This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(240, 33%, 33%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMG%3C/text%3E%3C/svg%3E)
I am configuring a pretty basic pipeline that Uses a GET Metadata activity -> Filter --> For Each (2 copy activities) running recursively.
Each time I try to debug the pipeline I get this error:
I have tried to configure this linked service without the fingerprint, however, the linked service is unable to test the connection successfully when removing the fingerprint.
Any assistance with this would be greatly appreciated!
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(54.400000000000006, 12%, 15%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESB%3C/text%3E%3C/svg%3E)
Hi @Maria Giugno ,
Please check this page for the recommendation from Microsoft to fix this error.
Try the resolution and let us know how it goes.
Thanks
This support page does not help, as I have already entered the accurate fingerprint the way it recommends. The fingerprint successfully validates through the Linked Service, however, fails in the pipeline run. Thank you for your help
Are you doing a debug run or using trigger now, @Maria Giugno ??
debug run
@ShaikMaheer-MSFT , any inputs??
@ShaikMaheer-MSFT This time an unrelated error is populating. The run is working for most files. We can close this request. I didn't modify any settings from last run so it's weird that it started working today. Regardless, thank you for your help!
Glad it worked. What's the other error you are encountering
Yeah, I think it is still weird that it was erroring out in the first place. I was wondering if maybe the sftp could be throttling the connection to data factory? That is just a guess of mine. The error that is showing now is because some of my files have a different column that is not the same as the other files. I was not aware of this until now. I am copying from an sftp to SQL with auto create table
Are you doing explicit mapping in copy activity ?
Yes....
There is nothing in the sink since it should auto create table. I even cleared the mapping and set the schema with a sample file adding in these columns: 
production_grade_code247 is a duplicate column. I removed it. Ran this pipeline three different ways: column name mapping, column number mapping, and with no mapping and a schema imported from a dummy file with all headers.
Hi,
Please post screenshot of your pipeline design, source and sink data set to narrow down the issue.
Thanks
@Suba Balaji the screenshots were too large for the comment- so I posted everything below in the Answers section. Thank you
All three pipelines are constructed the same, just with different mapping:
Lastly, tried no mapping and thought importing dummy csv with all of the headers at the source dataset.
All resulting in similar errors:
No mapping error:
With column name mapping:
With column number mapping:
Next course of action, we were thinking of building this as a data flow instead. We appreciate any feedback.
Thank you
Did you make sure there is no schema in the sink data set, when you don't do any mapping?
If there's a schema , clear it in the data set and try running your pipeline without mapping.
Also, try to create a simple copy (outside foreach loop) to see if that works, without mapping.
The sink schema is blank in each pipeline:
Source:
Ok, quick question , are you trying to use one copy activity to load different source data with different schema to different tables?
For instance,.
Src 1- 3 columns - shud be loaded to target1
Src 2 - 5 columns - should be loaded to target,2
But with one copy activity
Tried to execute the pipeline with no mapping
Not at all... we are copying from a file share that stores .csv(s) from a warehouse bank with the same schema to SQL on prem with auto create table. We have ran this pipeline successfully for two other banks. Even though I add ss_number to the schema at the source dataset, it still errors out during pipeline run. We have two separate copy activities in the For Each. One is responsible for copying from the stfp to SQL and the other copies from stfp to the archive folder.
Do you think a data flow would better achieve this?
Clear the schema from source data set too and run the pipeline.
It should work in copy activity itself. No need for a data flow.
Cleared the schema from the source data set and same error:
Did you check on the SQL database, by any chance the table is already there on the db?
Just published a video earlier today. Pls check if it's of help.
Thank you @Suba Balaji I make sure that I drop the table each time the debug run populates this error. I cancel recursive and then drop the table on the database side immediately following that. 
So did it work @Maria Giugno
nope... same error
I watched your video. Very educational, but not exactly what are we doing here. I am pulling from a secured FTP and copying to SQL on-prem and then copying from the SFTP/data to the SFTP/archive.
Can you check the copy activity by creating the table in the on prem sink and run the pipeline? This will help us isolate the issue
Here is the table created with all of the columns. The table is empty.
There is no schema set on the sink dataset. And I turned off the Auto create table since the table is already created.
Appears to be working now... Thank you!
Alright. Are we good then ? Or we need to use auto create table option.and Thanks
much for your patience.
Yes, we are all set! No worries - I appreciate your time and help!
