---
title: "Verify a partial dashboard failure preserves healthy panels"
description: "Inject one sandbox dashboard data failure externally and check that working panels stay accurate while the failed panel offers a scoped recovery."
canonical_url: "https://vostride.com/agent-library/network-recovery/partial-dashboard-failure.md"
md_url: "https://vostride.com/agent-library/network-recovery/partial-dashboard-failure.md"
content_kind: "recipe"
topic: "network-recovery"
---

# Verify a partial dashboard failure preserves healthy panels

Inject one sandbox dashboard data failure externally and check that working panels stay accurate while the failed panel offers a scoped recovery.

## Risk

A page-level success state can conceal a failed panel, while a global retry can reset useful filters elsewhere. Use distinct panel data and a single failing endpoint to test isolation and recovery semantics.

## Fixtures

- An owned sandbox dashboard with summary, recent activity, and queue panels, each backed by a distinguishable fixture dataset.
- An external fault rule affecting only the queue data request, with a report confirming the other panel responses remain healthy.
- A read-only dashboard oracle and a test account with permission to view all three fixture panels without modifying data.

## Steps

1. Open the dashboard under the prepared queue fault and verify the summary and recent activity panels match their fixture values.
2. Verify the queue panel shows a specific unavailable state and retry affordance instead of an empty queue presented as current data.
3. Change the activity panel filter and verify its filtered identifier set while the queue remains unavailable.
4. Remove the queue fault externally, use its retry action, and verify the expected queue entries appear without resetting the activity filter.
5. Refresh the page after recovery and verify all panels match their independent datasets; inspect the fault trace to confirm only the intended panel failed during the first load.

## Expected results

- Healthy panels retain accurate data throughout the queue failure.
- The failed queue is visibly distinguished from a legitimately empty queue.
- Scoped retry restores queue data and preserves the unrelated activity filter selection.

## False-pass traps

- An empty-state illustration can hide a failed network request and falsely pass an empty-queue check.
- Faulting every endpoint tests total outage rather than panel isolation.

## Cleanup

- Disable the queue fault and restore the dashboard filter baseline for the disposable account.

## Use with agent-qa

Adapt this test design to your application's acceptance criteria and configured target using the [recipe execution guide](/agent-library/software-factory/turn-a-recipe-into-an-agent-qa-test.md). External fixture setup and non-UI checks require your own supported hooks or test tooling. Validate the authored test, execute it, and inspect the completed evidence; this recipe is not a recorded test result.

## References

- [agent-qa first test](/docs/agent-qa/guides/first-test.md)
- [agent-qa hooks](/docs/agent-qa/guides/hooks.md)
- [agent-qa test](/docs/agent-qa/configuration/test.md)

- [Network recovery QA recipes](/agent-library/network-recovery.md)
- [Software factory QA library](/agent-library.md)