# Self-Healing CI: Auto-Retrying an Azure Blob Storage Throttle

> A 503 ServerBusy throttle from Azure Blob Storage is a temporary rate limit, not a build failure. See the manual fix and how self-healing CI backs off and retries.

Source: https://latchkey.dev/learn/self-healing-ci/self-healing-azure-blob-throttle  
Updated: 2026-06-26

A ServerBusy throttle from Azure Blob Storage is a temporary rate limit, not a build error -- the same request succeeds after a short backoff.

## What makes a failure safely retryable

Automatic retry is only correct for failures that are genuinely transient. Retrying a deterministic failure wastes minutes and hides a real defect, so the classification matters more than the retry mechanism.

- Safe to retry: network timeouts, registry 5xx, transient DNS failures, a service container that was not ready, a spot instance reclaimed mid-run.
- Not safe to retry: assertion failures, compile errors, lint violations, anything that fails identically on every attempt.
- Ambiguous, and worth investigating rather than retrying: out-of-memory kills, disk exhaustion, and flaky tests. These repeat under load and a retry only hides the trend.
- Always record that a retry happened. A pipeline that silently retries is a pipeline whose real failure rate you do not know.

## FAQ

### What causes Self-Healing CI: Auto-Retrying an Azure blob storage throttle?

An artifact transfer to/from Azure Blob Storage fails because the service returned a 503 ServerBusy (or timed out) under load. The build and the blob are fine; the request was briefly throttled. A human re-runs and the transfer completes unchanged.

### How do I fix Self-Healing CI: Auto-Retrying an Azure blob storage throttle manually?

[object Object]

### Can Self-Healing CI: Auto-Retrying an Azure blob storage throttle be fixed automatically?

An Azure ServerBusy throttle carries an unmistakable transient signature and a safe default action: back off and retry. A self-healing CI pipeline recognizes the throttle, waits with backoff, retries the transfer, and only escalates if the operation keeps failing after the rate should have eased, distinguishing a throttle from a genuine

---

Latchkey runs CI/CD that repairs its own failures. Agent entry points: https://latchkey.dev/agent.txt, https://latchkey.dev/openapi.json, https://latchkey.dev/llms.txt