To trace an MCP tool call in production, instrument the caller or MCP client, the MCP server, and the downstream services the tool invokes; propagate trace context across the MCP boundary; then export and inspect the resulting telemetry. OpenTelemetry provides a portable instrumentation and export layer, but no single integration should be assumed to capture every span, metric, log, or dependency automatically.
What an end-to-end MCP trace needs to show
A useful trace follows the work across each boundary, rather than stopping at the MCP request. Map the path from the agent or caller through its client SDK and transport, into the server handler, and onward to the tool’s dependencies. Instrument the components that perform that work: an MCP span identifies protocol activity, while HTTP, database, queue, or other dependency spans reveal what happened inside the tool.
As an Amazon Associate I earn from qualifying purchases.
The OpenTelemetry MCP service example pairs MCP instrumentation with separate HTTPX instrumentation for outbound API calls. That distinction matters: an MCP span alone can show that a tool call happened without explaining the time spent in an uninstrumented downstream request. The example also exports metrics, while its standard-library logs go to stdout and are collected by the container runtime rather than a LoggerProvider; those logs are not correlated with its traces. OpenTelemetry’s MCP service documentation describes this particular setup, not universal defaults.
Implement tracing in a deliberate sequence
- Map the call path and ownership. Record the caller, client SDK, transport, server handler, and dependencies each tool can reach. Note where each runs and who can instrument it.
- Instrument the client and server. Use instrumentation supported by the exact language SDK and release you deploy. For the TypeScript SDK v2, Telemetry.dev’s integration documentation says to instrument each transport before calling
client.connect()orserver.connect(); it also says both sides need instrumentation to record client and server spans. Verify these steps against your installed SDK version. Read the TypeScript integration details. - Propagate context across MCP. Confirm that the caller’s trace context is present in the MCP request metadata and that the server continues the trace. An MCP Blog release-candidate announcement dated July 28, 2026 describes W3C
traceparent,tracestate, andbaggagekeys in_meta. Because that post announces a release candidate, check the current specification and the behavior of your SDKs and gateways before relying on it. See the MCP announcement. - Instrument downstream work separately. Add appropriate instrumentation for the HTTP clients, databases, queues, or other systems the tool calls. The MCP trace boundary does not automatically reveal work performed by dependencies.
- Export and inspect the telemetry. Configure an OpenTelemetry export path and use your backend to find slow calls, errors, and gaps. For one concrete backend example, Google Cloud documents filtering MCP server spans with
mcp.method.nameset totools/call, then examining them in Trace Explorer. Its interface is an example, not a requirement for other backends. Google Cloud’s MCP tracing guide gives the details. - Configure metrics and logs intentionally. Verify that metrics are emitted and exported, logs are collected, and any desired trace-to-log correlation is configured. Span collection by itself does not establish coverage for either signal.
- Validate representative outcomes. In staging, exercise successful calls, tool-level errors, transport errors, cancellation, notifications, and downstream failures. Inspect the emitted telemetry to make sure it reflects the application outcome rather than only the transport status.
Check what each integration actually records
TypeScript SDK v2 integration
Telemetry.dev documents spans for JSON-RPC requests, with attributes including method, request ID, protocol version, session ID when available, errors, and method-specific details. It labels tools/call as execute_tool. The same documentation says notifications do not produce processing spans and that the integration does not emit metrics. These are properties of that integration, not rules for every MCP implementation. Check its stated coverage.
#1 Best Overall
- WIFI ENABLED TO CONTROL FROM ANYWHERE – Transform your home into a smart home with the Feit Electric Smart Wi-Fi Plug. Remotely turn on or off lights, fans, coffee makers, or other home appliances from your smartphone or tablet. Works seamlessly with Alexa and Google Home, giving you effortless voice control without needing a separate hub. Manage your devices anytime, whether you’re at home, at work, or traveling.
- SIMPLE SETUP, NO HUB REQUIRED – Enjoy the convenience of smart home automation without extra equipment. The plug connects directly to your 2.4 GHz Wi-Fi network, making installation fast and easy. Plug it in, download the Feit Electric app, follow the simple steps, and your devices are instantly connected. Perfect for beginners or anyone looking to expand their smart home ecosystem with minimal hassle.
- SET YOUR ROUTINE & SAVE ENERGY – Save energy, stay organized, and automate daily routines with customizable schedules and timers. Set your lamps, heaters, or appliances to turn on and off automatically at specific times, ensuring your home is always comfortable and efficient. Ideal for morning routines, evening wind-downs, or holiday lighting, giving you peace of mind and energy savings without constant manual operation.
- ENHANCED SAFETY & CONVENIENCE – Protect your home and appliances with the Feit Electric Smart Plug’s durable design and safety features. Its compact size fits easily into standard indoor outlets without blocking other sockets. With real-time app control and notifications, you can monitor appliance activity and prevent energy waste. Ideal for families, pet owners, or anyone seeking a smarter, safer, and more convenient home setup.
- RELIABLE 2.4GHz WI-FI PERFORMANCE – Designed to work exclusively on 2.4 GHz networks, this smart plug provides stable connectivity for smooth operation of all your devices. Avoid interruptions caused by incompatible networks, ensuring your appliances respond instantly when controlled via the app or voice commands. Perfect for indoor home use, it supports up to 15 amps, handling heavy-duty appliances safely and reliably.
Python server instrumentation
The MCP Python SDK documentation says each MCPServer and low-level Server emits one SERVER span per inbound message by default. It describes tools/call spans using OpenTelemetry GenAI semantic conventions. It also says that, when inbound context is absent, the server span parents to the current server context. Confirm the behavior for your installed Python SDK version; it should not be generalized to other language SDKs. Consult the Python SDK documentation.
Metrics and logs
Coverage varies by setup. In the OpenTelemetry MCP service example, metrics are enabled unless disabled in that demo configuration, but standard-library logs are written to stdout and are not correlated with traces. Grafana documents a separate MCP setup covering Prometheus metrics, OpenTelemetry distributed tracing, and log export, with traceparent and tracestate in _meta when present. Treat that as a vendor implementation and recheck its versioned configuration before applying it. Grafana’s developer documentation describes its approach.
Rank #2
- equipped with atom n2600 d2700 processor, compatible with many freebsd based router systems, linux distros, or win.os supported, easy configuration and management
- Please note, this is a barebone only. A system memory, a storage drive and an operating system are needed to complete this system
- 13-19 inches 1u, 50w power, with power cord, make sure to use a big brand memory and ssd/hdd with quality assurance
- Designed with console, 2 x usb, 4 x lan, vga, power switch, size at 290 x 180 x 44mm
- There are 2 inside reserved fans on chassis, which could be removed freely or be turned on in a high temperature environment to ensure the best function of the product
Choose an export and inspection setup that fits
OpenTelemetry is the instrumentation and export layer; the backend is where teams store and query what they export. When evaluating a setup, check the dimensions that determine whether it can answer your operational questions:
Recommended Free Tools
- Language and SDK coverage for every client and server in the path.
- Whether it records client, server, and downstream dependency spans.
- How trace context crosses transports, gateways, and service boundaries.
- Whether metrics and logs are supported and correlated as needed, in addition to traces.
- Which export protocols and backends are compatible with your deployment.
- How tool attributes and request data can be controlled to protect privacy.
- What sampling, retention, and operational-cost controls are available.
There is no evidence here for a universal backend ranking or comparable cost, retention, or performance figures. Select based on required signal coverage, privacy controls, and the query workflow your on-call team will use.
Rank #3
- Shelly Plus 1 PM is a Wi-Fi smart relay switch with 1 channel, up to 16A with power metering that can be used also as a WiFi repeater and Bluetooth gateway. Shelly Plus 1PM can be used to monitor the consumption and take control of home appliances, electric circuits, and office equipment individually.
- Automate electrical appliance and control - With Shelly Plus 1PM you can automate any electrical appliance in your home and control it remotely. Shelly Plus 1PM can control appliances with a large load which makes it perfect for kitchen appliances and domestic systems monitoring and control. You can get precise measurements of the power consumption of each appliance and switch in on/off remotely, no matter where you are.
- Set and be prepared for everything - Reveal the full potential of Shelly Plus 1PM by combining it with other devices from your home network! Set Shelly Plus 1PM to activate custom scenes based on hour, light, or various occurrences. For example, you can set Shelly Door/Window sensor to report a porch door opening and activate Shelly Plus 1PM to turn on the hot tub heaters only in the hours after 8 pm.
- Shelly Customer Service - Shelly is one of the fastest-growing Smart Home brands in the world with devices, providing solutions for the automation of private homes, buildings and businesses. We provide our customers with professional support and a 3 years device warranty.
- Shelly Smart Control App will help you control your Shelly devices remotely and will send notifications for all automated events in your home. You can easily configure devices and manage their settings individually, or you can create personalized scenes by combining Shelly devices to trigger certain actions in your home automation.
Find MCP tool calls in a trace backend
Google Cloud provides one specific inspection workflow: filter MCP server spans where mcp.method.name equals tools/call, then explore the matching spans in Trace Explorer. The result is useful only to the extent that instrumentation emits the attribute and the relevant spans are exported. Other backends may use different query interfaces or attribute handling. See the Google Cloud steps.
Quick Recap
Rank #4
- Portable 100M/1G Network TAP Appliance for remote capture of data traffic
- Integrated with a Raspberry Pi 4 module (8GB RAM and 64GB Micro SD Card)
- Can be used as a standalone 100M/1G network TAP with the external monitor port
- Dual DC power inputs for enhancing overall system availability
Protect coverage and data quality in production
- Do not equate a request span with a complete trace. Check that both MCP endpoints and each important dependency produce spans connected by propagated context.
- Distinguish requests from notifications. Some integrations omit notification processing spans, so an absence of a span does not necessarily mean the client sent nothing.
- Inspect errors at the right layer. A transport-level success does not prove the tool or its downstream work succeeded; test and review these outcomes separately.
- Review attributes before export. Tool arguments, identifiers, and request details can be sensitive. Decide which attributes are necessary for diagnosis and apply the controls supported by your instrumentation and backend.
- Check signal configuration independently. Confirm trace, metric, and log collection and correlation separately rather than treating one configured exporter as proof of complete observability.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
