tlyyxjz commented on issue #12230: URL: https://github.com/apache/gravitino/issues/12230#issuecomment-5473635390
Really thorough write-up — this class of bug (cache eviction racing policy writes, leaving a node permanently denied until restart) is nasty precisely because nothing throws. We've been working on tooling around exactly this: diagnosing JCasbin/Casbin deployments where policies drift out of sync with what the enforcer actually loaded — comparing what's in the adapter vs what's in memory, and surfacing "why did this request get denied" step by step. Since you clearly hit this in production across three pods — do you think something like that would have saved you time on this one? Curious whether you'd find a config/policy consistency checker useful for Gravitino's JCasbin integration, or if you'd rather keep this logic internal. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
