Cut back Imply Time to Decision with an observability agent

Clients of all sizes have been efficiently utilizing Amazon OpenSearch Service to energy their observability workflows and achieve visibility into their functions and infrastructure. Throughout incident investigation, Web site Reliability Engineers (SREs) and operations middle personnel depend on OpenSearch Service to question logs, study visualizations, analyze patterns, correlate traces to seek out the foundation explanation for the incident, and scale back Imply Time to Decision (MTTR). When an incident occurs that triggers alerts, SREs sometimes soar between a number of dashboards, write particular queries, verify current deployments, and correlate between logs and traces to piece collectively a timeline of occasions. Not solely is that this course of largely handbook, but it surely additionally creates a cognitive load on these personnel, even when all the information is available. That is the place agentic AI might help, by being an clever assistant that may perceive the right way to question, interpret varied telemetry alerts, and systematically examine an incident.

On this put up, we current an observability agent utilizing OpenSearch Service and Amazon Bedrock AgentCore that may assist floor root trigger and get insights sooner, deal with a number of query-correlation cycles, and finally scale back MTTR even additional.

Answer overview

The next diagram exhibits the general structure for the observability agent.

Functions and infrastructure emit telemetry alerts within the type of logs, traces, and metrics. These alerts are then gathered by OpenTelemetry Collector (Step 1) and exported to Amazon OpenSearch Ingestion utilizing particular person pipelines for each sign: logs, traces, and metrics (Step 2). These pipelines ship the sign information to an OpenSearch Service area and Amazon Managed Service for Prometheus (Step 3).

OpenTelemetry is the usual for instrumentation, and gives vendor-neutral information assortment throughout a broad vary of languages and frameworks. Enterprises of assorted sizes are adopting this structure sample utilizing OpenTelemetry for his or her observability wants, particularly these dedicated to open supply instruments. Extra notably, this structure builds on open supply foundations, serving to enterprises keep away from vendor lock-in, profit from the open supply neighborhood, and implement it throughout on-premises and varied cloud environments.

For this put up, we use the OpenTelemetry Demo software to show our observability use case. That is an ecommerce software powered by about 20 completely different microservices, and generates sensible telemetry information along with function units to generate load and simulate failures.

Mannequin Context Protocol servers for observability sign information

The Mannequin Context Protocol (MCP) gives a standardized mechanism to attach brokers to exterior information sources and instruments. On this answer, we constructed three distinct MCP servers, one for every kind of sign.

The Logs MCP server exposes device capabilities for looking, filtering, and choosing log information that’s saved in an OpenSearch Service area for log information. This permits the agent to question the logs utilizing varied standards like easy key phrase matching, service title filter, log degree, or time ranges. This mimics the standard queries you’d run throughout an investigation. The next snippet exhibits a pseudo code of what the device perform can appear to be:

# Logs MCP Server - Key Features
search_otel_logs(
    question: string,           # Textual content search question for log messages
    service: string,         # Service title to filter logs
    severity: string,        # Log degree (INFO, WARN, ERROR)
    startTime: string,       # Begin time (ISO format or relative e.g., 'now-1h')
    endTime: string,         # Finish time (ISO format or relative e.g., 'now')
    dimension: quantity             # Variety of outcomes to return
)
get_logs_by_trace_id(
    traceId: string,         # Hint ID to retrieve all correlated logs
    dimension: quantity             # Most variety of logs to return
)

The Traces MCP server exposes device capabilities for looking and retrieving details about distributed traces. These capabilities might help lookup traces by hint ID and discover traces for a selected service, the spans belonging to a hint, the service map data constructed based mostly on the spans, and the speed, error, and period (also called RED metrics). This permits the agent to comply with a request’s path throughout the companies and pinpoint the place failures occurred or latency originated.

# Traces MCP Server - Key Features
get_otel_spans(
    serviceName: string,     # Service title to filter spans
    traceId: string,         # Hint ID to filter spans
    spanId: string,          # Span ID to retrieve a particular span
    operationName: string,   # Operation/span title to filter
    startTime: string,       # Begin time (ISO format or relative)
    endTime: string,         # Finish time (ISO format or relative)
    dimension: quantity             # Variety of outcomes to return
)
get_spans_by_trace_id(
    traceId: string,         # Hint ID to retrieve all spans for
    dimension: quantity             # Most variety of spans to return
)
get_otel_service_map(
    serviceName: string,     # Service title to filter service map
    startTime: string,       # Begin time
    endTime: string,         # Finish time
    dimension: quantity             # Variety of outcomes to return
)
get_otel_rate_error_duration_metrics(
    startTime: string,       # Begin time (default: 'now-5m')
    endTime: string          # Finish time (default: 'now')
)

The Metrics MCP server exposes device capabilities for querying time collection metrics. The agent can use these capabilities to verify error fee percentiles and useful resource utilization, that are key alerts for understanding the general well being of the system and figuring out anomalous habits.

# Metrics MCP Server - Key Features
query_instant(
    question: string,           # PromQL question expression
    time: string,            # Analysis timestamp (optionally available)
    timeout: string          # Analysis timeout (optionally available)
)
query_range(
    question: string,           # PromQL question expression
    begin: string,           # Begin timestamp
    finish: string,             # Finish timestamp
    step: string,            # Question decision step (e.g., '15s', '1m')
    timeout: string          # Analysis timeout (optionally available)
)
get_timeseries(
    metric: string,          # Metric title or PromQL expression
    period: string,        # Time period to look again (e.g., '1h', '6h')
    step: string             # Step dimension (optionally available)
)
search_metrics(
    sample: string          # Search sample (helps regex e.g., 'http.*')
)
explore_metric(
    metric: string           # Metric title to discover (metadata + samples)
)

These three MCP servers span throughout the various kinds of information utilized by investigation engineers, offering an entire working set for an agent to conduct investigations with autonomous correlation throughout logs, traces, and metrics to find out the doable root causes for a problem. Moreover, a customized MCP server exposes device capabilities over enterprise information on income, gross sales, and different enterprise metrics. For the OpenTelemetry demo software, you may develop artificial information to help in offering context for affect and different enterprise degree metrics. For brevity, we don’t present that server as part of this structure.

Observability agent

The observability agent is central to the answer. It’s constructed to assist with incident investigation. Conventional automations and handbook runbooks sometimes comply with predefined working procedures, however with an observability agent, you don’t must outline them. The agent can analyze, motive based mostly on the information obtainable to it, and adapt its technique based mostly on what it discovers. It correlates findings throughout logs, traces, and metrics to reach at a root trigger.

The observability agent is constructed with the Strands Agent SDK, an open supply framework that simplifies growth of AI brokers. The SDK gives a model-driven strategy with flexibility to deal with underlying orchestration and reasoning (the agent loop) by invoking uncovered instruments and sustaining coherent, turn-based interactions. This implementation additionally discovers instruments dynamically, so if there’s a change within the capabilities, the agent could make choices based mostly on up-to-date data.

The agent runs on Amazon Bedrock AgentCore Runtime, which gives totally managed infrastructure for internet hosting and working brokers. The runtime helps well-liked agent frameworks, together with Stands, LangGraph, and CrewAI. The runtime additionally gives scaling availability and compute that many enterprises require to run production-grade brokers.

We use Amazon Bedrock AgentCore Gateway to hook up with all three MCP servers. When deploying brokers at scale, gateways are indispensable parts to cut back administration duties like customized code growth, infrastructure provisioning, complete ingress and egress safety, and unified entry. These are important enterprise capabilities wanted when bringing a workload to manufacturing. On this software, we create gateways that join all three MCP servers as targets utilizing server-sent occasions. Gateways work alongside Amazon Bedrock AgentCore Identities to offer safe credentials administration and safe identification propagation from the person to the speaking entities. The pattern software makes use of AWS Identification and Entry Administration (IAM) for identification administration and propagation.

Incident investigation is usually a multi-step course of. It includes iterative speculation testing, a number of rounds of querying, and constructing context over time. We use Amazon Bedrock AgentCore Reminiscence for this function. On this answer, we use session-based namespaces to take care of separate dialog threads for various investigations. For instance, when a person asks “What about Cost service?” throughout an investigation, the agent retrieves current dialog historical past from reminiscence to take care of consciousness of prior findings. We retailer each person questions and agent responses with timestamps to assist the agent reconstruct the dialog chronologically and motive about already accomplished findings.

We configured the observability agent to make use of Anthropic’s Claude Sonnet v4.5 in Amazon Bedrock for reasoning. The mannequin interprets questions, decides which MCP device to invoke, analyzes the outcomes, and formulates the set of questions or conclusions. We use a system immediate to instruct the mannequin to assume like an skilled SRE or an operation middle engineer: “Beginning with a high-level verify, narrowing down affected parts, correlate throughout telemetry sign sorts and derive conclusion with substantiation. You ask the mannequin to additionally counsel logical subsequent steps akin to performing a drill down to analyze inter service dependencies.” This makes the agent versatile to investigate and motive about widespread forms of incident investigations.

Observability agent in motion

We constructed a real-time RED (fee, errors, period) metrics dashboards for the complete software, as proven within the following determine.

To ascertain a baseline, we requested the agent the next query: “Are there any errors in my software within the final 5 minutes?”The agent queries the traces and metrics, analyzes the outcomes, and responds saying there are not any errors within the system. It notes that every one the companies are energetic, traces are wholesome, and the system is processing requests usually. The agent additionally proactively suggests subsequent steps that is likely to be helpful for additional investigation.

Introducing failures

The OpenTelemetry demo software has a function flag that we are able to use to introduce deliberate failures within the system. It additionally contains load technology so these errors can floor prominently. We use these options to introduce a number of failures with the cost service. The true-time RED metrics dashboards within the earlier determine replicate the affect and present the error charges climbing.

Investigation and root trigger evaluation

Now that we’re producing errors, we interact the agent once more. That is sometimes the beginning of the investigation session. Additionally, now we have workflows like alarms triggering or pages going out that may set off the beginning of an investigation.

We ask the query “Customers are complaining that it’s taking a very long time to purchase objects. Are you able to verify to see what’s going on?”

The agent retrieves the dialog historical past from reminiscence (if there’s any), invokes instruments to question RED metrics throughout companies, and analyzes the outcomes. It identifies a vital buy circulate efficiency situation: cost service is in a connectivity disaster and fully unavailable, with excessive latency noticed in fraud detection, advert service, and suggestion service. The agent gives instant motion suggestions—restore cost service connectivity as the highest precedence—and suggests subsequent steps, together with investigating cost service logs.

Following the agent’s suggestion, we ask it to analyze the logs: “Examine cost service logs to know the connectivity situation.”

The agent searches logs for the checkout and cost companies, correlates them with hint information, and analyzes service dependencies from the service map. It confirms that though cart service, product catalog service, and foreign money service are wholesome, the cost service is totally unreachable, efficiently figuring out the foundation explanation for our intentionally launched failure.

Past root trigger: Analyzing enterprise affect

As talked about earlier, now we have artificial enterprise gross sales and income information in a separate MCP server, so when the person asks the agent “Analyze the enterprise affect of the checkout and cost service failures,” the agent makes use of this enterprise information, examines the transaction information from traces, calculates estimated income affect, and assesses buyer abandonment charges because of checkout failures. This exhibits how the agent can transcend figuring out the foundation trigger and supply assist with operational actions like making a runbook for situation decision sooner or later, which could be first the step to offering automated remediation with out involving SREs.

Advantages and outcomes

Though the failure state of affairs on this put up is simplified for illustration, it highlights a number of key advantages that instantly contribute to decreasing MTTR.

Accelerated investigation cycles

Conventional workflows for troubleshooting contain a number of iterations of hypotheses, verification, querying, and information evaluation at every step, requiring context switching and consuming hours of effort. The observability agent reduces these drastically to a couple minutes by autonomous reasoning, correlation, and actioning, which in flip reduces MTTR.

Dealing with complicated workflows

Actual-world manufacturing eventualities typically contain cascading failures and a number of system failures. The observability agent’s capabilities can lengthen to those eventualities through the use of historic information and sample recognition. As an example, it could possibly distinguish associated points from false positives utilizing temporal or identity-based correlation, dependency graphs, and different methods, serving to SREs keep away from wasted investigation effort on unrelated anomalies.

Fairly than present a single reply, the agent can present probabilistic distribution throughout potential root causes, serving to SREs prioritize remediation strategies; for instance:

• Cost service community connectivity situation: 75%
• Downstream cost gateway timeout: 15%
• Database connection pool exhaustion: 8%
• Different/Unknown: 2%

The agent can examine present signs towards previous incidents, figuring out whether or not comparable patterns have occurred up to now, thereby evolving from a reactive question device right into a proactive diagnostic assistant.

Conclusion

Incident investigation stays largely handbook. SREs juggle dashboards, craft queries, and correlate alerts below stress, even when all the information is available. On this put up, we confirmed how an observability agent constructed with Amazon Bedrock AgentCore and OpenSearch Service can alleviate this cognitive burden by autonomously querying logs, traces, and metrics; correlating findings; and guiding SREs towards root trigger sooner. Though this sample represents one strategy, the flexibleness of Amazon Bedrock AgentCore mixed with the search and analytics capabilities of OpenSearch Service permits brokers to be designed and deployed in quite a few methods—at completely different phases of the incident lifecycle, with various ranges of autonomy, or centered on particular investigation duties—to fit your group’s distinctive operational wants. Agentic AI doesn’t change current observability funding, however amplifies them by offering an efficient manner to make use of your information throughout incident investigations.

In regards to the authors