2. Information
  3. maintenance
  4. 【Supercomputer】 Special Maintenance on the Compute Nodes in the System B and C



【Supercomputer】 Special Maintenance on the Compute Nodes in the System B and C

publication date : Oct.13, 2017

When updating the OS of the systems B and C to the latest RHEL 7.4, on the maintenance of October, calculation error was confirmed and the calculation of the program using Intel compiler resulted in NaN.

In addition, the function to refer to group authority is unstable and the file cannot be accessed due to insufficient privileges.

In order to solve these problems, we will return the OS of the compute node to the state before maintenance in October. We will do the work from 3:00 p.m. today. Since the calculation may result in the abnormal value, we will reboot the systems B and C all at once.

Along with this, all currently running jobs are re-executed. Please re-submit if re-execution is prohibited.

It will take about 40 minutes to reboot.
While rebooting, we will temporarily terminate the new job execution.

This work has no effect on the login node.

We apologize for any inconvenience this may cause.

The system maintenance has been completed, and all services are now restored as norma. We apologize for any inconvenience that this may have caused you.
Period 2017.10.13 15:00 ~2017.10.13 16:10
Range affected
Inquiry Supercomputing Section, IT Services Division, Information Management Department, Kyoto University
Inquiry Form

Back to Maintenance


Copyright © Institute for Information Management and Communication, Kyoto University, all rights reserved.