Table of Contents

Search

  1. Abstract
  2. Installation and Upgrade
  3. 10.2 HF1 Fixed Limitations and Closed Enhancements
  4. 10.2 HF1 Known Limitations
  5. 10.2 Fixed Limitations and Closed Enhancements
  6. 10.2 Known Limitations
  7. Emergency Bug Fixes Merged into 10.2
  8. Informatica Global Customer Support

Workflows Known Limitations (10.2)

Workflows Known Limitations (10.2)

The following table describes known limitations:
Bug
Description
MWF-1448
If a domain fails over from one gateway node to another gateway node while a workflow runs, the Data Integration Service re-runs any Mapping task that began before the domain failed. Any mapping that runs again fails during workflow recovery.
The issue arises when the following conditions are true:
  • The workflow contains multiple parallel Mapping tasks and Command tasks between two Inclusive gateways.
  • The batch execution pool size specifies that the mappings in the Mapping tasks must run in two or more batches.
  • You manually recover the workflow or the workflow recovers automatically after the domain fails over.
  • You configure the nodes in a grid, and the grid Data Integration Service runs the workflow.
MWF-1419
If a domain fails over from one gateway node to another gateway node while a workflow runs, the Data Integration Service might re-run a mapping that completed before the domain failed.
The issue can arise when the following conditions are true:
  • The workflow contains multiple parallel Mapping tasks between two Inclusive gateways.
  • The batch execution pool size specifies that the mappings in the Mapping tasks must run in two or more batches.
  • The domain fails during the Mapping task execution and before the final batch of mappings starts to run.
  • You configure the nodes in a grid, and the grid Data Integration Service runs the workflow.
MWF-1414
If the Data Integration Service restarts while a workflow is running, the Monitoring tool might fail to report that the workflow entered a cancelled state.
The issue can arise when the domain uses different Model Repository Services to store the workflow metadata and to monitor the workflow.
MWF-1396
Workflow metadata might become corrupted in a Blaze or Hive environment and the workflow might fail to run.
MWF-1369
The Monitoring tool is slow to update the status of multiple workflows and workflow tasks following a domain restart.
The issue can arise when the following conditions are true:
  • The domain uses different Model Repository Services to store the workflow metadata and to monitor the workflow.
  • When the domain restarts, the Model Repository Service that performs monitoring restarts after the Data Integration Service and the workflow engine restart.
  • Multiple workflows were running when the domain failed. For example, monitoring does not reflect the correct status for ten workflow that each contain ten parallel Mapping tasks until 90 seconds after the application services restart.
MWF-1340
If a workflow contains parallel Mapping tasks and you cancel the workflow while the Mapping tasks are running, any mapping that has yet to start proceeds to run.
The issue arises when the following conditions are true:
  • The number of mappings that the Mapping tasks specify is greater than the Maximum Worker Threads value on the Data Integration Service.
  • You cancel the workflow while the mappings that the Maximum Worker Threads value specifies are running and other mappings have yet to start.
Workaround: Increase the Maximum Worker Threads value.


Updated June 29, 2020