Columbia University Information Technology

Service Alerts

RESOLVED SERVICE ALERTS

Resolved Service Alerts

Resolved Service Alerts

  • Yeti HPC
    05/26/2015 - 09:00am to 05/26/2015 - 08:00pm

    **** Access to the cluster has been restored, as this work was completed ahead of schedule. ****

     

    CUIT is performing necessary maintenance on this service.  This scheduled maintenance is expected to be completed on-time.

    The Yeti HPC cluster will be down for maintenance and updates from 9am to 8pm on Tuesday,  May 26. Running jobs will be terminated Tuesday 5/26 at 9 am and users will not  have access to the cluster or to any files located in the cluster storage area during the downtime.

    CUIT apologizes for any inconvenience this may cause, but this action is necessary to ensure a more reliable service.  If clients continue to experience issues after the service-interruption hours, please contact the CUIT Helpdesk at:

    Web:     http://cuit.columbia.edu/support

    Phone:   212-854-1919

    Email:     askcuit@columbia.edu

     

    For a list of active and future Service Alerts, please visit http://cuitalerts.columbia.edu/.

     

  • Yeti HPC
    11/03/2015 - 08:30pm to 11/04/2015 - 07:36pm

    Access to this service was restored after moving this functionality to another server, during the scheduled maintenance downtime which ended around 7:30 this evening.

     

    The Yeti head node became unreachable starting at approximately 8:30 pm. Active user jobs are continuing to run but users cannot access the cluster and no new jobs can start. The problem is being investigated. *Update* The cause was most likely a hardware failure and the system will be serviced in the morning. As the cluster has a previously scheduled maintenance window starting at 8 am it has been decided that in order to minimize the effect on running jobs the head node will remain offline until then. The users have been notified.

  • Yeti HPC
    11/04/2015 - 08:00am to 11/04/2015 - 08:00pm

    **** Update ***** 

    The downtime maintenance was completed and the cluster returned to production slightly before 7:30 PM this evening. 

     ******************** 

     CUIT is performing necessary maintenance on this service.  This scheduled maintenance is expected to be completed on-time.


    The Yeti HPC cluster will be down for maintenance and updates from 8am to 8pm on Wednesday, November 4. Users will not  have access to the cluster or to any files located in the cluster storage area during the downtime.

    CUIT apologizes for any inconvenience this may cause, but this action is necessary to ensure a more reliable service.  If clients continue to experience issues after the service-interruption hours, please contact the CUIT Helpdesk at:

    Web:     http://cuit.columbia.edu/support

    Phone:   212-854-1919

    Email:     askcuit@columbia.edu

     

    For a list of active and future Service Alerts, please visit http://cuitalerts.columbia.edu/.

     

  • Yeti HPC
    12/05/2015 - 05:01pm to 12/05/2015 - 07:45pm

    Beginning around 5pm, The Yeti HPC execute nodes became unavailable due to A problem with the LDAP authentication servers.

    This problem was resolved as of 7:45 pm and the service is functioning normally. 

    For a complete list of recent and upcoming scheduled service changes, please visit http://cuitalerts.columbia.edu/.

  • Yeti HPC
    02/02/2016 - 08:00am to 02/02/2016 - 08:00pm

    Update: This maintenance is complete and the Yeti service is back online.

    CUIT is performing necessary maintenance on this service.   This scheduled maintenance is expected to be completed on-time.

    During the scheduled time of this Service Alert, the cluster will be unavailable to researchers and no compute jobs will run.

    CUIT apologizes for any inconvenience this may cause, but this action is necessary to ensure a more reliable service.  If clients continue to experience issues after the service-interruption hours, please contact the CUIT Helpdesk at:

    Web:     http://cuit.columbia.edu/support

    Phone:   212-854-1919

    Email:     askcuit@columbia.edu

     

    For a list of active and future Service Alerts, please visit http://cuitalerts.columbia.edu/.