Educloud Research Operational Log
We are currently experiencing issues with login attempts to Educloud Selfservice using ID-porten.?
Login with Educloud credentials is not affected.
We are working to resolve this issue.
?
?
There are currently issues preventing login to Educloud Self Service with ID-porten (BankID) authentication. Work is ongoing to resolve this.
Update: the issue was resolved on 2026-04-02 21:25.
The network switch that the GPU nodes in Fox are connected to has died. ?This means the GPU nodes are down. ?We are working on installing a replacement switch.
Update, 2026-01-26 12:56: The GPU nodes have been moved to the replacement switch and are back in production again.
We are currently experiencing issues with the selfservice portal, selfservice.educloud.no.
Users will be met with "selfservice.educloud.no is not responding"
We are working on a solution and will update this message when it is resolved.
Update: fixed at 10AM, same day.
The network infrastructure and Slurm will be upgraded during this 3-day maintenance window. (Originally planned as a 2-day maintenance window)
This affects login, interactive and all compute, gpu and bigmem nodes. No jobs can be submitted or run during this time.
Educloud VDIs and storage are unaffected
2026-01-13 16:29: the downtime has been extended until 2026-01-14 16:00. If possible, we'll bring up parts of the cluster throughout the day.?
2026-01-14 16:14: the downtime has been extended until 2026-01-15.
2026-01-15 14:19: Login nodes are opened for access again, and Fox is back in production. ?A few nodes are still unavailable, notably the interactive nodes.
2026-01-19: All the interactive nodes are available again.
2026-01-19: The remaining compute nodes are now in production again.
?
Educloud On Demand is not responding for the moment, and users will therefore not be able to access On Demand.
We are working to fix the issue.
circa 13:00: The problem is solved
We are currently experiencing issues with Educloud on Demand.?
Users trying to access the site receives a message saying the service is not responding (Tjenesten ondemand.educloud.no svarer ikke).
The network infrastructure and Slurm will be upgraded during the 2-day maintenance window.
This affects login, interactive and all compute/gpu/bigmem nodes. No jobs can be submitted or run during this time.
Educloud VDIs and storage are unaffected.
Update: postponed from 8-9 Dec to 10-11 Dec.
[14.07.25] The issues with storage in Educloud are now partly resolved and all users should have regained access to files.
Beware that sign in and loading of files might be working a bit slower than usual.
We are working on a permanent solution.
[10.07.25] We are currently experiencing issues with the storage services in Educloud. This means that access to both personal and project folders is unavailable at the moment.
We are working to identify and resolve the issue, and we will update this message as soon as we know more.
We apologize for the inconvenience this may cause.
Update: Maintenance was unsuccessful and has to be rescheduled.
view.educloud.no will experience scheduled downtime on Thursday, 11 September, from 15:00 to 18:00. Connecting to a new session during the downtime might not be possible. If you've connected to a machine since Friday, 5 September, you will be able to reconnect to this machine.
Emergency downgrade causing 1-5 minutes downtime for NFS and SMB on file storage in Educloud.
We're rebuilding the following CUDA modules on Fox:
- PyTorch/2.1.2-foss-2023a-CUDA-12.1.1
- PyTorch-bundle/2.1.2-foss-2023a-CUDA-12.1.1
- AlphaPulldown/2.0.3-foss-2023a-CUDA-12.1.1
- dm-haiku/0.0.12-foss-2023a-CUDA-12.1.1
- LLAMA-CPP/b3580-foss-2023a-CUDA-12.1.1
- llama_cpp_python/0.2.89-foss-2023a-CUDA-12.1.1
- Uni-Core/0.0.3-foss-2023a-CUDA-12.1.1
- OpenMM/8.0.0-foss-2023a-CUDA-12.1.1
- Optax/0.2.2-gfbf-2023a-CUDA-12.1.1
- dm-haiku/0.0.11-foss-2022b-CUDA-12.2.0
- jax/0.4.19-foss-2022b-CUDA-12.2.0
- GraphCast/0.1-foss-2022b-CUDA-12.2.0
- GraphCast/0.1.1-foss-2022b-CUDA-12.2.0
If you use these in your jobs, please check your job output to make sure the jobs were not affected.
250506: Updated list of modules.
Educloud share (share.educloud.no) is currently not responding. We are working on solving the issue and will update this message when we know more.
We are currently experiencing issues with inviting foreign users from the selfservice portal.
We are investigating the issue and will update this message when it is resolved.
The license for the program F4 transcript is expired.
We are working on a solution, and will update the program as soon as possible.
This message will be updated when the issue is solved.
We are currently experiencing issues with the selfservice portal, selfservice.educloud.no.
We are working on a solution and will update this message when it is resolved.
Update: This issue has been resolved.
We are currently experiencing issues with renewing Nvivo licences in Educloud.
We are working on a solution, and will update the licences as soon as possible.
This message will be updated when the issue is solved.
Update: The maintenance has been completed.
view.educloud.no will experience scheduled downtime on Monday, 24 February, from 07:00 to 09:00. While the service may be intermittently available during this time, you may experience periods of downtime.
Please note that you will be able to reconnect to the same machine you were using before the maintenance once the service is restored.
Galaxy on Fox has been upgraded to version 24.1.
The Galaxy platfom on FOX was down due to some issues in the slurm-drmaa library. They have been now fixed and Galaxy is working properly.
We are going to upgrade Educloud Share on Wednesday, October 9th 2024 at 10:30. The service will be unavailable for about 15-30 minutes.
On 2024-10-09 from 16:00 we'll perform maintenance on the storage system.
During this period the home and project directories on Educloud VDIs will be inaccessible. Processes will hang and continue afterwards or terminate unexpectedly. To be on the safe side, save your work and log off prior to the downtime.
A maintenance reservation for the downtime has been set on Fox starting at 15:00. Jobs that cannot complete before the downtime will not be scheduled until after the downtime. Any running jobs and processes will be killed. (Note that this is one hour before the storage maintenance starts - we will use the opportunity to adjust the Slurm setup instead of having a separate downtime for it.) Users will be kicked out of the login and interactive nodes at 16:00. If you had running jobs or processes at the start of the downtime, please check your job output for errors. To be on the safe side, save your work, cancel running jobs and...
Update, 2024-09-25 09:17: The upgrade has been completed.
The queue system (Slurm) on Fox will be upgraded on Wednesday, September 25 at 09:00.
The upgrade is expected to last only five to ten minutes. Running jobs will be suspended during the upgrade, and the Slurm commands (squeue, sbatch, etc.) will be unavailable during parts of the upgrade. Apart from that, users should not be affected by the upgrade.
The upgrade will bring a few user visible new features, like squeue --notme.
03rd September from 08:00 to 12:00 CET share.educloud.no (Nextcloud) will be down for maintenance as its storage (S3) will be upgraded.
This message will be updated once the downtime is completed.
Issues occurred while listing users in research.educloud.no. We are working to solve the issue promptly