Flow Problems and Workarounds - Teradata Vantage

Teradata® VantageCloud Lake
Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
Language
English (United States)
Last Update
2024-04-03
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905
Problem Solution
Scheduled and continuous flows have "catch up" delays after database restart.

Sources continue to create new data, but Flow is unable to load it.

 Edit flows to increase microbatch sizes.

Select desired source, select Update from menu, and change number of checkpoint files: Update flow > Sources and Targets > Update > Advanced Options > Checkpoint files

For information about checkpoint files, see Creating a Flow Using the Console (AWS).
Block File System limits trigger Object File System spool, reducing throughput.
  • In large Parquet source files, use multiple rowgroups.
  • Compute cluster administrator can increase spool limit of flow owner, using MODIFY USER.

    Especially important if flow has multiple sources or user runs multiple concurrent flows.
Large number of bad record errors. Flow inferred inaccurate schema. Use schema hints. See Schema Inference.
Flow stops running due to error. Correct error and restart flow.

To restart flow, select FLOWS tab, select flow you want to restart, and from its menu, select Run.
Other See Database Query Log Tables and Views.