Mapping is Always Lost After Updating BAPI Message Type #
An issue arises when attempting to update the message type by reimporting the BAPI Metadata. After reimporting, the existing message type gets updated, leading to the loss of mapping. However, this issue has been resolved by ensuring that the ‘Source/Target Message Type’ is not removed from the interface during the metadata re-import process.
Fixed Various Issues Updated incorrect Message Status #
- We have encountered an issue during the Request and Response process with the SAP CPI Adapter. Specifically, when receiving a failed response, the API message does not update its status from ‘completed’ to ‘Partial Completed.’ This issue occurs when using the invokeCalloutV3 method in various modes such as Future, Queueable, Batch, Auto, and message_only.
- Incorrect Update of API Message: After reprocessing the message, the API Message incorrectly updates as ‘failed’ even though some business messages are completed successfully while others have failed. This issue has been resolved by updating the status to ‘Partial Completed.
- We encountered another issue when calling the ‘SAP CPI’ adapter and performing the ‘Get’ operation without mapping (only querying data for inbound processing) using callout modes such as Future, Queueable, Batch, Auto, and message_only. As a result, both the request business message and API Message remained in a pending status. However, we have resolved this issue by updating the message statuses correctly.
Unable to Add Interface to Interface Group #
This issue occurs when attempting to add an interface to an interface group that contains multiple interface records within the integration. It results in an error message alert stating, ‘Please enter the value in Sequence.’ The issue arises during the interface trigger when verifying and validating the source name of the interface. We have resolved this problem by using Type.forName(‘xxxxxxx’) instead of Schema.getGlobalDescribe().get(‘xxxxxxx’). This approach provides faster performance and prevents validation checks during interface updates within the Interface group.
Fixing the Issue of the MessageReprocessPerInterfaceGroup Scheduler #
We have resolved the issue with the MessageReprocessPerInterfaceGroup scheduler that was causing a mix-up between inbound and outbound reprocessing messages that caused an incorrect setting of Processing Package size to outbound.
To fix this issue, we implemented the following solutions:
- Ensured that inbound and outbound processes are not mixed within the same scheduler of the interface group. When the interface group contains only outbound interfaces, the scheduler will execute the outbound worker job (IServiceBatchCallout) to reprocess the outbound messages.
- Similarly, when the interface group contains only inbound interfaces, the scheduler will execute an inbound worker job (IServiceBatchReprocess) to reprocess the inbound messages.
- The issue with the scheduler on the interface group that reprocesses outbound messages with a processing package size greater than 1, even when the Processing Package Size is blank, has been resolved. The issue occurred due to a mix-up in the scheduler’s reprocessing of request/response messages (outbound/inbound messages) within the same group. When the ‘Processing Package Size’ setting is blank, the following default values will apply:
- For outbound message reprocessing, the setting will be 1.
- For inbound message reprocessing, the setting will be 10.
Fixed exceeded 100000 distinct who/what’s of inbound processing with the event object #
This issue may arise when using ‘List<SObject> lst = [SELECT Id FROM SObjectX WHERE IsDeleted = false LIMIT 0];’ in Inbound Processing with Event sObject. If you exceed the limit 100000, you will encounter an error message stating ‘First error: exceeded 100000 distinct who/what’s’. To resolve this issue, consider using ‘List<SObject> lst = new List<SObject>();’ instead.
Overlapping Interface Text on View Screen on Compact Mode #
We have encountered an issue where the text on the interface view screen overlaps when the screen size is small and the input text field contains a long value on Compact Mode. This problem has been addressed by implementing a responsive interface detail screen, ensuring that the text view is displayed properly without any overlapping.
Incorrect Order of Nodes and Fields in Generated JSON Payload #
You may encounter a problem with the mapping order incorrect, leading to the generation of an incorrect JSON payload. In certain cases, specific fields may be misplaced and end up in the wrong node. However, we have resolved this issue by appropriately enhancing our JSON builder. This ensures the correct order and field population in the JSON payload.
Added Filter and Pagination for Generic Repository List View #
We have enhanced the Generic Repository by adding filter and pagination functionalities. These enhancements will assist you in searching your repository and easily viewing each repository record in a paginated manner.
Added Various Functions for Generic Baskets List View #
We have implemented several functions on the Generic Baskets List View, as outlined below:
- Refresh button: Added a refresh button to facilitate the refreshing of Basket Records.
- Process button: Introduced a process button to enable the processing of Baskets.
- New field – Processing Type: Included a new field called Processing Type, which provides information on the origin of the basket generation. Examples include batch, bulk, or raw basket processing.
- We have added a Download option that enables you to download all baskets
Resolution of Error When Deleting Pending Event Baskets #
There was an issue encountered when attempting to delete a Pending Event Basket that was under processing. This issue triggered an error message on the pop-up screen, stating “Cannot read properties of undefined [reading ‘exceptionType’].” We are pleased to inform you that this issue has been successfully resolved.
Resolved Frozen Data Loader Screen When Selecting All Records, Appending Their Child Records, and Pushing Records with Raw Upload File #
We have identified an issue that occurs when selecting all records and appending their child records using the Raw Upload File. After pushing these selected records, you may experience a frozen Data Loader screen. We are pleased to inform you that this issue has been successfully resolved.
Fixed ‘Please Select a Record’ Warning Message in V2 Interface Design for Importing Records #
When importing records using the V2 interface design, a warning message stating “Please select a record” may appear, even if you have already selected it. However, this issue has been resolved by allowing the selection of records.
Fixed Filter Baskets with Specific Integration #
We have identified an issue where, when you filter the basket by a specific Integration, it displays all baskets from all integrations, instead of just the baskets for the selected integration. This issue has now been resolved.
Fixed VLOOKUP on Workflow Function and Mapping Issue Getting Error ‘Too many SOQL queries: 201’ #
An issue was encountered when using VLOOKUP on the workflow function and Mapping with multiple child records to look up to a parent. This resulted in an error message: “System.LimitException: skyvvasolutions:Too many SOQL queries: 201.” However, this issue has been resolved.
Fixed Various Issues Regarding DoAlert Function #
- When all Agent Schedulers start, a triggered error prevents the Agent from inserting the interfaceControlRunTime. Due to this issue on the Salesforce side, it is unable to generate an alert notification about the scheduler not running since xxx.
- The issue was caused by the Alert Configuration being set at the integration level, but it did not check the NextRun of interfaceControlRunTime for the Interface Group. This resulted in the system failing to generate an alert notification when the scheduler stop running since xxx. This issue has been fixed by updating the Alert Configuration to check the interfaceControlRunTime at the Integration, Interface Group, and Interface levels.
- Fixed the issue of lost Alerting Setup for Scheduler Processing on integration caused by an unexpected deletion of the scheduler. This issue has now been resolved.
- The issue of accidentally losing the alert configuration record has been resolved. This issue could occur when creating a new integration, interface, interface group, or scheduler. The problem arose when creating a new record, the ID was initially set to null. This caused the trigger for creating the new integration, interface, interface group, or scheduler to query the wrong alert configuration record with a null ID, and then attempt to update the wrong record.
Fixed Various Issues Regarding Find & Replace #
- The Interface may not refresh the interfaces on the Find & Replace the second time cause you may make a change to the interface. For example, you remove or create an interface. This issue has been fixed by refreshing the interface after clicking the Find & Replace Tab again.
- Previously, when there were failed messages with no iData on the outbound interface, an error alert was displayed after selecting the outbound interface. This issue has now been fixed, and instead of the error alert, a warning message “No Failed or Pending Messages are Available” is shown.
- In the previous version, the formula headers in both the mapping and workflow functionalities displayed the complete formula. This resulted in some issues. However, in the current version, the formula headers in both the mapping and workflow functionalities now only show the field names of the formula, rather than the full formula, in the header of the find and replace function.
