During this downtime, we will connect our new water-cooled rack to water. All compute, including the GPUs, will be down (jobs will not run), but it will still be possible to log in, submit jobs, and access storage.

Queued jobs overlapping the maintenance window will not start. Running jobs will be re-queued at the start of the maintenance.

During this urgent downtime we will replace the faulty storage server that has been causing several crashes yesterday and today.

Please note that:

• MOMA VarSeq will also be affected by this downtime.
• This maintenance will not affect the clinical data dump.

At the start of the maintenance we will close all open connections and shut off access to the cluster for all users.

Queued jobs overlapping the maintenance window will not start. Running jobs will be re-queued at the start of the maintenance.

Faststorage was down due to a storage server hardware fault. We will take actions to replace the failing hardware.

Faststorage is currently down due to a hardware fault in the core storage switch. We're working on it.

Faststorage is currently down due to a hardware error. We're working on it.