# Astronomer: KPO podspec changes for GPU-enabled nodes may cause KPO pod startup failures

- Impact (vendor’s label): minor
- Status: resolved
- Started: 2026-08-04 15:11 UTC · Resolved: 2026-08-04 16:40 UTC · Duration: 1.5 h
- Updates: 3 · Affected components: Scheduling and Running DAGs and Tasks
- [Vendor status page: https://status.astronomer.io/incidents/01KZ6N8GP1FVGTB8WPZ7WT98PR](https://status.astronomer.io/incidents/01KZ6N8GP1FVGTB8WPZ7WT98PR)
- [Company: Astronomer](https://rewindex.com/company/astronomer.md)

## Timeline of updates

- **identified** 2026-08-04 15:11 UTC: We've identified a change that enabled GPU support for KubernetesPodOperator pods had an unintended side effect when reading pod specs without proper limits set in the expected format. We're working on a fix, if you notice any unexpected failures for KPO tasks in your Astro deployments, please open a support ticket.
- **monitoring** 2026-08-04 15:15 UTC: We have rolled back the changes that triggered this issue and are monitoring the issue for confirmation that this issue is resolved.
- **resolved** 2026-08-04 16:40 UTC: Mitigations have been successful. A Platform release that corrects this errant behavior is being worked on now. Issue is resolved.


---
Canonical: https://rewindex.com/company/astronomer/incidents/01KZ6N8GP1FVGTB8WPZ7WT98PR · Markdown: https://rewindex.com/company/astronomer/incidents/01KZ6N8GP1FVGTB8WPZ7WT98PR.md · JSON: https://rewindex.com/api/v1/companies/astronomer/incidents/01KZ6N8GP1FVGTB8WPZ7WT98PR · Generated 2026-10-18T09:55:11.406Z