Export Objects using Script causes error if attempting to export 1000+ objects
book
Article ID: 87597
calendar_today
Updated On:
Products
CA Automic Workload Automation - Automation Engine
Issue/Introduction
Error Message : U0020568 runtime errror in object 'FRAGMENT_JOBP(GET)', Zeile '00022': DS-return code: '0001001708': 'U1001708 DOM error: NOT_FOUND_ERR', error position in UCDSFUN: 'DS_DB2XML_JOBP33'.
While exporting a large number of Objects (from 500 - 1000+) the above error message will appear during the export process.
Investigation
While the documentation states that the maximum number of exported objects can be set to a maximum of 100000 (UC_SYSTEM_SETTINGS - MAX_EXPORT_COUNT), it is not recommended to use the Import/Export function for mass transfers of objects.
This is due to the structure of the .xml files which use a load of memory while parsing the nodes of the file which is done during the export.
Example Parsing of a 2 MB export file will consume 4 GB of memory during the process, calculating this for a file of 18 MB (which is a size of about 2000 objects) will increase the memory consumption accordingly, which will cause the export process to crash and the error message to be generated.
A counter check, besides monitoring the Automic processes, would be to try to open a 10 MB .xml File in Internet Explorer and watch the memory consumption of the Internet Explorer process.
Cause type: By design Root Cause: Import size of 2MB requires up to 4GB memory just for parsing the xml. The XML parser has a limit on the node handles apx. 3mio handles in one XML structure. This MUST be done with the DB UNLOAD/LOAD using he TRANSPORT CASE.
Resolution
This field was added on 30/03/2017. This article has not been updated yet. Refer to the "Description" or "Workaround" sections for solution information.
Fix Status: No Fix
Additional Information
Workaround : N/ASOLUTIONUse the transport case to import/export a large number of objects out or into your automic system(s). By using the transport case instead, all the processes are handled via the database and no .xml parsing is necessary.