System Updates Archive

  • UPDATE - Degraded filesystem performance on Farnam

    Filesystem performance on Farnam has improved. If you continue to experience any problems, please let us know at hpc@yale.edu.

  • Degraded filesystem performance on Farnam

    We are currently experiencing degraded filesystem performance on Farnam, and are looking into the issue. Sorry for the inconvenience.

  • UPDATE - Loomis Storage on Grace / Omega / Farnam Available

    UPDATE - 11/13/2018 12:10 - 

    Grace, Omega and Farnam temporarily lost contact with the Loomis GPFS storage system due to a networking issue. The issue has been resolved but most running jobs crashed and should be checked on. Sorry for the inconvenience.

    11/13/2018 - 12:05 - 

    Loomis storage mounted on Grace, Omega and Farnam is currently unavailable. Logins to Grace and Omega will be unavailable during this time. Updates to this issue will be posted here.  

  • Scheduled Maintenance on Ruddle

    Scheduled maintenance is currently being performed on Ruddle. The cluster is expected to be available by 5pm on Friday, 11/9. Please contact hpc@yale.edu if you have any questions.   

  • Data Center Maintenance - All Clusters

    11/4 - 4:15pm
     
    Due to required maintenance to the cooling system at the West Campus Data Center, all HPC clusters (including storage) will be unavailable starting at 4:00pm on Sunday, November 4, 2018.
     
    Please ensure that you have exited cleanly from any logins or interactive sessions before that time. We expect that the clusters except for Ruddle will be returned to service by the end of the day on Thursday, November 8, 2018.
     
    Immediately following the completion of the work on the cooling system, normal scheduled maintenance will be performed on Ruddle. We expect that Ruddle will be returned to service by the end of the day on Friday, November 9. We will send a communication to users of each cluster once it is available.
     
    For all clusters except Ruddle, as the maintenance window approaches, the Slurm scheduler will not start any job if the job’s requested wallclock time extends past the start of the maintenance period (4:00pm on Sunday, November 4, 2018).
     
    Please visit the status page on research.computing.yale.edu for the latest updates. If you have questions, comments, or concerns, please contact us at hpc@yale.edu.
     

  • Farnam Scheduled Maintenance

    08:00 - 9/10/2018 - 
     
    We will perform scheduled maintenance on Farnam starting on Monday, September 10, 2018 at 8:00 am through the end of the day on Wednesday, September 12, 2018.  During this time logins will be disabled. An email notification will be sent when the maintenance has been completed and the cluster is available.
     
    Please note that we have set a limit that prevents the scheduler from running any job whose walltime limit implies that it could still be running when the maintenance begins at 8:00am on September 10.  Jobs that could overlap will not run, but instead will wait with the reason “ReqNodeNotAvail”. You will either need to resubmit with a shorter walltime request or wait until the maintenance period ends.
     
    We will be retiring all M610 nodes during the next maintenance period. PIs with these nodes in their partitions have been notified.
     
    For more info on the hardware on Farnam, please go to research.computing.yale.edu/farnam#compute-hardware.
     
    Please visit the status page on research.computing.yale.edu for the latest updates. If you have questions, comments, or concerns, please contact us at hpc@yale.edu.
     

  • Scheduled Maintenance on Grace

    08:00 - 8/6/2018 - 8/8/2018

    Scheduled maintenance on Grace will be performed on Monday, August 6, 2018 at 8:00am, through the end of the day on Wednesday, August 8, 2018.  We will be performing preventative maintenance to ensure stable operation of the cluster.  During this time logins will be disabled on Grace. The Loomis GPFS storage will remain available and will be accessible from Omega and Farnam. An email notification will be sent when the maintenance has been completed, and the cluster is available.

  • Farnam Scheduler

    10:25 6/26 - YCRC is currently working bringing the Farnam scheduler back online. Updates will be posted here as more information becomes available. 

    10:35 6/26 - The Farnam scheduler is now back online. 

  • Milgram Scheduled Maintenance 6/4/2018

    Scheduled maintenance will be performed on Milgram beginning Monday, June 4th, 2018, at 8:00am.  Maintenance is expected to be completed by the end of day, Wednesday, June 6th.   During this time, logins will be disabled, including from lab workstations, and Milgram’s storage will not be available.  An email notification will be sent when the maintenance has been completed, and the cluster is available.

     

  • Scheduled Maintenance on Ruddle

    5/7/2018 - 8am - We will be performing scheduled maintenance on Ruddle beginning Monday, May 7th, 2018, at 8:00 am. Logins will be disabled, running jobs will be killed, and Ruddle’s storage (including all YCGA sequencing data) will be unavailable during this time. We expect maintenance to be completed by the end of Wednesday, May 9th. When we are finished, we will notify you via email when the cluster is available again.
     
    If you have questions, comments, or concerns please contact us at hpc@yale.edu.
     
     
     

Pages