Ask OpsAI

Ask OpsAI lets you ask questions about your Middleware observability data in natural language.

You can ask about infrastructure metrics, application traces, Kubernetes resources, and logs. OpsAI answers using the telemetry available in your account and the context provided in the conversation.

Use Ask OpsAI to understand system health, investigate errors, identify performance bottlenecks, and get suggested solutions.

Conversations are available from OpsAI Home and through Ask OpsAI on supported Middleware screens.

How Ask OpsAI Works#

Ask OpsAI uses the observability data available in your Middleware account to answer questions and investigate problems.

You can ask questions about:

• Infrastructure hosts and resource utilization
• Kubernetes clusters, nodes, and pods
• APM services, traces, and application errors
• Logs and related events
• RUM applications and frontend issues
• Detected issues and their proposed fixes
• Data from supported connected integrations

OpsAI analyzes the relevant telemetry and returns an explanation, investigation findings, or suggested next steps.

Open Ask OpsAI#

You can access Ask OpsAI in multiple ways.

From OpsAI Home#

Open OpsAI Home.

The page provides access to OpsAI conversations.

From here, you can:

• Start a new conversation.
• View previous conversations.
• Continue an existing conversation.
• Ask questions about your monitored environment.

From Infrastructure Monitoring#

Open a supported Infrastructure Monitoring page and click Ask OpsAI.

For example, while reviewing a Kubernetes cluster, you can ask:

• Which pods are currently unhealthy?
• Which node has the highest memory utilization?
• Why is this pod restarting?
• What should I investigate first?

This helps you investigate infrastructure issues without leaving the monitoring workflow.

From APM#

Open the relevant APM page and select Ask OpsAI.

You can ask questions about application performance, errors, traces, and service behavior.

For example:

• Why is this service experiencing high latency?
• What is causing these application errors?
• Which endpoint has performance issues?
• What could be causing this exception?

From Log Monitoring#

Open Log Monitoring and select Ask OpsAI.

You can ask questions about log activity and errors.

For example:

• What does this error mean?
• Why are these logs appearing repeatedly?
• What could be causing this exception?
• How can I fix this issue?

Ask Questions About Your Data#

Ask OpsAI supports natural-language questions about the telemetry collected in Middleware.

You do not need to write a query to start an investigation.

For example:

OpsAI uses the available data to provide an answer and, where appropriate, recommends further investigation or a solution.

Pass Context to Ask OpsAI#

When you open Ask OpsAI from a supported monitoring module, you can provide relevant telemetry as context.

For example, while viewing a log entry, you can pass that log to OpsAI and ask it to analyze the error.

Example: Analyze a Log#

1. Go to Log Monitoring.
2. Open the relevant log entry.
3. Select Ask OpsAI.
4. Pass the log information as context.
5. Ask a question such as: What caused this error, and how can I resolve it?
6. Review OpsAI's analysis and suggested solution.

Passing a log as context keeps the question tied to the specific event, including its message and available properties.

Investigate Infrastructure Issues#

Ask OpsAI can help you investigate infrastructure performance using the available host and Kubernetes telemetry.

For example, if a host has high CPU utilization:

1. Open Infrastructure Monitoring → Host.
2. Select the affected host.
3. Review its CPU utilization.
4. Open Ask OpsAI.
5. Ask why the host is consuming high CPU.
6. Review the findings and recommended next steps.

You can then ask follow-up questions to narrow down the investigation.

Investigate Application Issues#

Ask OpsAI can analyze application errors and performance issues using APM telemetry.

For example:

1. Open APM.
2. Select the affected service or trace.
3. Open Ask OpsAI.
4. Ask about the error or performance problem.
5. Review the analysis and suggested solution.

If the issue requires deeper investigation, you can open the corresponding detected issue in OpsAI and use Fix with OpsAI.

Continue a Conversation#

Ask OpsAI supports follow-up questions, allowing you to investigate a problem through a conversation.

For example:

Initial question Why is my checkout service experiencing high latency?

Follow-up question Which database operation is contributing the most latency?

Next question What changes would help reduce the response time?

OpsAI can use the conversation context and available telemetry to refine its analysis.

This helps you move from identifying a problem to understanding its possible cause and solution.

View Previous Conversations#

From the OpsAI Home screen, you can access previous conversations.

Use conversation history to return to an earlier investigation or continue asking questions about an existing issue.

You can also start a new conversation when investigating an unrelated problem.

Ask OpsAI from Slack#

You can also interact with OpsAI directly from Slack when the Middleware Slack integration is configured.

Mention @Middleware in a channel or send it a direct message to ask questions about your observability data.

For example:

@Middleware Which Kubernetes pods are currently experiencing errors?

You can also ask questions within an OpsAI investigation thread.

For example:

@Middleware Can you explain the root cause and suggested fix?

OpsAI replies in the Slack conversation, allowing your team to collaborate on the investigation.

For Slack setup and permissions, see the OpsAI Settings documentation.