imbajin opened a new pull request, #3403: URL: https://github.com/apache/hugegraph/pull/3403
## Before ➡️ After **Before:** Deleting a graph space leaves PD quota/usage records behind, and stale quota collection can recreate deleted state. **After:** Add authenticated, idempotent cleanup of the exact PD record, refusing cleanup while PD graphs remain. Coordinate cleanup and quota writes, propagate HStore deletion failures, and retain Server configuration when PD cleanup fails so the operation can be retried. Existing empty spaces keep their limits and zero usage. ## Why Make deletion failures explicit and prevent stale local work from recreating deleted state without removing another space's records. ## Validation boundary The source author reports focused PD/Server lifecycle and concurrency tests, not production multi-replica acceptance. Upgrade every PD node before Server: older PD versions reject the new RPC, and previously removed graphs are not restored. Historical orphan records are not automatically purged. Cross-replica concurrent creation/deletion is not a distributed transaction, and schema synchronization is outside scope. Source proposal: [🔗 hugegraph/hugegraph#336](https://github.com/hugegraph/hugegraph/pull/336). -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
