mengw15 opened a new issue, #8995: URL: https://github.com/apache/texera/issues/8995
### Task Summary Part of #5363. Depends on #8994. Add a global service that serves the results of executions that have finished (completed, failed or killed), so they can be read without a computing unit. It mirrors `file-service`: a Dropwizard service in front of a catalog (Lakekeeper here, LakeFS there), with its own sbt module, image, Helm deployment, gateway route and local-dev entry. **Endpoints** - The latest finished execution of a workflow started by the requesting user. - The results of an execution: each operator that has a result, with its output mode and row count, read from the database only. - One page of an operator's result for an execution named by id, with the existing column offset, limit and search options. The read path is the one `ExecutionResultService.handleResultPagination` uses today — a database lookup for the result URI, then an Iceberg read — except that the execution is named explicitly instead of resolved as the latest on a CU. **Access.** The caller needs read access to the workflow, and the execution must belong to that workflow, matching the existing execution endpoints. Reads of per-user-warehouse results go through `WarehouseReadGuard` as elsewhere. **Configuration.** The service reaches Lakekeeper and object storage the same way the CU does (`STORAGE_ICEBERG_CATALOG_REST_URI`, `STORAGE_S3_*`). The per-warehouse catalog cache's size bound and idle expiry, hardcoded today, become configurable, since this service touches far more warehouses than a CU does. **Out of scope.** The frontend (next task), runtime statistics and result export (follow-ups), and any change to deletion. ### Task Type - [ ] Refactor / Cleanup - [x] DevOps / Deployment / CI - [ ] Testing / QA - [ ] Documentation - [ ] Performance - [x] Other -- 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]
