UTHPC - Notice history

All systems operational

99.99% uptime

kubernetes.hpc.ut.ee - Operational

99.98% uptime
Sep 2026 · 99.95%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

galaxy.hpc.ut.ee - Operational

99.96% uptime
Sep 2026 · 99.95%Oct · 99.93%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

registry.hpc.ut.ee - Operational

99.98% uptime
Sep 2026 · 99.95%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

s3.hpc.ut.ee - Operational

100.0% uptime
Sep 2026 · 100.0%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

llm.hpc.ut.ee - Operational

100.0% uptime
Sep 2026 · 100.0%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

export.hpc.ut.ee - Operational

100.0% uptime
Sep 2026 · 100.0%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

sftp.hpc.ut.ee - Operational

100.0% uptime
Sep 2026 · 100.0%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

gitlab.cs.ut.ee - Operational

100.0% uptime
Sep 2026 · 100.0%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026
99.98% uptime

hpc.ut.ee - Operational

99.98% uptime
Sep 2026 · 99.95%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

docs.hpc.ut.ee - Operational

99.98% uptime
Sep 2026 · 99.95%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

support.hpc.ut.ee - Operational

99.98% uptime
Sep 2026 · 99.95%Oct · 100.0%Nov · 100.0%

Previous page

Next page

Sep 2026
Oct 2026
Nov 2026

Notice history

View current status

Nov 2026

No notices reported this month

Oct 2026

MyAccessID Service Is Currently Unavailable
ResolvedMajor outage17 minutes
  • ResolvedUTC

    The MyAccessID service is operational again. Users can now log in to minu.etais.ee and other Waldur portals using MyAccessID.

    Thank you for your patience.

  • InvestigatingUTC

    The MyAccessID service is currently experiencing an outage, preventing users from logging in to minu.etais.ee and other Waldur portals.

    If TARA is already linked to your account, you can use it to log in while MyAccessID is unavailable.

    We are monitoring the situation and will provide an update once the service has been restored.

OCT26
UTHPC LLM Inference service API keys must be re-provisioned on Monday (26.10.2026).
CompletedMaintenance10 minutes
  • Completed 26. October 2026 at 13:10:00UTC
    Maintenance has completed successfully
  • In progress 26. October 2026 at 13:00:01UTC
    Maintenance is now in progress
  • Planned 23. October 2026 at 12:53:13UTC

    All API keys previously requested and created for the LLM Inference API service will stop working on Monday, 26th October, when API key management will be switched to the ETAIS self-service portal.
    Everyone who wishes to continue using the service should provision a new API key for themselves through the self-service environment.

    Documentation on how to use the self-service environment and get new API credentials can be found in the following documentation:
    https://docs.hpc.ut.ee/public/services/LLM_Inference/inference/

    https://docs.hpc.ut.ee/public/ETAIS/adding_llm_inference/

    All current Inference service users have also received personal e-mail notices about the migration.

Security restriction: HPC web cluster access restricted to Estonia (CVE-2026-109618 & CVE-2026-109619)
ResolvedPartial outage6 hours 39 minutes
  • ResolvedUTC
    This incident has been resolved.
  • IdentifiedUTC
    Due to critical security vulnerabilities associated with CVE-2026-109618 and CVE-2026-109619, access to the entire HPC web cluster has been temporarily restricted to Estonian IP addresses. Users connecting from outside Estonia must use the UT VPN to access these services.
GPU nodes on our Rocket cluster are currently unavailable
ResolvedMajor outage1 hour 14 minutes
  • ResolvedUTC

    The emergency maintenance has been completed, and the GPU nodes on the Rocket cluster are available again.

    Affected jobs have been requeued and will run again as resources become available.

    Thank you for your patience and understanding.

  • IdentifiedUTC

    GPU nodes on our Rocket cluster are currently unavailable due to a recently disclosed vulnerability. All ongoing jobs on these nodes will be stopped as the nodes undergo emergency maintenance. Once GPU node availability is restored, the affected jobs will be returned to the job queue.

    We apologize for any inconvenience and will inform you when the nodes are available again.

  • InvestigatingUTC

    GPU nodes on our Rocket cluster are currently unavailable due to a recently disclosed vulnerability. All ongoing jobs on them will be killed as the nodes will undergo emergency maintenance. We apologize for any inconvenience and will inform you when the nodes are available for submission.

Sep 2026

Service disruption: "/gpfs/helios/software/cluster_software/modules/spack/linux-rhel9-x86_64/Core/" module path
ResolvedDegraded performance44 hours 36 minutes
  • ResolvedUTC

    The issue affecting the /gpfs/helios/software/cluster_software/modules/spack/linux-rhel9-x86_64/Core/ module path has been resolved.

    Thank you for your patience.

  • IdentifiedUTC

    The "/gpfs/helios/software/cluster_software/modules/spack/linux-rhel9-x86_64/Core/" module path is currently affected by a technical issue involving a broken zlib dependency in the gcc-runtime module. Around 2000 packages are affected.

    Please do not use this module path until further notice. We will update this page once the issue has been resolved.

Services Currently Unavailable
ResolvedMajor outage23 minutes
  • ResolvedUTC
    This incident has been resolved.
  • IdentifiedUTC

    Our services are beginning to recover. We’re continuing to monitor the situation and restore full functionality as quickly as possible.

  • InvestigatingUTC

    Our infrastructure is currently experiencing an outage. We are actively investigating the issue and working to restore service as quickly as possible. We apologize for the disruption and will share updates as they become available.

SEP4
Due to Slurm upgrade there might be short interruptions running jobs on Rocket cluster
CompletedMaintenance1 hour 47 minutes
  • Completed 04. September 2026 at 15:47:06UTC

    Slurm upgrade has been completed successfully.

  • In progress 04. September 2026 at 14:00:00UTC

    Due to an upcoming SLURM update, running jobs on the cluster may be interrupted in the next few hours. Job submissions may fail or hang. We apologize for any inconvenience.

Previous

Sep 2026 to Nov 2026

Next