tSalesforceBulkExec: missing salesforce_id / error in output

[FrancoisP_CH]

Hello,

We upgrade this summer our Talend Cloud Studio from 7.3.1 to 8.0.1R2026-06.

I'm trying to create a new job to put data in Salesforce using the tSalesforce bulk components, and specifically the tSalesforceBulkExec.

I put it on the designer, select the existing connection, choose the module to update and import the schema from my tSalesforceOutputBulk in the "Schema".

When I use the MAIN and Reject output of the tSalesforceBulkExec (or check the "Schema MAIN" & "Reject Schema" in Basic Settings), I have the input schema, but the following automatic columns are missing : 

• MAIN : salesforce_id & salesforce_created
• Reject : error

Is this a standard behavior or a regression ?

I try to click on the "Guess schema", but I got compilation error for this specific action (the job compile without error)

Edit Sept 07 :

It also seems that it does not handle well when one/some fields contains lines returns. When everythings was alright in 7.3.1

We generate the file with a tFileOutputDelimited (with a specific configuration) and use the API Bulk V2 to avoid this problem, but the global values NB_SUCCESS / NB_REJECT / NB_LINE aren't set in V2.

Edit Sept 08

The issue with the line return has been handle in the R2026-08 Studio patch, according to the associate release note update.talend.com/Studio/8/updates/R2026-08/PATCH_RELEASE_NOTE.html

• QTDI-3147 - tSalesforceOutputBulk: double-quote characters in field values not escaped in generated CSV after R2026-06 TCK migration

I have been able to automatically retrieve the technical output columns be clicking on the "Guess schema" bouton. Its more or less indicate in the documentation, section "Limitation" : tSalesforceBulkExec | Talend Components for Jobs Help

It lets us with "only" the global var issue when we are in Bulk V2, which isn't the subject anymore. I close this topic as resolve.

[craig42son]

Hello,
Unlike standard real-time output components (such as tSalesforceOutput), which process records synchronously and can immediately inject tracking fields like salesforce_id or salesforce_created into the flow, tSalesforceBulkExec operates asynchronously via the Salesforce Bulk API. It takes a pre-generated CSV file (from tSalesforceOutputBulk), submits it as a batch job, and handles processing server-side.
TPT login

Best Regards

[FrancoisP_CH]

Hello. Thanks for the reply. I'm aware about the specifities of the bulk components, I used them for a while now. The missing fields are send in response by Salesforce, even in bulk mode. And it worked fine until we migrate from 7.3.1 to 8.0.1R2026-06.

[FrancoisP_CH · Accepted solution]

The issue with the line return has been handle in the R2026-08 Studio patch, according to the associate release note update.talend.com/Studio/8/updates/R2026-08/PATCH_RELEASE_NOTE.html

• QTDI-3147 - tSalesforceOutputBulk: double-quote characters in field values not escaped in generated CSV after R2026-06 TCK migration

I have been able to automatically retrieve the technical output columns be clicking on the "Guess schema" bouton. Its more or less indicated in the documentation, section "Limitation" : tSalesforceBulkExec | Talend Components for Jobs Help

It lets us with "only" the global var issue when we are in Bulk V2, which isn't the subject anymore. I close this topic as resolve.