Forum Discussion

aidelasks's avatar
aidelasks
New Member
1 month ago
Solved

Power BI Service Refresh Error: "We cannot convert the value '[Table]' to type Table", Desktop works

Hi everyone,

I'm running into a Power BI Service refresh issue that I can't seem to solve.

 

The Error

The dataset refreshes successfully in Power BI Desktop, but fails in Power BI Service with:

 

Expression.Error: We cannot convert the value "[Table]" to type Table.

Details:
Reason = Expression.Error
ErrorCode = 10277
Value = [Table]

 

The frustrating part is that Desktop works perfectly, so this appears to be a Service-only evaluation issue.

 

My Query Structure

I have a query that:

  1. Loads a custom data collection table.
  2. Filters active parameters.
  3. Joins to a parameter mapping table.
  4. Groups source records by parameter value.
  5. Uses a custom function (fnExpandMappedParameter) to dynamically expand parameter data.
  6. Combines all returned tables into a single table.

A simplified version:

Source = base_table_custom_data_collection,

FilteredParams =
Table.SelectRows(
dim_parameters_tablenames,
each [Parameter Value] <> null
),

JoinMapping =
Table.NestedJoin(
FilteredParams,
{"Parameter Table Name"},
dim_newpara_w_tables,
{"Parameter Table Name"},
"Mapping",
JoinKind.LeftOuter
),

GroupedSource =
Table.Group(
Source,
{"Item"},
{{"Data", each _, type table}}
),

Joined =
Table.NestedJoin(
JoinMapping,
{"Parameter Value"},
GroupedSource,
{"Item"},
"Matched",
JoinKind.LeftOuter
),

AddTables =
Table.AddColumn(
Joined,
"Result",
each
let
tbl = [Matched],
mapping = [Mapping],
param = [Parameter Value]
in
if tbl = null or Table.IsEmpty(tbl)
then null
else fnExpandMappedParameter(
tbl{0}[Data],
param,
mapping
)
),

Combined =
Table.Combine(List.RemoveNulls(AddTables[Result]))

 

What I've Already Tried
1. Table.Buffer()

I buffered both source and parameter tables, no change.

 

2. Verify Nested Tables

I checked the columns involved, both appear as nested tables in Desktop. No obvious issues appeared in preview.

 

3. Expand Mapping Earlier

I tried expanding the Mapping column, however, downstream logic expects: mapping = [Mapping], which no longer exists after expansion. This didn't solve the refresh issue and likely breaks the original function design.

 

4. Considered Moving Logic to SQL

I considered moving the transformation to SQL Views or Stored Procedures. However, the solution depends on multiple parameter tables, dynamic mappings, custom functions and dynamic column generation. So it doesn't feel like a practical SQL rewrite without substantial redevelopment.

 

I'm wondering if Power BI Service is evaluating one of the nested table objects differently than Desktop. Perhaps somewhere a nested table is being interpreted as the text "[Table]" rather than an actual table object.

 

The part I'm most suspicious of is the below, since this is where nested tables are being passed into a custom function:

fnExpandMappedParameter(
tbl{0}[Data],
param,
mapping
)

Has anyone encountered a Desktop vs Service issue where a nested table becomes the text value "[Table]" during refresh? Is there a better architectural approach for handling dynamic parameter expansion like this that avoids nested table evaluation issues? 

Are there known limitations with some of the syntaxes used, that behave differently between Desktop and Service?

 

Any suggestions would be greatly appreciated. I've spent quite a bit of time troubleshooting this and have reached the point where Desktop works consistently but Service continues to fail with the same error.

 

Thanks!

  • Its usually happens due to Privacy level of your data source, update the web privacy level same as your desktop file and you should be good

6 Replies

  • arvindsingh802's avatar
    arvindsingh802
    Community Champion

    Its usually happens due to Privacy level of your data source, update the web privacy level same as your desktop file and you should be good

    • aidelasks's avatar
      aidelasks
      New Member

      Hi Arvind, great shout but the problem is still unresolved. I have set both as 'Organizational' and it is still failed in service.

  • Hi aidelasks 

    Ensure to match privacy levels for data source in Power BI Service dataset settings (Settings - Datasets - Data source credentials - Privacy level) since Desktop and Service can evaluate nested table firewall rules differently

    • aidelasks's avatar
      aidelasks
      New Member

      Hi Krishna, thank you, butthe problem is still unresolved. I have set both as 'Organizational' and it is still failed in service.