Info Panel

Overview

The Info Panel helps you quickly understand why a request was slow, failed, or behaved unexpectedly. It provides a single, searchable view of all context for the selected span, including metadata, errors, infrastructure, and related attributes, so you can move faster from observation to root cause.

The power of the Info Panel

In Coralogix, span metadata is composed of span fields and tags. The Info Panel brings this metadata together in one place. It provides a structured, searchable source of truth for a selected span, including identifiers, timing, errors, type-specific attributes (HTTP, DB, messaging, and so on), and the full set of user-driven tags — so you do not need to switch between multiple views while troubleshooting.

Use the Info Panel when you need to quickly answer questions like:

• Is this span slow compared to recent history?
• Did this span fail, and why?
• What request, database, or messaging details were involved?
• Which environment, region, version, host, or container produced this telemetry?
• How do I pivot from this span to related views (Logs, Databases, Service Catalog, Infrastructure, RUM, and so on)?

How to use the Info Panel effectively

1. Start by examining the Span Duration Distribution to determine whether the span is an outlier.
2. Check Errors to quickly identify failures and exceptions.
3. Use Metadata to confirm service, operation, environment, and ownership.
4. Review type-specific sections (for example, HTTP, DB, Messaging) to inspect domain-relevant details.
5. Use Tags to pivot into related views (logs, metrics, infrastructure) or refine your investigation.

Span fields

Span fields are standardized attributes generated by instrumentation and semantic conventions. They describe the operation itself — including core identifiers, timing details, and operation-specific data such as requests, databases, messaging, and events.

In the Info Panel, fields are organized into sections so you can quickly find what you need during an investigation.

Span tags

Span tags are user-defined or custom attributes that add business, environment, or deployment context.

Tags are displayed in a dedicated Tags section, where you can search, pin, and take actions directly from keys and values.

Begin your investigation

Select a trace or span in Explore tracing to begin your investigation. You will navigate to the drilldown with the Info Panel on the right-hand side.

The panel is organized into collapsible sections. Sections appear only when relevant attributes exist for the selected trace or span.

Typical sections include:

• Span Duration Distribution
• Errors (only when the span indicates an error)
• Metadata
• Performance
• Type-specific sections (based on OTel semantic conventions):
  • Requests (HTTP / RPC)
  • Database
  • Messaging
  • Object Stores
  • CloudEvents
  • Network
  • Gen-AI
  • Infrastructure
  • Serverless
  • RUM
  • Runtime / Process
• Tags (everything not already shown elsewhere; grouped and searchable)
• Events (span events, when present)

Span duration distribution

The heatmap shows where the duration of this span falls within the distribution of spans for the same service and operation. It visualizes span duration over time and highlights where the current span falls in that distribution.

The heatmap covers the last 6 hours by default, measured from the end time of the selected span. You can adjust the time window to 12 or 24 hours using the selector in the heatmap header.

How to read the heatmap

• The X-axis represents time over the selected window.
• The Y-axis shows duration ranges calculated from the recent distribution of spans for this service and operation. Buckets adapt automatically to reflect meaningful performance differences.
• Each cell shows the count of spans and the average duration for that time and duration bucket. Darker cells indicate more spans.
• A yellow perforated line marks the duration of the span you are investigating. Hover the line to see a tooltip showing whether the span is slower or faster than recent spans with the same service.name and operation.name.

Cell click actions

Click any cell to open quick actions scoped to that cell's time range and duration bucket:

• View Explore Tracing: opens a new tab with a DataPrime query filtered by service, operation, time range, and duration range.
• View Highlights: filters by the cell's time frame, duration range, service, and operation, and opens the Duration view.
• View Service Operation: navigates to APM for the relevant service and operation within the cell's time range.
• Create Tracing Alert: creates an alert preconfigured with the cell's duration, service, and operation.
• View Query: displays both queries behind the heatmap: the heatmap query and the related spans query.
• Copy: copies the cell's time range, duration bucket, and span count.

What the heatmap can tell you

• If the span falls above p90, investigate downstream dependencies or recent changes.
• If it is within the median but the end-to-end trace is still slow, the bottleneck may be elsewhere in the trace.
• If many spans cluster at higher durations, consider capacity or configuration issues.

Errors

The Errors section appears at the top when the span is considered an error.

Common fields surfaced:

• Status code / outcome (color-coded):
  • HTTP: http.status_code / http.response.status_code
  • RPC: rpc.status_code / rpc.grpc.status_code
• Error type / message / stacktrace (when present), typically from exception attributes:
  • exception.type
  • exception.message
  • exception.stacktrace

This section is intended to answer: "What failed and why?" — with minimal scrolling.

From this section, you can pivot directly to:

• Related Logs for the same trace or span — to see log events emitted around the time of the error.
• Parent span: to see whether the error propagated up the call chain.
• Infrastructure: to check for resource pressure (CPU, memory, disk) around the time of the failure.

Metadata

Metadata provides the identifiers and core context needed for navigation and correlation.

Typical fields:

• Trace ID: copyable; typically links back to the full trace
• Root Span ID: for trace-level selection
• Span ID
• Parent ID: with a link to the parent span when applicable
• Span kind: server, client, producer, consumer, internal
• Span type
• Operation name / action: for example, POST /checkout
• Service name
• Subsystem / module (if tagged)
• Environment: for example, deployment.environment.name

Span type

Span type identifies the type of operation the span represents (for example HTTP request, database query, or messaging call). It helps users quickly understand the purpose of a span without requiring them to inspect raw attributes.

• Span types are inferred using OpenTelemetry semantic conventions (for example, db.system, http.method, rpc.system, messaging.system, aws.s3.*).
• A single span may map to one or multiple span types, depending on its attributes.
• Span types are ordered by relevance.
• The most relevant type-specific section is shown immediately after Metadata.

Span kind

Span kind (server, client, producer, consumer, internal) describes the role of the span in the execution flow, according to OpenTelemetry semantics.

Span kind helps users quickly understand how the operation participates in the request lifecycle:

• Server: handling an incoming request (entry point).
• Client: making an outbound call to another service.
• Producer / Consumer: asynchronous messaging flows.
• Internal: in-process work without external interaction.

Together with Span type, span kind provides both:

• What the operation is (HTTP, DB, Messaging, and so on)
• How it behaves in the flow (entry, exit, async, internal)

This allows users to reason faster about request paths, ownership boundaries, and failure propagation when debugging or analyzing traces.

CloudEvents in Metadata (when present)

If CloudEvents attributes exist, they are surfaced here with the domain cloudevents.event*:

• cloudevents.event_id
• cloudevents.event_source
• cloudevents.event_spec_version
• cloudevents.event_subject
• cloudevents.event_type

Performance

Performance focuses on when the span ran and how long it took.

Typical fields:

• Start time
• End time
• Duration
• Relative start offset: helps users quickly understand execution order and timing relationships between spans within the trace.

Advanced performance indicators (when enabled):

• Duration percentile (p50 / p90 / p95 / p99 compared to similar spans)

Type-specific sections

The Info Panel dynamically surfaces the most relevant fields based on OpenTelemetry semantic conventions and commonly used attribute prefixes.

HTTP (when http.* attributes exist)

• http.method / http.request.method
• http.url / url.full
• http.target / route
• http.status_code
• http.user_agent

RPC / gRPC (when rpc.* attributes exist)

• rpc.system, rpc.service, rpc.method
• rpc.status_code, rpc.grpc.status_code

FaaS / serverless (when faas.* attributes exist)

• faas.name, faas.id, faas.trigger, faas.execution, faas.coldstart
• cloud.resource_id (when available)

Server / network (when server.* / network.* exist)

• server.address, server.port
• network.peer.address, network.peer.port
• network.protocol.name, network.protocol.version, network.transport

Database (when db.* attributes exist)

• db.system, db.name, db.operation, db.user
• db.statement (may be truncated in the UI)

Messaging (when messaging.* attributes exist)

• messaging.system (Kafka, SQS, RabbitMQ, Pub/Sub, and so on)
• messaging.destination, messaging.destination_kind
• messaging.operation, messaging.message_id

Object Stores (when objectstore.* attributes exist)

• objectstore.bucket, objectstore.key, objectstore.operation
• objectstore.size, objectstore.version

Checkly (when checkly.* attributes exist)

When Checkly identifiers are available, the Info Panel generates a deep link back to the relevant Checkly result page.

Gen-AI (when gen.* attributes exist)

When spans contain GenAI attributes (for example, LLM requests or embeddings), they appear in a dedicated section.

Infrastructure (when k8s.*, ecs.*, cloud.*, host.*, deployment.* exist)

This section connects spans to runtime and deployment context, including cloud region/zone, host/pod/container identifiers, and deployment version/build metadata.

Runtime / Processes (when process.*, thread.*, telemetry.*, os.* exist)

• Process ID, Thread ID
• Runtime version (JVM / Node / Go, if available)
• SDK / instrumentation version (if available)

Events

Span events represent things that happened during the span — for example, internal logs, checkpoints, or domain events. When present, they are displayed in the Info Panel under the Events section, as well as in the Gantt view.

Tags

Span tags are user-defined or custom attributes that add business, environment, or deployment context. Displayed in a dedicated Tags section, where you can search, pin, and take actions directly from keys and values.

Actions and quick operations

The Info Panel supports inline actions on most values (via a context menu or action icon). Common actions:

• Copy value (to clipboard)
• Copy key:value
• Copy full path (useful when building queries)
• Include in / Exclude from query: modifies the main query in Explore tracing or Visual Explorer.
• Pin field: pins fields and tags so they are always visible at the top of your Info Panel.

Header-level actions:

• Export trace or span attributes (JSON; CSV may also be available depending on your deployment)
• Share a trace or span link (permalink)

Limitations