    Due to mounting and access issues, holylabs needs to be rebooted. It will be unavailable during this time. We will update once it is back up.

FASRC websites unavailable
Gelöst6 Minuten
JUN26
FASRC websites -Unplanned maintenance (www.rc.fas.harvard.edu and docs.rc.fas.harvard.edu)
AbgeschlossenWartung11 Minuten
  • AbgeschlossenJuni 26, 2024 um 04:29UTC

    This update was completed successfully.

  • GeplantJuni 26, 2024 um 04:17UTC

    An unplanned maintenance on www.rc.fas.harvard.edu and docs.rc.fas.harvard.edu is required.

    ETA is approximately 1 hour . We apologize for any inconvenience.

Scratch instability
GelöstBeeinträchtigte Leistung41 Minuten
  • GelöstUTC
    This incident has been resolved.
  • AnalyseUTC

    Global scratch (/n/holyscratch01) is currently experiencing instability.

    We are currently investigating this incident.

JUN24
MGHPCC Pod 8A Power Upgrade June 24 will idle some Cannon nodes
AbgeschlossenWartung23 Stunden 59 Minuten
  • AbgeschlossenJuni 25, 2024 um 04:00UTC
    Maintenance has completed successfully
  • In BearbeitungJuni 24, 2024 um 16:01UTC
    Maintenance is now in progress
  • GeplantJuni 24, 2024 um 04:01UTC

    MGHPCC will be performing power upgrades on Pod 8A in order to increase density and allow more nodes to be added in that Pod's rows.  Similar to the May 13th work, this means that we will be idling half the nodes in 8A on two dates: June 17 and June 24th.

    These are all day events, meaning that the nodes in question will not be available for the 24 hours of that day.  This is being accomplished via reservations. So no jobs will be canceled but nodes will be drained and users may notice that their jobs may pend longer than normal as the scheduler idles these nodes.

    Where possible, please use or include other partitions in your job scripts and plan accordingly for any new or long-running jobs during that period: https://docs.rc.fas.harvard.edu/kb/running-jobs/#Slurm_partitions

    This affects the Cannon cluster. FASSE is not affected.

    Impacted partitions are:

    arguelles_delgado_gpu

    bigmem_intermediate

    bigmem

    blackhole_gpu

    eddy

    enos

    gershman gpu

    hejazi hernquist_ice

    hoekstra hsph

    huce_ice

    iaifi_gpu

    iaifi_gpu_priority

    iaifi_gpu_requeue

    intermediate

    itc_gpu

    itc_gpu_requeue

    joonholee

    jshapiro

    jshapiro_priority

    jshapiro_sapphire

    kempner

    kempner_dev

    kempner_h100

    kempner_requeue

    kempner_reservation

    kovac

    kozinsky

    kozinsky_gpu

    kozinsky_priority

    kozinsky_requeue

    murphy_ice

    ortegahernandez_ice

    sapphire

    seas_compute

    seas_gpu siag

    siag_combo

    siag_gpu

    sur test

    yao

    yao_priority

    zhuang

Zurück

Juni 2024 zu Aug. 2024

Nächste

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