system news

Maintenance window, batch jobs will not run (2017-09-04 08:00)

  • Posted on: 25 August 2017
  • By: ake

On Sep 4th 08:00 CEST we will have a maintenance window to change some parameters for the lustre file system.

To be able to do this we need to empty the clusters from running jobs and reboot all nodes including the login nodes.

As we get closer to that point in time, jobs will not be allowed to start if their requested runtime is too long to fit before 2017-09-04 08:00.

 

In other words, submitting jobs with shorter runtimes will be a good idea.

2017-06-29: Disruptive Lustre server issues, queues suspended (22:45 queues now re-enabled)

  • Posted on: 29 June 2017
  • By: nikke

Node reboots to activate security updates seems to have triggered strange disruptive behavior in the Lustre service affecting all compute resources.

We are currently diagnosing the issue in order to come up with a solution.

 

*UPDATE 2017-06-29 22:45*

Lustre now looks stable again. Queues have been re-enabled.

2017-06-28: Maintenance on the Lustre file system finished, queues running again

  • Posted on: 20 June 2017
  • By: ake

On Wednesday, June 28 08:00 - 17:00, we will perform yet another maintenance on our Lustre filesystem.

This time two of the internal UPSes need to be replaced.

To be on the safe side we will take the Lustre file system offline during this process.

This means that no jobs will be allowed to run after 08:00.

 

*Update 2017-06-28 15:00*

The maintenance is finished and the queues are running again.

2017-06-13: Resolved: PFS outage. Systems back to normal.

  • Posted on: 13 June 2017
  • By: zao

The PFS file system is having some server problems and is currently not accessible.
Due to this, it is not possible to log in to the login nodes.

The batch queues are suspended as we work on this.

We will update this news entry as we make progress and/or resolve the problem.

2017-06-13 17:45
The failing hardware have been reported to our hardware vendor.

2017-06-14 12:24
We are working with the vendor to resolve the problem.

srun works again

  • Posted on: 24 May 2017
  • By: bbrydsoe

srun works again for parallel jobs on both Abisko and Kebnekaise, so in general you should no longer use mpirun.

Note that srun and mpirun are now equivalent in most cases, but mpirun does not always work when using standard input (mpirun program < file).

Pages

Updated: 2017-09-21, 11:05