diff --git a/log-collector-script/linux/eks-log-collector.sh b/log-collector-script/linux/eks-log-collector.sh index 7453b05a2..4b74232f2 100755 --- a/log-collector-script/linux/eks-log-collector.sh +++ b/log-collector-script/linux/eks-log-collector.sh @@ -63,6 +63,7 @@ COMMON_DIRECTORIES=( kubelet # eks nodeadm # eks cni # eks + gpu # eks ) COMMON_LOGS=( @@ -287,6 +288,7 @@ collect() { get_sandboxImage_info get_cpu_throttled_processes get_io_throttled_processes + get_nvidia_bug_report } pack() { @@ -796,6 +798,16 @@ get_io_throttled_processes() { ok } +get_nvidia_bug_report() { + try "Collect Nvidia Bug report" + if ! command -v nvidia-bug-report.sh &> /dev/null; then + echo "No Nvidia drivers found, nothing to do." + else + timeout 75 nvidia-bug-report.sh --output-file "${COLLECT_DIR}"/gpu/nvidia-bug-report.log &> /dev/null + fi + ok +} + # ----------------------------------------------------------------------------- # Entrypoint parse_options "$@"