April 30, 2026 11:07 · January 15, 2016 19:38
 From 33f13f5c6cf75f8afa1d6a79983ab96d3eeb419d Mon Sep 17 00:00:00 2001
 From: Chris Fregly <cfregly@coreweave.com>
 Date: Thu, 30 Apr 2026 04:06:52 -0700
 Subject: [PATCH] runner: GAMBLE_SKIP_K8S=1 / --no-k8s-context escape hatch for
 Slurm-side runs (#276)

 gamble/runner.py hardcodes a `kubectl get node` probe in build_node_context().
 On Slurm-side runs (where no Kubernetes apiserver is reachable from the
 compute node), this fails with the canonical:
 # Note – this is not a bash script (some of the steps require reboot)
 # I named it .sh just so Github does correct syntax highlighting.
 #
 # This is also available as an AMI in us-east-1 (virginia): ami-cf5028a5
 #
 # The CUDA part is mostly based on this excellent blog post:
 # http://tleyden.github.io/blog/2014/10/25/cuda-6-dot-5-on-aws-gpu-instance-running-ubuntu-14-dot-04/

 # Install various packages
 sudo apt-get update
	# Note – this is not a bash script (some of the steps require reboot)
	# I named it .sh just so Github does correct syntax highlighting.
	#
	# This is also available as an AMI in us-east-1 (virginia): ami-cf5028a5
	#
	# The CUDA part is mostly based on this excellent blog post:
	# http://tleyden.github.io/blog/2014/10/25/cuda-6-dot-5-on-aws-gpu-instance-running-ubuntu-14-dot-04/

	# Install various packages
	sudo apt-get update