Mongo db mirroring in fabric throwing 'Running with Warnings'

[rsheltondavid]

Hello Team, 

We tried to mirror the mongo db into fabric and the set up is done but we've some tables that says 'Running with Warnings' and it is because of the datatypes in the same column. 

ErrorCode: SchemaMergeFailure, Message: Failed to write table 'dbo\opportunity' due to schema mismatch. Schema merge failure due to potential invalid input. Cause: Type mismatch for column 'last_access'. Incoming type: 'string' (Nullable: True), existing type: 'double' (Nullable: True). ArtifactId: a38c79cb-7743-4e31-ad22-baebeda554a9

Knowing that mongo has now become compatible for mirroring, Is there a way to fix this issue? 

We follow this github link for setup: GitHub - mongodb-partners/MongoDB_Fabric_Mirroring: Code to enable mirroring in Microsoft Fabric for...

[v-saisrao-msft]

Hi @rsheltondavid,

The error appears to be caused by a data type mismatch. Please verify that the column’s data type is consistent. After correcting it, try rerunning the database mirroring to see if the issue is resolved.

Thank  you.

[rsheltondavid]

Hey @v-saisrao-msft 

Is fixing the source the only option or can we edit the source code before the mirroring to store the column as strings? We used this to set up the mirroring:  GitHub - mongodb-partners/MongoDB_Fabric_Mirroring: Code to enable mirroring in Microsoft Fabric for...

[v-saisrao-msft]

Hi @rsheltondavid,

Checking in to see if your issue has been resolved. let us know if you still need any assistance.

Thank you.

[v-saisrao-msft]

Hi @rsheltondavid,

We haven’t heard back from you in a while regarding your issue. let us know if your issue has been resolved or if you still require support.

Thank you.

[v-saisrao-msft]

Hi @rsheltondavid,

Have you had a chance to review the solution we shared earlier? If the issue persists, feel free to reply so we can help further.

Thank you.

[v-saisrao-msft]

Hi @rsheltondavid,

Kindly check if there is a schema mismatch for the last_access column in your MongoDB collection. Check whether the values are stored as strings, doubles, or a combination of both. It’s best to store all values as strings to support both numeric and date formats. After the necessary changes for consistency, please restart the database mirroring process.

Thank you.

[rsheltondavid]

Hey,

Thanks for getting back to us. We've a mix of both on our mongo db end but the source is so large that we cannot change the source as yet. We've several columns with a combination data type in different collections. The data is landing fine as parquet files. The issues is only with the sync to the table. What would you suggest in such a scenario?