[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-07。"],[[["\u003cp\u003eThe migration of Oracle workloads to AlloyDB for PostgreSQL is eventually consistent, meaning data order may vary during transfer but will align once source writes cease and the buffer is cleared.\u003c/p\u003e\n"],["\u003cp\u003eTransactions that are rolled back in Oracle might still be temporarily visible in AlloyDB, and \u003ccode\u003eSAVEPOINT\u003c/code\u003e statements are not supported, potentially causing data discrepancies on rollback.\u003c/p\u003e\n"],["\u003cp\u003eAll destination tables in AlloyDB must have a primary key, and tables without one in the source should be adjusted using specified methods.\u003c/p\u003e\n"],["\u003cp\u003eSchema changes during migration are not automatically applied, requiring updates to the conversion workspace and relevant migration jobs for any schema changes.\u003c/p\u003e\n"],["\u003cp\u003eCertain Oracle data types like \u003ccode\u003eANYDATA\u003c/code\u003e, \u003ccode\u003eLONG/LONG RAW\u003c/code\u003e, \u003ccode\u003eNCLOB\u003c/code\u003e, \u003ccode\u003eUDT\u003c/code\u003e, \u003ccode\u003eUROWID\u003c/code\u003e, \u003ccode\u003eXMLTYPE\u003c/code\u003e, as well as zero dates and Index-organized tables (IOTs) are unsupported and will either be replaced by \u003ccode\u003eNULL\u003c/code\u003e values, not migrated or the table will not be replicated.\u003c/p\u003e\n"]]],[],null,[]]