feat: Tracing support

[maciaszczykm]

Fixes PROD-5129 too.

Test Plan

Test environment: https://console.plrl-dev-aws.onplural.sh/
Test image: sha-d14b2a4

Checklist

• I have added a meaningful title and summary to convey the impact of this PR to a user.
• If required, I have updated the Plural documentation accordingly.
• I have added tests to cover my changes.
• I have deployed the agent to a test environment and verified that it works as expected (required only when changing agent code).

Plural Flow: console

[linear]

PROD-4759

PROD-5129

[plural-copilot]

✅ Deploy preview for docs-pr-4110 is ready!

Name	Details
⚡ Service	docs-pr-4110 (5 / 5 ready)
☁️ Cluster	plural (eks)
Commit sha	a8fed4a
Plural URL	https://console.mgmt.plural.sh/cd/clusters/446acdef-8524-425a-a49f-5336a...
🔍 Logs URL	https://console.mgmt.plural.sh/cd/clusters/446acdef-8524-425a-a49f-5336a...
🌊 Flow URL	https://console.mgmt.plural.sh/flows/08a17af1-8578-449e-9ae1-28ccd97eafc...
Preview Link	Preview

---

You can learn more about preview environments and Plural Flows here

[soffi-ai]

Soffi AI Summary

This PR introduces end-to-end distributed tracing support for the AI Workbench, addressing PROD-5129. On the backend, it adds opt-in OpenTelemetry instrumentation to the Elixir server (including Ecto query tracing, Git command tracing, and workbench execution span tracking via a new Console.AI.Workbench.Tracking module). On the frontend, it delivers a rich trace visualization experience within the Workbench job view: a Gantt-style waterfall timeline, a topology graph showing service dependencies, semantic status colors (ok/warning/error), fullscreen mode, span detail panels, and multi-trace selection. Utility modules handle span ordering, tree metadata, tick generation, duration formatting, and color assignment per service. Tests cover span shape, topology rendering, canvas summaries, and user interactions.

---

Updated: 2026-09-08 10:01 UTC

Deploy in Soffi

[maciaszczykm]

@greptileai

[greptile-apps]

Greptile Summary

Adds end-to-end tracing support and new Workbench trace visualization:

• Configures opt-in OpenTelemetry export and instruments HTTP, Phoenix, GraphQL, Ecto, ReqLLM, Git, Helm, and Workbench execution.
• Sanitizes URL-bearing telemetry attributes and disables capture of GraphQL documents, variables, results, and errors.
• Adds Workbench trace timelines, span and service topology views, fullscreen details, trace canvas blocks, and supporting design-system controls.
• Improves canvas metrics, logs, summaries, legends, and time-range controls, with frontend and backend tests.

Confidence Score: 5/5

The PR appears safe to merge; no new actionable correctness, security, or repository-rule violations remain.

The current changes preserve Bandit handler behavior while stripping request query strings before span creation, disable sensitive GraphQL payload capture, sanitize telemetry URL attributes, and keep Workbench trace rendering and canvas layout consistent. The five earlier findings were manually resolved without explanatory replies and therefore are not outstanding.

Important Files Changed

Filename	Overview
lib/console/otel/tracing.ex	Configures tracing integrations and centralizes URL, GraphQL, database, and HTTP query-string redaction.
lib/console/ai/workbench/tracking.ex	Adds parent run spans, typed activity spans, and propagated context for concurrent Workbench activities.
lib/console/ai/workbench/engine.ex	Wraps Workbench execution and activity dispatch in the new tracing boundaries.
js/console/src/components/workbenches/workbench/job/WorkbenchJobTraces.tsx	Implements trace selection, timeline rendering, fullscreen span details, and topology view controls.
js/console/src/components/workbenches/workbench/job/workbenchJobTraceUtils.ts	Provides trace grouping, ordering, timing, severity, and formatting utilities used by the visualization.
js/console/src/components/workbenches/workbench/job/workbenchJobTraceGraph.ts	Builds span-level and aggregated service-level topology nodes and edges.
js/console/src/components/workbenches/workbench/job/WorkbenchJobActivityResults.tsx	Adds trace result loading and rendering while improving metrics legends, summaries, and time-range filtering.
js/console/src/components/workbenches/workbench/job/WorkbenchJobCanvas.tsx	Adds trace canvas blocks and refines metrics, logs, and summary presentation without overriding stored grid layout.
rel/runtime.exs	Enables batch OTLP trace export only when a runtime endpoint is configured.
mix.exs	Adds the OpenTelemetry instrumentation, exporter, and process-propagation dependencies required by tracing.

_{Reviews (7): Last reviewed commit: "drop query strings from bandit spans" | Re-trigger Greptile}

[github-actions]

Storybook preview: https://6a8c1fbd32d6c1cc8fc131a5-imbszzyujs.chromatic.com/

[maciaszczykm]

@greptileai

[maciaszczykm]

@greptileai