Two systems hit the same usage limit. Only one of them was spending it.

Description and use

Two automated pipelines failed due to a shared weekly usage limit, but only one was actually consuming the quota. Analysis revealed the limit was shared, so the best response is to wait for reset or adjust the account limit rather than throttling the pipeline.

Problem

Two automated pipelines failed on the same day due to a shared weekly usage limit, but only one was actually consuming the quota.

Outcome

The failure was caused by a shared account limit, not by the pipelines themselves. Adjusting per-run size would have minimal impact; the appropriate response is to wait for reset or adjust account provisioning.

Workflow

• 1. Observe HTTP 429 failures on both pipelines.
• 2. Check run ledger for failure_class: usage_limit.
• 3. Review account usage records for the week.
• 4. Determine that automation accounted for a small portion of total usage.
• 5. Update escalation rules to note that reducing per-run size is least effective.
• 6. Decide to either wait for reset, raise account limits, or adjust other account activity.
• 7. Document findings for future incidents.

Evidence

Both returned the same response: HTTP 429 You've hit your weekly limit · resets Aug 31, 11pm (UTC)

Source

The limit is scoped to the whole account, and the same account also carries interactive work unrelated to the automation.

Source

The pipeline that stopped was not the one that had filled the quota.

Source