Data source error: Not a legal OleAut date.. The exception was raised by the IDataReader interface.

[ganaa0810]

Hello guys, Im new here.
I Configured everything and checked every date columns in my table which might out of zone or with time with zone. Its returned 0 so i think the data itself is okay and native query is folding, The on-premises gateways is up to data what other would be case ?
I set incremental refresh only on one table so focusing just on one table.. Still gettin errors Any other way to check exactly what is caussing error there

{ "timeStart": "2025-11-10T07:08:13.4450000Z", "timeEnd": "2025-11-10T07:22:29.7900000Z", "durationMs": 856346, "externalQueryExecutionTimeMs": 967260, "vertipaqJobCpuTimeMs": 80641, "mEngineCpuTimeMs": 63, "queryProcessingCpuTimeMs": 219, "totalCpuTimeMs": 294469, "executionDelayMs": 18, "approximatePeakMemConsumptionKB": 1365561, "mEnginePeakMemoryKB": 159044, "tabularConnectionTimeoutMs": 17999000, "commandType": "Batch", "discoverType": 129, "queryDialect": 4, "refreshParallelism": 6, "vertipaqTotalRows": 17985569, "intendedUsage": 2, "errorCount": 14 }

[v-ssriganesh]

Hello @ganaa0810,

Hope everything’s going great with you. Just checking in has the issue been resolved or are you still running into problems? Sharing an update can really help others facing the same thing.

Thank you.

[v-ssriganesh]

Hi @ganaa0810,

Thank you for posting your query in the Microsoft Fabric Community Forum, and thanks to @lbendlin & @ibarrau for sharing valuable insights.

Could you please confirm if your query has been resolved by the provided solutions? This would be helpful for other members who may encounter similar issues.

Thank you for being part of the Microsoft Fabric Community.

[lbendlin]

Incremental refresh requires datetime or dateinteger values.  You provided datetimezone values.

[ibarrau · Accepted solution]

Hi. How are you checking for valid dates. That error means you are using a wrong date, it could be something like 22/22/0100 with wrong month or year. You can check it re ordering by that date, filtering for keeping errors or even using a custom function like: 

= try Value.Is(Date.From([Date]), type date) otherwise false

Do a check for every date column in the semantic model. Reordering usually shows a wrong value when it's too high, too low or totally wrong. But if the issue is "10/22/2025" instead of "22/10/2025", that would be found with keep errors filter after converted to date.

I hope that helps, make sure after you refresh the whole dataset of desktop, you don't have any warning.