Linux CUDA builds are failing to configure the system #74968
Description
Current Status
ongoing.
Error looks like
All Xenial based builds are failing with:
Failed to fetch https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64/Packages 404 Not Found
Incident timeline (all times pacific)
Include when the incident began, when it was detected, mitigated, root caused, and finally closed.
Click for example
e.g.
- 10/30 7:27a incident began
- 10/30 8:30a detected by
- 10/30 9:00 pm root caused as…
- 10/30 9:10 pm mitigated by…
- 10/31 10: am closed by…
User impact
How does this affect users of PyTorch CI?
Root cause
What was the root cause of this issue?
Mitigation
How did we mitigate the issue?
Prevention/followups
How do we prevent issues like this in the future?