What will happen?

During 2023 NSC will upgrade the operating system of Tetralith from CentOS 7 to Rocky Linux 9.

Other changes will be kept to a minimum. You will still use SSH or Thinlinc to log in, you will still submit your jobs to Slurm, etc.

Most applications provided by NSC will continue to work. Minor adjustments to job scripts might be needed. Some of these applications will run natively in the new operating system, some will have been recompiled, others will be run in a CentOS 7 container using Apptainer.

Regardless of which method has been used to get an application running in the new environment, we have not seen any significant performance impact during our testing. Some applications are even running faster than before.

If you have built or installed your own application you will need to choose a suitable way to move the application to the new environment. There are several ways to do this. Which one to use depends on the application. NSC Support can assist you with this. Documentation will also be available on this page later.

There will not be any significant downtime due to the upgrade, only a short stop (approximately a day, probably early 2024) to upgrade the cluster "system server" (where the job scheduler and other similar components run).

During the upgrade process Tetralith will be split into two parts for several weeks:

  • One login node (tetralith-el7.nsc.liu.se, a.k.a tetralith1) and a number of compute nodes running CentOS 7 (the old operating system)

  • One login node (tetralith-el9.nsc.liu.se, a.k.a tetralith2) and a number of compute nodes running Rocky Linux 9 (the new operating system)

This split is done so that all users will have some time to test and move their jobs to the new environment.

The upgrade window (from when you will get access to the first upgraded compute nodes to when all compute nodes have been upgraded) will be approximately six weeks.

Jobs submitted from the CentOS 7 login node will (unless you override it by choosing a reservation) run on CentOS 7 compute nodes.

Jobs submitted from the Rocky 9 login node will (unless you override it by choosing a reservation) run on Rocky 9 compute nodes.

How to access e.g devel or lsda reservation nodes will be documented here later.

The number of upgraded compute nodes will gradually increase during the upgrade window. The rate at which this happens will be adjusted based on how fast users are moving their jobs.

At the end of the upgrade, all compute nodes will run the new operating system.

If we find applications that the user and NSC cannot easily get working in the new environment, we have the the option to keep some CentOS 7 compute nodes online until 2024-06-30 for those applications. However, we will do our best to make sure this does not happen.

When will things happen?

We aim to start the upgrade in the end of October. We will inform users as soon as the exact date is known, at least two weeks before the start of the upgrade.

The upgrade window (during which you need to move your jobs) will be approximately six weeks long.

Why are we doing this?

The current operating system CentOS 7 (which is based on RedHat Enterprise Linux 7) will not receive any security updates after 2024-06-30.

As Tetralith is planned to continue operating at least until January 2025, we need to upgrade the operating system.

This is not the final version of this document!

We will update this page as more information becomes available. Important information (e.g when we have decided on a date for this) will also be sent via email to all Tetralith users.

If you have questions not answered on this page

Contact NSC Support and ask us!

User Area

User support

Guides, documentation and FAQ.

Getting access

Applying for projects and login accounts.

System status

Everything OK!

No reported problems


NSC Express