The following is a list of known issues for which a workaround is provided.
| Category | Known Issue | Workaround |
| CMIS Export | When a batch class is imported with CMIS export configured, the CMIS mapping files are missing in the imported batch class, which results in failures at CMIS export. | To resolve this issue, you will need to manually copy the CMIS mapping files from the original batch class to the new batch class.
For help, refer to CMIS Export Failing on Imported Batch Class. |
| EText | OCR results may be incorrect or missing data for non-EText operations. | This issue may occur because of an incorrect entry in the associated RSP file for the batch class.
For help, refer to Poor OCR Results in HOCR.xml. |
| Logging | The following error is logged in the dcma-all.log file if SharedFolders is not accessible:
Uncaught Exception in readOCR method Invalid Batch Instance Folder Path is passed for copying EText Page(s) Hocr File(s) This can impact functionality in a production environment, including setup failures during installation. |
To prevent this issue, ensure SharedFolders is accessible from all nodes.
For help, refer to Migrating the Ephesoft SharedFolders Location. |
| Pickup Service | After upgrading from 2020.1.02 to 2020.1.03, the following error is logged each time you pick up a batch, or when a batch finishes processing:
/dcma/ws/activity/pickup/notification access cannot be granted. com.ephesoft.dcma.authorization.exception.AuthorizationException: None of the rules granted access for the incoming request |
To resolve this issue, you will need to manually edit an entry in the web.xml file.
For help, refer to Error Logged when Picking up Batches After Upgrade to 2020.1.03. |
Known Issues
The following is a list of known issues for which there is not yet a workaround or fix.
| Category | Known Issue |
| Installation | There is a duplicate JAR file in the [Ephesoft_Directory]\Application\WEB-INF\lib folder: jai_core-1.1.3.jar.
This has no impact on Transact functionality during our testing. |
| Logging | The following error is logged in the dcma-all.log file when processing batches over a long period of time:
Exception during job acquisition: execution <UUID> doesn't exist We encountered this error during our endurance testing. This had no impact on batch instance processing. |
| Logging | The following error is logged in the dcma-all.log file when processing batches over a long period of time:
Server <ID> not able to release lock over batch class ID BCX for operation R for Machine Learning We encountered this error during our endurance testing. This had no impact on batch instance processing. |
| Logging | In a multi-server Windows environment, the following error may be intermittently logged in the dcma-all.log:
“Remote job notification failed. <hostname:port> failed to respond” If this error is observed, a scheduled pickup will occur in place of the Activiti job notification process. We observed this error in our testing. This issue had no impact on the functionality of Transact. |
| Logging | In a multi-server Windows environment, the following error may be intermittently logged in the dcma-all.log:
“Batch Instance with ID: <BI#> is already locked by null” We observed this error in our testing. It had no impact on the functionality of Transact. |
| User Interface (UI) | When using MSSQL as your database, multiple options in different plugin configuration screens are displayed several times in the drop-down.
This is due to an issue in the SQL insert queries. The duplicate options are not selectable, and this had no impact on any functionality during our testing. |