Repository navigation
fix(cluster): leave a level someone else changed for the cooldown - #128
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
The watcher's calm streak and cooldown belonged to its own steps only. A server stepped down by hand, or by another watcher, while the watcher had been calm since its first sample was therefore stepped straight back up on the next sample, undoing the step. A level that is not the one the last sample saw, and that this watcher did not step to, now ends both streaks and starts the cooldown.
Mearman
changed the title
|
@codex security review |
|
🎉 This PR is included in version 2.7.4 🎉 The release is available on:
Your semantic-release bot 📦🚀 |
🛡️ Codex Security ReviewSecurity review completed. No security issues were found in this pull request. Reviewed commit: Only the user who started this review can view the report in Codex. ℹ️ About Codex security reviews in GitHubThis is an experimental Codex feature. Security reviews are triggered when:
Once complete, Codex will leave suggestions, or a comment if no findings are found. |
1 participant
The regrade scenario flaked three times on different checks (another server stepped while node 2 stepped up, the role moved node 2 off etcd-only, node 3 did not read etcd-only after a step by hand). The cause is in the core, not the scenario: the watcher's calm streak began at its first sample and its cooldown counted only its own steps, so any server stepped down by hand or by another watcher was stepped straight back up on the next sample, by the service's own watcher running beside the one the test drives. A level that is not the one the last sample saw, and that the watcher did not step to, now ends both streaks and starts the cooldown. The scenario also counts only a known reduced level as a step, waits for a level to read instead of asserting a single read, and prints what regrading recorded for each node when it fails. The core change is a self-contained diff of fleet-regrade-core.sh with its test, for the fleet image's in-tree copy.