Active Incident

Updated a few seconds ago

Incident Status

Degraded Performance

Components

Primary Cloud

Locations

Primary Data Center



April 1, 2025 11:39AM EDT
April 1, 2025 3:39PM UTC
[Investigating] Jetstream2 is currently experiencing extremely limited availability of g3.xl instances due to scarce supply; usage of g3.xl instances has rapidly approached our maximum capacity. As placements are regularly exhausted, we expect that creating new g3.xl instances, unshelving existing ones, and resizing instances to the g3.xl flavor may result in errors. Jetstream2 engineers are currently working toward a solution, but at this time, we recommend using other flavors (e.g. g3.medium or g3.large) where possible. The dashboard at https://docs.jetstream-cloud.org/overview/status/#availability-of-scarce-resources shows a live-updating count of vacancy for scarce resources such as g3.xl instances. We thank you for your patience and support while we address this issue. Jetstream2 Support can be reached via help@jetstream-cloud.org for any questions, comments, or concerns.
Jetstream2 Research and Education Cloud JS2 Docs are here: https://docs.jetstream-cloud.org/

Primary Cloud

Operational

Exosphere User Interface

Operational

ASU Regional Cloud

Operational

Cornell Regional Cloud

Operational

Hawaii Regional Cloud

Operational

TACC Regional Cloud

Operational

Docs Site

Operational

Website

Operational

Jetstream2 Support

Operational

0

Upcoming Maintenances

2

Incidents Last 30 Days

0

Maintenances Last 30 Days

External Services

History (Last 7 days)

Incident Status

Partial Service Disruption


Components

Primary Cloud


Locations

Primary Data Center




April 13, 2025 10:47PM EDT
April 14, 2025 2:47AM UTC
[Resolved] All vGPU instances came back online several hours ago save seven that were in ERROR status. Engineers have been monitoring the situation and feel confident clearing this issue at this time. We will continue to work to resolve those remaining seven instances but cloud operations are normal at this time and have been for some time.

April 13, 2025 5:30PM EDT
April 13, 2025 9:30PM UTC
[Monitoring] Large Memory nodes and g3.xl nodes are back online. vGPU nodes are in process of returning to service.

April 13, 2025 12:38PM EDT
April 13, 2025 4:38PM UTC
[Identified] There was a power event at the IU Data Center at approximately 12:24pm EDT. The power event caused a brief loss of power and all nodes not on battery backup (GPU and Large Memory nodes) went offline. Engineers are working to bring those nodes back up now. There is no ETA at this time but we hope to have them up as quickly as possible.