Conversation
There was a problem hiding this comment.
Pull Request Overview
This PR improves the cluster readiness checking mechanism by verifying topology consensus rather than just individual node status. The changes replace CLUSTER INFO checks with CLUSTER SLOTS comparisons to ensure all nodes agree on the cluster topology before proceeding.
- Changed
waitClusterto useCLUSTER SLOTScommand instead ofCLUSTER INFOto verify topology consensus - Added
waitClustermethod to theEnvclass to expose cluster waiting functionality - Fixed trailing whitespace in
getInformationAfterDispose
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| RLTest/redis_cluster.py | Updated waitCluster logic to compare CLUSTER SLOTS responses across nodes and removed trailing whitespace |
| RLTest/env.py | Added waitCluster method to delegate to cluster environment implementations when applicable |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Pull Request Overview
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
* refactor to use cluster slots to ensure agreement * expose to Env * wait for both OK and same topo * minor improvements
* refactor to use cluster slots to ensure agreement * expose to Env * wait for both OK and same topo * minor improvements
* initial commit * Add message and depth parameters to all Query assertions (#237) * add depth and message for Query, and some related fixes * remove unused import * CI update (#238) * split nightly from generic CI * version bump actions * update poetry lock * remove usage of pkg_resources * bump poetry version * update toml and poetry lock again * fix whitespace * Rename nightly-build job to build in biweekly.yml * support version on python 3.7 * run CI on older and newer python version * add old versions back and lock poetry * fix run * another attempt * another attempt * another attempt * another attempt * another attempt * another attempt * another attempt * fix poetry lock * CI improvement * improve comment and concurrency group * add failure notification * Improve waitCluster (#239) * refactor to use cluster slots to ensure agreement * expose to Env * wait for both OK and same topo * minor improvements * print rltest args * code review comments * code review comments * fixes * small fix * remove redundant function call * Apply suggestions from code review Co-authored-by: GuyAv46 <[email protected]> --------- Co-authored-by: GuyAv46 <[email protected]>
Improve
waitClusterby using bothCLUSTER INFOandCLUSTER SLOTS, so we do not just wait for all nodes claiming 'ok', but they also agree on the topology.Also, expose this function in
Env, so cluster envs can use it to wait with no additional implementations. Stand-alone Envs will return immediately.