Downtime 11/2 - 13/2

We need some downtime in the server rooms Friday February 11. to Sunday February 13. to do maintenance.

Status:

13.02.22, 19:00 We are done for the weekend and you can now log in. All linux machines now run StorNext 7.0.2. Infiniband drivers (OFED) have been upgraded on compute nodes. CUDA has been upgraded to 11.6 on all GPU-nodes. MacOS has been upgraded to 12.2.1 (Monterey) on all Mac workstations. Some work remain on the Mac's. We have some problems with beehive30, so login will be disabled.

11.02.22, 23:00 StorNext servers upgraded

11.02.22, 18:00 Upgrades progress as planned.

11.02.22, 08.30 StorNext server upgrade in progress.

11.02.22, 07:00 StorNext metadata backup.

11.02.22, 00.00 Login disabled. StorNext unmounted and shut down. Waiting for storage queues to quiet down.

Downtime:

During the downtime we will:

  • Upgrade StorNext to v. 7.0.2
  • Upgrade to the latest RedHat 7.9 kernels
  • Upgrade Infiniband OFED drivers on the clusters
  • Upgrade CUDA on our GPU-machines
  • Do maintenance and upgrades on Infiniband switches

We will start shutdown at 00.00 on Friday February 11.

You will not be able to log on to the system from Friday through Sunday. Email will be available as normal

At the moment we do not have a detailed schedule for the work, so we reserve the whole weekend. We may end up finishing earlier. We will get back with more detailed plans closer to the downtime.

 

 

By Torben Leifsen
Published Jan. 19, 2022 5:44 PM - Last modified Feb. 13, 2022 7:09 PM