Sentry Debug Bundle Skill

Sentry Debug Bundle

Overview

Collect SDK versions, configuration state, network connectivity, and event delivery status into a single diagnostic report. Attach the output to Sentry support tickets or use it to systematically isolate why events are not reaching the dashboard.

Current State

!node --version 2>/dev/null || echo 'Node.js not found' !python3 --version 2>/dev/null || echo 'Python3 not found' !npm list @sentry/node @sentry/browser @sentry/react @sentry/cli 2>/dev/null | grep sentry || pip show sentry-sdk 2>/dev/null | grep -E '^(Name|Version)' || echo 'No Sentry SDK found' !sentry-cli --version 2>/dev/null || echo 'sentry-cli not installed' !sentry-cli info 2>/dev/null || echo 'sentry-cli not authenticated'

Prerequisites

At least one Sentry SDK installed (@sentry/node, @sentry/browser, @sentry/react, or sentry-sdk for Python)
SENTRY_DSN environment variable set (or DSN configured in application code)
For API checks: SENTRY_AUTH_TOKEN with project:read scope (generate token)
Optional: sentry-cli installed for source map diagnostics and send-event tests

Instructions

Step 1 — Gather SDK Version, Configuration, and Init Hooks

Identify the installed SDK, verify all @sentry/* packages share the same version (mismatches cause silent failures), and extract the runtime configuration.

Check installed packages:

# Node.js — list all Sentry packages and flag version mismatches
npm ls @sentry/node @sentry/browser @sentry/react @sentry/nextjs @sentry/cli 2>/dev/null | grep sentry

# Python — show sentry-sdk version and installed extras
pip show sentry-sdk 2>/dev/null

Extract runtime configuration (Node.js):

import * as Sentry from '@sentry/node';

const client = Sentry.getClient();
if (!client) {
  console.error('ERROR: Sentry client not initialized — Sentry.init() may not have been called');
  process.exit(1);
}

const opts = client.getOptions();
const diagnostics = {
  sdk_version: Sentry.SDK_VERSION,
  dsn_configured: !!opts.dsn,
  dsn_host: opts.dsn ? new URL(opts.dsn).hostname : 'N/A',
  dsn_project_id: opts.dsn ? new URL(opts.dsn).pathname.replace('/', '') : 'N/A',
  environment: opts.environment ?? '(default)',
  release: opts.release ?? '(auto-detect)',
  debug: opts.debug ?? false,
  sample_rate: opts.sampleRate ?? 1.0,
  traces_sample_rate: opts.tracesSampleRate ?? '(not set)',
  profiles_sample_rate: opts.profilesSampleRate ?? '(not set)',
  send_default_pii: opts.sendDefaultPii ?? false,
  max_breadcrumbs: opts.maxBreadcrumbs ?? 100,
  before_send: typeof opts.beforeSend === 'function' ? 'CONFIGURED' : 'none',
  before_send_transaction: typeof opts.beforeSendTransaction === 'function' ? 'CONFIGURED' : 'none',
  before_breadcrumb: typeof opts.beforeBreadcrumb === 'function' ? 'CONFIGURED' : 'none',
  integrations: client.getOptions().integrations?.map(i => i.name) ?? [],
  transport: opts.transport ? 'custom' : 'default',
};

console.log(JSON.stringify(diagnostics, null, 2));

Extract runtime configuration (Python):

import sentry_sdk
from sentry_sdk import Hub

client = Hub.current.client
if not client:
    print("ERROR: Sentry client not initialized")
    exit(1)

opts = client.options
print(f"SDK version:       {sentry_sdk.VERSION}")
print(f"DSN configured:    {bool(opts.get('dsn'))}")
print(f"Environment:       {opts.get('environment', '(default)')}")
print(f"Release:           {opts.get('release', '(auto-detect)')}")
print(f"Debug:             {opts.get('debug', False)}")
print(f"Sample rate:       {opts.get('sample_rate', 1.0)}")
print(f"Traces sample rate:{opts.get('traces_sample_rate', '(not set)')}")
print(f"Send default PII:  {opts.get('send_default_pii', False)}")
print(f"before_send:       {'CONFIGURED' if opts.get('before_send') else 'none'}")
print(f"before_breadcrumb: {'CONFIGURED' if opts.get('before_breadcrumb') else 'none'}")
print(f"Integrations:      {[i.identifier for i in client.integrations.values()]}")

Key check: If beforeSend is CONFIGURED, inspect the function — a beforeSend that returns null will silently drop every event.

Step 2 — Verify DSN Connectivity and Sentry Service Status

Confirm the application can reach Sentry's ingest endpoint and that Sentry itself is operational.

Test DSN reachability:

# Test the Sentry API root (should return HTTP 200)
curl -s -o /dev/null -w "sentry.io API: HTTP %{http_code} (%{time_total}s)\n" \
  https://sentry.io/api/0/

# Test the ingest endpoint derived from your DSN
# DSN format: https://<PUBLIC_KEY>@o<ORG_ID>.ingest.sentry.io/<PROJECT_ID>
# Extract host from DSN and test the envelope endpoint
curl -s -o /dev/null -w "Ingest endpoint: HTTP %{http_code} (%{time_total}s)\n" \
  "https://o0.ingest.sentry.io/api/0/envelope/"

# DNS resolution check
dig +short o0.ingest.sentry.io 2>/dev/null || nslookup o0.ingest.sentry.io

# Check for proxy/firewall interference
curl -v https://sentry.io/api/0/ 2>&1 | grep -iE 'proxy|blocked|forbidden|connect'

Check Sentry service status:

Visit https://status.sentry.io or:

# Programmatic status check
curl -s https://status.sentry.io/api/v2/status.json | python3 -c "
import sys, json
d = json.load(sys.stdin)
print(f\"Sentry Status: {d['status']['description']}\")
print(f\"Updated:       {d['page']['updated_at']}\")
"

Verify auth token and project access (requires SENTRY_AUTH_TOKEN):

# Check token validity and list accessible projects
curl -s -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  https://sentry.io/api/0/projects/ | python3 -c "
import sys, json
projects = json.load(sys.stdin)
if isinstance(projects, dict) and 'detail' in projects:
    print(f'Auth error: {projects[\"detail\"]}')
else:
    for p in projects[:10]:
        print(f\"  {p['organization']['slug']}/{p['slug']} (platform: {p.get('platform', 'N/A')})\")"

# sentry-cli auth check
sentry-cli info 2>/dev/null || echo "sentry-cli not authenticated — run: sentry-cli login"

Step 3 — Test Event Capture, Verify Delivery, and Generate Report

Send a diagnostic test event, confirm it arrives in Sentry, and produce the final debug bundle report.

Send test event via sentry-cli:

# Quick test — sends a test message event
sentry-cli send-event -m "diagnostic test from debug-bundle $(date -u +%Y-%m-%dT%H:%M:%SZ)"

Send test event programmatically (Node.js):

import * as Sentry from '@sentry/node';

Sentry.init({
  dsn: process.env.SENTRY_DSN,
  debug: true, // Enables console logging of SDK internals
});

const eventId = Sentry.captureMessage('sentry-debug-bundle diagnostic test', 'info');
console.log(`Test event ID: ${eventId}`);
console.log(`View at: https://sentry.io/organizations/YOUR_ORG/issues/?query=${eventId}`);

// CRITICAL: Node.js will exit before async transport completes without flush
const flushed = await Sentry.flush(10000);
console.log(`Flush: ${flushed ? 'SUCCESS — event delivered' : 'TIMEOUT — event may not have been sent'}`);

if (!flushed) {
  console.error('Flush timeout. Possible causes:');
  console.error('  - Network blocking outbound HTTPS to sentry.io');
  console.error('  - DSN is invalid or project has been deleted');
  console.error('  - Rate limit exceeded (HTTP 429)');
}

Send test event programmatically (Python):

import sentry_sdk
import time

sentry_sdk.init(dsn="YOUR_DSN", debug=True)

event_id = sentry_sdk.capture_message("sentry-debug-bundle diagnostic test", level="info")
print(f"Test event ID: {event_id}")

# Python SDK flushes automatically on exit, but explicit flush is safer
sentry_sdk.flush(timeout=10)
print("Flush complete — check Sentry dashboard for event")

Generate the debug bundle report:

#!/bin/bash
set -euo pipefail

REPORT="sentry-debug-$(date +%Y%m%d-%H%M%S).md"

cat > "$REPORT" << 'HEADER'
# Sentry Debug Bundle
HEADER

cat >> "$REPORT" << EOF
**Generated:** $(date -u +"%Y-%m-%dT%H:%M:%SZ")
**Node.js:** $(node --version 2>/dev/null || echo N/A)
**Python:** $(python3 --version 2>/dev/null || echo N/A)
**OS:** $(uname -srm)
**sentry-cli:** $(sentry-cli --version 2>/dev/null || echo 'not installed')

## SDK Packages
\`\`\`
$(npm list 2>/dev/null | grep -i sentry || echo "No npm Sentry packages")
$(pip show sentry-sdk 2>/dev/null | grep -E '^(Name|Version|Location)' || echo "No pip sentry-sdk")
\`\`\`

## Environment Variables (sanitized)
| Variable | Status |
|----------|--------|
| SENTRY_DSN | $([ -n "${SENTRY_DSN:-}" ] && echo "SET (\`$(echo "$SENTRY_DSN" | sed 's|//[^@]*@|//***@|')\`)" || echo "NOT SET") |
| SENTRY_ORG | ${SENTRY_ORG:-NOT SET} |
| SENTRY_PROJECT | ${SENTRY_PROJECT:-NOT SET} |
| SENTRY_AUTH_TOKEN | $([ -n "${SENTRY_AUTH_TOKEN:-}" ] && echo "SET (${#SENTRY_AUTH_TOKEN} chars)" || echo "NOT SET") |
| SENTRY_RELEASE | ${SENTRY_RELEASE:-NOT SET} |
| SENTRY_ENVIRONMENT | ${SENTRY_ENVIRONMENT:-NOT SET} |
| NODE_ENV | ${NODE_ENV:-NOT SET} |

## Network Connectivity
$(curl -s -o /dev/null -w "- sentry.io API: HTTP %{http_code} (%{time_total}s)" https://sentry.io/api/0/ 2>/dev/null || echo "- sentry.io: UNREACHABLE")
$(curl -s -o /dev/null -w "\n- Ingest endpoint: HTTP %{http_code} (%{time_total}s)" https://o0.ingest.sentry.io/api/0/envelope/ 2>/dev/null || echo "- Ingest: UNREACHABLE")

## Sentry Status
$(curl -s https://status.sentry.io/api/v2/status.json 2>/dev/null | python3 -c "import sys,json; d=json.load(sys.stdin); print(f\"- Status: {d['status']['description']}\")" 2>/dev/null || echo "- Status: Could not fetch")

## CLI Authentication Status
\`\`\`
$(sentry-cli info 2>/dev/null || echo "sentry-cli not authenticated — run: sentry-cli login")
\`\`\`

## Source Map Artifacts
\`\`\`
$(sentry-cli releases files "${SENTRY_RELEASE:-unknown}" list 2>/dev/null || echo "No release artifacts found (set SENTRY_RELEASE)")
\`\`\`
EOF

echo "Debug bundle saved to: $REPORT"
echo "Attach this file to your Sentry support ticket at https://sentry.io/support/"

Output

The debug bundle produces:

SDK inventory — all @sentry/* or sentry-sdk versions with mismatch detection
Configuration snapshot — DSN (redacted), environment, release, sample rates, hooks status
Connectivity report — HTTP status and latency to sentry.io API and ingest endpoints
Service status — current Sentry platform health from status.sentry.io
Test event confirmation — event ID with flush result (SUCCESS or TIMEOUT)
Markdown report file — sentry-debug-YYYYMMDD-HHMMSS.md ready to attach to support tickets

Error Handling

| Error | Cause | Solution | |-------|-------|----------| | Sentry client not initialized | Sentry.init() not called or called after diagnostic code | Ensure Sentry.init() runs at application entry point before any capture calls | | Flush: TIMEOUT | Network blocking outbound HTTPS to *.ingest.sentry.io | Check firewall rules, corporate proxy, VPN — allow outbound 443 to *.ingest.sentry.io | | sentry-cli info reports no auth | Token not set or expired | Run sentry-cli login or set SENTRY_AUTH_TOKEN (generate) | | Package version mismatch (@sentry/* at different versions) | Partial upgrade or transitive dependency conflict | Align all @sentry/* packages to the same version: npm i @sentry/node@latest @sentry/browser@latest and run npm dedupe | | sourcemaps explain shows no match | --url-prefix does not match the stack trace URLs | Compare the abs_path in your error event's stack frame with the --url-prefix used during upload | | Events dropped by beforeSend | Hook returns null for matching events | Log inside beforeSend to verify: console.log('beforeSend:', event.exception?.values?.[0]?.type) | | DSN returns HTTP 401 / 403 | Project deleted, DSN revoked, or wrong organization | Verify DSN at Settings > Projects > Client Keys in the Sentry dashboard | | 429 Too Many Requests | Organization or project rate limit exceeded | Check quotas at Settings > Subscription, reduce sampleRate/tracesSampleRate, or use beforeSend to filter noise | | Process exits before events sent (Node.js) | No await Sentry.flush() before process.exit() | Always call await Sentry.flush(timeout) before exiting; in serverless, use the platform's Sentry.wrapHandler() |

Examples

TypeScript — Full Diagnostic Script

// scripts/sentry-diagnostics.ts
import * as Sentry from '@sentry/node';

async function runDiagnostics() {
  // 1. Check SDK is initialized
  const client = Sentry.getClient();
  if (!client) {
    console.error('[FAIL] Sentry client not initialized');
    console.error('       Ensure Sentry.init({ dsn: "..." }) is called before this script');
    return;
  }
  console.log(`[OK]   SDK version: ${Sentry.SDK_VERSION}`);

  // 2. Check configuration
  const opts = client.getOptions();
  console.log(`[OK]   DSN configured: ${!!opts.dsn}`);
  console.log(`[INFO] Environment: ${opts.environment ?? '(default)'}`);
  console.log(`[INFO] Release: ${opts.release ?? '(auto-detect)'}`);
  console.log(`[INFO] Debug mode: ${opts.debug ?? false}`);
  console.log(`[INFO] Sample rate: ${opts.sampleRate ?? 1.0}`);
  console.log(`[INFO] Traces sample rate: ${opts.tracesSampleRate ?? '(not set)'}`);
  console.log(`[INFO] beforeSend: ${typeof opts.beforeSend === 'function' ? 'CONFIGURED' : 'none'}`);

  // 3. List active integrations
  const integrations = opts.integrations?.map(i => i.name) ?? [];
  console.log(`[INFO] Integrations (${integrations.length}): ${integrations.join(', ')}`);

  // 4. Send test event
  const eventId = Sentry.captureMessage('sentry-diagnostics test event', 'info');
  console.log(`[INFO] Test event ID: ${eventId}`);

  // 5. Flush and verify
  const flushed = await Sentry.flush(10000);
  if (flushed) {
    console.log('[OK]   Flush succeeded — event delivered to Sentry');
  } else {
    console.error('[FAIL] Flush timed out — check network connectivity to Sentry');
  }
}

runDiagnostics().catch(err => console.error('[FAIL] Diagnostic error:', err));

Python — Full Diagnostic Script

# scripts/sentry_diagnostics.py
import sentry_sdk
from sentry_sdk import Hub
import urllib.request
import json
import sys

def run_diagnostics():
    print("=== Sentry Debug Bundle (Python) ===\n")

    # 1. SDK version
    print(f"SDK version: {sentry_sdk.VERSION}")

    # 2. Client check
    client = Hub.current.client
    if not client:
        print("[FAIL] Sentry client not initialized")
        print("       Call sentry_sdk.init(dsn='...') before running diagnostics")
        sys.exit(1)
    print("[OK]   Client initialized")

    # 3. Configuration
    opts = client.options
    print(f"[INFO] DSN configured: {bool(opts.get('dsn'))}")
    print(f"[INFO] Environment: {opts.get('environment', '(default)')}")
    print(f"[INFO] Release: {opts.get('release', '(auto-detect)')}")
    print(f"[INFO] Debug: {opts.get('debug', False)}")
    print(f"[INFO] Sample rate: {opts.get('sample_rate', 1.0)}")
    print(f"[INFO] Traces sample rate: {opts.get('traces_sample_rate', '(not set)')}")
    print(f"[INFO] before_send: {'CONFIGURED' if opts.get('before_send') else 'none'}")
    print(f"[INFO] before_breadcrumb: {'CONFIGURED' if opts.get('before_breadcrumb') else 'none'}")

    # 4. Integrations
    integrations = [i.identifier for i in client.integrations.values()]
    print(f"[INFO] Integrations ({len(integrations)}): {', '.join(integrations)}")

    # 5. Network connectivity
    try:
        req = urllib.request.urlopen("https://sentry.io/api/0/", timeout=5)
        print(f"[OK]   sentry.io reachable (HTTP {req.status})")
    except Exception as e:
        print(f"[FAIL] sentry.io unreachable: {e}")

    # 6. Service status
    try:
        req = urllib.request.urlopen("https://status.sentry.io/api/v2/status.json", timeout=5)
        data = json.loads(req.read())
        print(f"[INFO] Sentry status: {data['status']['description']}")
    except Exception:
        print("[WARN] Could not fetch Sentry status")

    # 7. Test event
    event_id = sentry_sdk.capture_message("sentry-diagnostics test event", level="info")
    print(f"[INFO] Test event ID: {event_id}")

    # 8. Flush
    sentry_sdk.flush(timeout=10)
    print("[OK]   Flush complete")

if __name__ == "__main__":
    run_diagnostics()

Resources

Sentry Troubleshooting Guide (JavaScript) — official SDK debugging steps
Sentry Troubleshooting Guide (Python) — Python-specific diagnostics
Source Maps Troubleshooting — sourcemaps explain and URL prefix debugging
sentry-cli Reference — CLI installation, auth, send-event, sourcemaps commands
Sentry Status Page — real-time platform health
Sentry Support — submit tickets with your debug bundle attached
sentry-javascript GitHub Issues — known bugs and community reports

Next Steps

After collecting the debug bundle:

Events missing? Check beforeSend hook — add logging to confirm events are not being dropped
Source maps broken? Run sentry-cli sourcemaps explain --org ORG --project PROJ EVENT_ID to diagnose URL prefix mismatches
Performance data missing? Verify tracesSampleRate is non-zero and the instrumentation integration is active
Rate limited? Review org quotas at Settings > Subscription and reduce sample rates
Need deeper tracing? Enable debug: true in Sentry.init() to see full SDK lifecycle in console
Ready to file a ticket? Attach the generated sentry-debug-*.md file to your request at sentry.io/support

Agent Skills: Sentry Debug Bundle

Install this agent skill to your local

Skill Files