Constrain total fetch time across DNS and redirects
Each redirect gets a fresh timeout, extending one import indefinitely.
- Focused work estimate
- 3h + prerequisites
- Priority in the scenario
- High
- Engineering practice
- Cancellation · Timeout design
Estimated field mix
- Networking50%
- Site reliability30%
- Backend20%
Field percentages are editorial estimates of the ticket's engineering focus. They total 100%; they are not measured time, proficiency scores, or ownership evidence.
Review it, then add it to your workspace.
The board opens an editable draft; nothing is saved until you confirm it. Sign-in and workspace permissions apply, and Demo boards remain ephemeral.
Project context
A fictional content-import service accepts document URLs. The team needs explicit destination approval, redirect handling, and response limits before enabling imports.
Setup prerequisites
- Create a local HTTP destination simulator and DNS double with authorized address cases; contact no real external hosts.
Preceding work
Complete these dependencies, or supply their agreed outputs before taking this ticket.
- BEGRESS-101 · Define the approved destination contract for imports
- BEGRESS-102 · Normalize import URLs without changing their authority
- BEGRESS-103 · Bind resolved addresses to the approved hostname policy
- BEGRESS-104 · Revalidate every redirect before following it
- BEGRESS-105 · Enforce response byte limits during streaming
Acceptance criteria
- Use one total deadline for all stages.
- Propagate cancellation to owned operations.
- Settle even when a destination double ignores cancellation.
Implementation constraints
- No automatic retry after the total budget expires.
Verification to include
- Complete a multi-stage fetch within budget.
- Exhaust the deadline during redirects and verify cleanup.
Deliverables
- End-to-end fetch deadline.
Rollout and recovery
Return a retryable unavailable result only under the documented import policy.
Value of the work
For the engineer: Practice URL interpretation, network policy enforcement, and resource controls.
For the team: Provide a reusable fetch boundary that makes destination authority and failure behavior inspectable.
Evidence boundaries
Outcome Evidence: Tests, patches, and runbooks are requested deliverables. They become Outcome Evidence only through a qualified Mission and immutable Evidence IDs.
Ownership Evidence: Independent adaptation must be observed under a declared verification policy and cite immutable Evidence IDs. Completing a planning ticket establishes no Ownership Evidence.