Skip to content

[CP 1337] [Fix] Add workflow and workflow-triggered pod collection to techsupport #522

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
ci-penbot-01 wants to merge 1 commit into
base: main
Choose a base branch
from
+62 −2
Open

[CP 1337] [Fix] Add workflow and workflow-triggered pod collection to techsupport #522

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
64 changes: 62 additions & 2 deletions tools/techsupport_dump.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ NFD_RESOURCES="pods daemonsets deployments configmap"
KMM_RESOURCES="pods daemonsets deployments modules configmap"
GPUOPER_RESOURCES="pods daemonsets deployments deviceconfig configmap"
NPD_RESOURCES="pods daemonsets configmap"
WORKFLOW_RESOURCES="workflows"

OUTPUT_FORMAT="json"
WIDE=""
Expand Down Expand Up @@ -66,8 +67,8 @@ pod_logs() {
for lpod in ${PODS}; do
pod=$(basename ${lpod})
log " ${NS}/${pod}"
${KNS} logs "${pod}" --all-containers >${TECH_SUPPORT_FILE}/${NODE}/${FEATURE}/${NS}_${pod}.txt
${KNS} logs -p "${pod}" --all-containers --tail 1 >/dev/null 2>&1 && ${KNS} logs -p "${pod}" >${TECH_SUPPORT_FILE}/${NODE}/${FEATURE}/${NS}_${pod}_previous.txt
${KNS} logs "${pod}" --all-containers >${TECH_SUPPORT_FILE}/${NODE}/${FEATURE}/${NS}_${pod}.txt 2>&1 || true
${KNS} logs -p "${pod}" --all-containers --tail 1 >/dev/null 2>&1 && ${KNS} logs -p "${pod}" >${TECH_SUPPORT_FILE}/${NODE}/${FEATURE}/${NS}_${pod}_previous.txt 2>&1 || true
done
echo "${PODS}" >${TECH_SUPPORT_FILE}/${node}/${FEATURE}/pods.txt
}
Expand Down Expand Up @@ -250,6 +251,49 @@ if [ -n "${NPD_NS}" ]; then
done
fi

# workflow resources (auto node remediation feature)
log "workflows:"
for resource in ${WORKFLOW_RESOURCES}; do
log " ${GPUOPER_NS}/${resource}"
mkdir -p ${TECH_SUPPORT_FILE}/workflow/
${KUBECTL} get -n ${GPUOPER_NS} ${resource} ${WIDE} >${TECH_SUPPORT_FILE}/workflow/${resource}.txt 2>&1 || true
${KUBECTL} describe -n ${GPUOPER_NS} ${resource} >>${TECH_SUPPORT_FILE}/workflow/${resource}.txt 2>&1 || true
${KUBECTL} get -n ${GPUOPER_NS} ${resource} -o ${OUTPUT_FORMAT} >${TECH_SUPPORT_FILE}/workflow/${resource}.${OUTPUT_FORMAT} 2>&1 || true
done

# workflow controller pod
log "workflow controller:"
mkdir -p ${TECH_SUPPORT_FILE}/workflow/
WORKFLOW_CONTROLLER_PODS=$(${KUBECTL} get pods -n ${GPUOPER_NS} -l app=amd-gpu-operator-workflow-controller --no-headers -o custom-columns=NAME:.metadata.name 2>/dev/null || true)
if [ -n "$WORKFLOW_CONTROLLER_PODS" ]; then
echo "" >${TECH_SUPPORT_FILE}/workflow/workflow-controller-pods.txt
echo "" >${TECH_SUPPORT_FILE}/workflow/workflow-controller-pods.${OUTPUT_FORMAT}
while IFS= read -r pod_name; do
[ -z "$pod_name" ] && continue
log " ${GPUOPER_NS}/${pod_name}"
${KUBECTL} describe -n ${GPUOPER_NS} pod "$pod_name" >>${TECH_SUPPORT_FILE}/workflow/workflow-controller-pods.txt 2>&1 || true
${KUBECTL} get -n ${GPUOPER_NS} pod "$pod_name" -o ${OUTPUT_FORMAT} >>${TECH_SUPPORT_FILE}/workflow/workflow-controller-pods.${OUTPUT_FORMAT} 2>&1 || true
done <<< "$WORKFLOW_CONTROLLER_PODS"
fi

# workflow-triggered test runner pods (pods with workflow labels)
log "workflow-triggered pods:"
mkdir -p ${TECH_SUPPORT_FILE}/workflow/
WORKFLOW_PODS=$(${KUBECTL} get pods -n ${GPUOPER_NS} -l workflows.argoproj.io/workflow --no-headers -o custom-columns=NAME:.metadata.name 2>/dev/null || true)
if [ -n "$WORKFLOW_PODS" ]; then
echo "" >${TECH_SUPPORT_FILE}/workflow/workflow-pods.txt
echo "" >${TECH_SUPPORT_FILE}/workflow/workflow-pods.${OUTPUT_FORMAT}
POD_COUNT=$(echo "$WORKFLOW_PODS" | wc -l)
POD_NUM=0
while IFS= read -r pod_name; do
[ -z "$pod_name" ] && continue
POD_NUM=$((POD_NUM + 1))
log " ($POD_NUM/$POD_COUNT) ${GPUOPER_NS}/${pod_name}"
${KUBECTL} describe -n ${GPUOPER_NS} pod "$pod_name" >>${TECH_SUPPORT_FILE}/workflow/workflow-pods.txt 2>&1 || true
${KUBECTL} get -n ${GPUOPER_NS} pod "$pod_name" -o ${OUTPUT_FORMAT} >>${TECH_SUPPORT_FILE}/workflow/workflow-pods.${OUTPUT_FORMAT} 2>&1 || true
done <<< "$WORKFLOW_PODS"
fi

CONTROL_PLANE=$(${KUBECTL} get nodes -l node-role.kubernetes.io/control-plane | grep -w Ready | awk '{print $1}')
# logs
if [ "${NODES}" == "all" ]; then
Expand Down Expand Up @@ -424,6 +468,22 @@ for node in "${nodeList[@]}"; do
fi
fi

# workflow controller pod logs
WORKFLOW_CONTROLLER_PODS=$(${KNS} get pods -o name --field-selector spec.nodeName=${node} -l app=amd-gpu-operator-workflow-controller 2>/dev/null || true)
if [ -n "$WORKFLOW_CONTROLLER_PODS" ]; then
if ! pod_logs $GPUOPER_NS "workflow-controller" $node "$WORKFLOW_CONTROLLER_PODS"; then
log "Failed to collect logs for workflow controller on node ${node}"
fi
fi

# workflow-triggered pod logs (pods with workflow labels on this node)
WORKFLOW_NODE_PODS=$(${KNS} get pods -o name --field-selector spec.nodeName=${node} -l workflows.argoproj.io/workflow 2>/dev/null || true)
if [ -n "$WORKFLOW_NODE_PODS" ]; then
if ! pod_logs $GPUOPER_NS "workflow-pods" $node "$WORKFLOW_NODE_PODS"; then
log "Failed to collect logs for workflow-triggered pods on node ${node}"
fi
fi

# operator pod logs
GPUOPER_PODS=$(${KNS} get pods -o name --field-selector spec.nodeName=${node} -l "app.kubernetes.io/name=gpu-operator-charts" 2>/dev/null || true)
if [ -z "$GPUOPER_PODS" ]; then
Expand Down