All systems are operational.

Last updated 1 month ago
    • Mass Storage (/pnfs)

      Grid-enabled distributed mass storage filesystem.

      Operational
    • Batch System

      Job submission infrastructure for local users as well as grid experiments.

      Operational
    • User Interfaces - mX machines

      Interfaces for the users to access the cluster’s compute and storage resources.

      Operational
    • Network

      Internal cluster network and its Internet access.

      Operational

Past incidents

7 years ago 2019-10-14 10:58 UTC
Fixed

Fixed

7 years ago 2019-10-14 12:56 UTC

Batch service is back. Unfortunately all jobs were purged during the investigation, so you will have to resubmit your jobs.

The issue was due to a remnant job file defined as array that created a segmentation fault, probably because the batch software could not handle the number of created jobs by the array. Once the file was removed, the batch system did not crash during start anymore.

Reported

7 years ago 2019-10-14 10:58 UTC

The batch system software is currently failing to restart. We are investigating. No local submissions are possible. Grid jobs will continue running.

No incidents reported.

No incidents reported.

No incidents reported.

No incidents reported.

No incidents reported.

No incidents reported.