imbajin opened a new pull request, #3294: URL: https://github.com/apache/hugegraph/pull/3294
Requests can leave scan work or native resources alive after cancellation, and Store can close its engine before in-flight RPCs finish. This change releases request-owned resources on failure/cancellation and drains Store RPC work before storage shutdown. - Release every transaction owner and preserve cleanup failures. - Close scan iterators and child streams when requests fail or are cancelled. - Stop accepting new Store work, finish or cancel in-flight work, then close storage. Based on current Apache master, including the merged ToplingDB integration. This is a general lifecycle fix for Server/Store; it does not add a provider or change how users switch to ToplingDB. Changes already merged in [apache/hugegraph#3275](https://github.com/apache/hugegraph/pull/3275) are retained rather than duplicated. Validation: Java 17 full-repository compilation and focused transaction, scan, cancellation, shutdown and legacy serialization regressions pass. Existing review threads are resolved; CI for this updated commit remains to be confirmed. See `docs/transaction-lifecycle.md` for lifecycle behavior. Doc - Done. -- 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]
