[tombeckenham]

Summary

• A models poll that hits a spent D1 window (d1WindowExhausted) waits D1_WINDOW_WAIT_MS and retries that provider once. A second overload still records ingest_failed. A recovered retry records none, and the poll still does not write providers.status.
• The spec-sync shard does the same. degraded is written only when the retry also fails, and only after the wait, so the status write is not inside the spent window. The 05:11 pair (worldlabs, zai-coding-plan) came from this shard.

Fixes #635

Test plan

• One D1 overload then a successful poll: second call after the wait, no ingest_failed, status stays active
• Two overloads in a row: one ingest_failed, no third attempt
• Spec sync recovers the same way and leaves the provider active
• Spec sync writes degraded only after a second overload, and the next provider runs after that wait
• A non-window failure is not retried

[changeset-bot]

⚠️ No Changeset found

Latest commit: c1276e0

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[coderabbitai]

Caution

Review failed

The pull request is closed.

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

Share

• X
• Mastodon
• Reddit
• LinkedIn