databricks-debug-bundle（databricks 调试包）

Databricks Debug Bundle

Current State

!databricks --version 2>/dev/null || echo 'CLI not installed'!python3 -c "import databricks.sdk; print(f'SDK {databricks.sdk.__version__}')" 2>/dev/null || echo 'SDK not installed'

Overview

Collect all diagnostic information needed for Databricks support tickets: environment info, cluster state, cluster events, job run details, Spark driver logs, and Delta table history. Produces a redacted tar.gz bundle safe to share with support.

Prerequisites

• Databricks CLI installed and configured
• Access to cluster logs (admin or cluster owner)
• Permission to access job run details

Instructions

Step 1: Create Debug Collection Script

#!/bin/bash
set -euo pipefail
# databricks-debug-bundle.sh [cluster_id] [run_id] [table_name]

BUNDLE_DIR="databricks-debug-$(date +%Y%m%d-%H%M%S)"
mkdir -p "$BUNDLE_DIR"

CLUSTER_ID="${1:-}"
RUN_ID="${2:-}"
TABLE_NAME="${3:-}"

echo "=== Databricks Debug Bundle ===" | tee "$BUNDLE_DIR/summary.txt"
echo "Generated: $(date -u +%Y-%m-%dT%H:%M:%SZ)" >> "$BUNDLE_DIR/summary.txt"
echo "Workspace: ${DATABRICKS_HOST:-unset}" >> "$BUNDLE_DIR/summary.txt"

Step 2: Collect Environment Info

{
    echo ""
    echo "--- Environment ---"
    echo "CLI: $(databricks --version 2>&1)"
    echo "SDK: $(pip show databricks-sdk 2>/dev/null | grep Version || echo 'not installed')"
    echo "Python: $(python3 --version 2>&1)"
    echo "OS: $(uname -srm)"
    echo ""
    echo "--- Current User ---"
    databricks current-user me --output json 2>&1 | jq '{userName, active}' || echo "Auth failed"
} >> "$BUNDLE_DIR/summary.txt"

Step 3: Collect Cluster Information

if [ -n "$CLUSTER_ID" ]; then
    echo "" >> "$BUNDLE_DIR/summary.txt"
    echo "--- Cluster: $CLUSTER_ID ---" >> "$BUNDLE_DIR/summary.txt"

    # Full cluster config
    databricks clusters get --cluster-id "$CLUSTER_ID" --output json \
        > "$BUNDLE_DIR/cluster_config.json" 2>&1

    # Key fields summary
    jq '{state, spark_version, node_type_id, num_workers,
         autotermination_minutes, termination_reason}' \
        "$BUNDLE_DIR/cluster_config.json" >> "$BUNDLE_DIR/summary.txt"

    # Recent cluster events (state changes, errors, resizing)
    databricks clusters events --cluster-id "$CLUSTER_ID" --limit 30 --output json \
        > "$BUNDLE_DIR/cluster_events.json" 2>&1

    # Extract event timeline
    jq -r '.events[]? | "\(.timestamp): \(.type) — \(.details // "no details")"' \
        "$BUNDLE_DIR/cluster_events.json" >> "$BUNDLE_DIR/summary.txt" 2>/dev/null
fi

Step 4: Collect Job Run Information

if [ -n "$RUN_ID" ]; then
    echo "" >> "$BUNDLE_DIR/summary.txt"
    echo "--- Run: $RUN_ID ---" >> "$BUNDLE_DIR/summary.txt"

    # Full run details
    databricks runs get --run-id "$RUN_ID" --output json \
        > "$BUNDLE_DIR/run_details.json" 2>&1

    # Run state summary
    jq '{state: .state, start_time, end_time, run_duration}' \
        "$BUNDLE_DIR/run_details.json" >> "$BUNDLE_DIR/summary.txt"

    # Task-level breakdown
    jq -r '.tasks[]? | "  Task \(.task_key): \(.state.result_state // "RUNNING") — \(.state.state_message // "ok")"' \
        "$BUNDLE_DIR/run_details.json" >> "$BUNDLE_DIR/summary.txt"

    # Run output (error messages, stdout)
    databricks runs get-output --run-id "$RUN_ID" --output json \
        > "$BUNDLE_DIR/run_output.json" 2>&1

    jq '{error, error_trace: (.error_trace // "" | .[0:2000])}' \
        "$BUNDLE_DIR/run_output.json" >> "$BUNDLE_DIR/summary.txt" 2>/dev/null
fi

Step 5: Collect Spark Driver Logs

if [ -n "$CLUSTER_ID" ]; then
    echo "" >> "$BUNDLE_DIR/summary.txt"
    echo "--- Spark Driver Logs (last 500 lines) ---" >> "$BUNDLE_DIR/summary.txt"

    python3 << 'PYEOF' > "$BUNDLE_DIR/driver_logs.txt" 2>&1
from databricks.sdk import WorkspaceClient
w = WorkspaceClient()
try:
    content = w.dbfs.read("/cluster-logs/${CLUSTER_ID}/driver/log4j-active.log")
    # Take last 500 lines
    lines = content.data.decode().splitlines()[-500:]
    print("\n".join(lines))
except Exception as e:
    print(f"Could not fetch driver logs: {e}")
    print("Tip: Enable cluster log delivery in cluster config for persistent logs")
PYEOF
fi

Step 6: Collect Delta Table Diagnostics

if [ -n "$TABLE_NAME" ]; then
    echo "" >> "$BUNDLE_DIR/summary.txt"
    echo "--- Delta Table: $TABLE_NAME ---" >> "$BUNDLE_DIR/summary.txt"

    python3 << PYEOF > "$BUNDLE_DIR/delta_diagnostics.txt" 2>&1
from databricks.connect import DatabricksSession
spark = DatabricksSession.builder.getOrCreate()

print("=== Table Details ===")
spark.sql("DESCRIBE DETAIL ${TABLE_NAME}").show(truncate=False)

print("\n=== Recent History (last 20 operations) ===")
spark.sql("DESCRIBE HISTORY ${TABLE_NAME} LIMIT 20").show(truncate=False)

print("\n=== Schema ===")
spark.sql("DESCRIBE ${TABLE_NAME}").show(truncate=False)

print("\n=== File Stats ===")
detail = spark.sql("DESCRIBE DETAIL ${TABLE_NAME}").first()
print(f"Files: {detail.numFiles}, Size: {detail.sizeInBytes / 1024 / 1024:.1f} MB")
PYEOF
fi

Step 7: Package Bundle (Redacted)

# Redact sensitive data from config snapshot
echo "" >> "$BUNDLE_DIR/summary.txt"
echo "--- Config (redacted) ---" >> "$BUNDLE_DIR/summary.txt"
if [ -f ~/.databrickscfg ]; then
    sed 's/token = .*/token = ***REDACTED***/' \
        ~/.databrickscfg > "$BUNDLE_DIR/config-redacted.txt"
    sed -i 's/client_secret = .*/client_secret = ***REDACTED***/' \
        "$BUNDLE_DIR/config-redacted.txt"
fi

# Network connectivity test
echo "--- Network ---" >> "$BUNDLE_DIR/summary.txt"
echo -n "API reachable: " >> "$BUNDLE_DIR/summary.txt"
curl -s -o /dev/null -w "%{http_code}" \
    "${DATABRICKS_HOST}/api/2.0/clusters/list" \
    -H "Authorization: Bearer ${DATABRICKS_TOKEN}" >> "$BUNDLE_DIR/summary.txt"
echo "" >> "$BUNDLE_DIR/summary.txt"

# Create archive
tar -czf "$BUNDLE_DIR.tar.gz" "$BUNDLE_DIR"
rm -rf "$BUNDLE_DIR"

echo ""
echo "Bundle created: $BUNDLE_DIR.tar.gz"
echo "Contents: summary.txt, cluster_config.json, cluster_events.json,"
echo "  run_details.json, run_output.json, driver_logs.txt,"
echo "  delta_diagnostics.txt, config-redacted.txt"

Output

• databricks-debug-YYYYMMDD-HHMMSS.tar.gz containing:

- summary.txt — Human-readable diagnostic summary - cluster_config.json — Full cluster configuration - cluster_events.json — State changes, errors, resizing events - run_details.json — Job run with task-level breakdown - run_output.json — Stdout/stderr and error traces - driver_logs.txt — Last 500 lines of Spark driver log - delta_diagnostics.txt — Table details, history, schema - config-redacted.txt — CLI config with secrets removed

Error Handling

Examples

Usage

# Environment only
bash databricks-debug-bundle.sh

# With cluster diagnostics
bash databricks-debug-bundle.sh 0123-456789-abcde

# With cluster + job run
bash databricks-debug-bundle.sh 0123-456789-abcde 12345

# Full diagnostics including Delta table
bash databricks-debug-bundle.sh 0123-456789-abcde 12345 catalog.schema.table

Submit to Support

1. Generate bundle: bash databricks-debug-bundle.sh [args]
2. Review summary.txt for sensitive data
3. Open ticket at help.databricks.com
4. Attach the .tar.gz bundle
5. Include workspace ID (found in workspace URL: adb-<workspace-id>)

Resources

• Databricks Support
• Status Page
• Community Forum

Next Steps

For rate limit issues, see databricks-rate-limits.