{"version":"2.1.282","anchor":"api-error-records-take-the-request-id-from-the-x-amzn-reques","canonical_anchor":"api-error-records-take-the-request-id-from-the-x-amzn-reques","heading":"API errors now pick up a request id from Amazon's response header","tier":"notice","area":"Elsewhere","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.282\/e\/api-error-records-take-the-request-id-from-the-x-amzn-reques","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.282","markdown":"### API errors now pick up a request id from Amazon's response header\n\nWhen an API error carries no request id of its own, Claude Code now reads one from the `x-amzn-requestid` response header\n\n**What**\n\nWhen a request to the model fails, Claude Code records a request id, a label that identifies that exact request, in its logs and error reports. It used to take this id only from the error itself. When the error has none, it now falls back to the `x-amzn-requestid` header in the response, which Amazon-hosted endpoints send.\n\n**Why**\n\nErrors from those endpoints can now carry a request id. That makes a failure easier to match up with AWS support.\n\n- Area: Elsewhere\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5"}