Cfxtrjtrk

I've been beating my head against the wall of the Salesforce Bulk API v2.0 for a couple of weeks now and I'm at my wits end.

I've been able to use Postman to test most of the calls and they appear to be working with correct values being returned. I can login, create a job, upload a file, close the job, and check the status all in Postman and it works correctly.

I can duplicate this code in C# and it all works as expected until I try to check the status of the job. At that point it returns this error, along with the normal fields.

"InvalidBatch : InvalidBatch : Field name not found : ", there is no field name listed as I have seen on other posts.

I've double checked the session token, the url, the http method, and the headers. Everything appears to be normal and matches what I'm using in Postman with the obvious values replaced.

This is the url I'm calling to return the job status, https://instance.my.salesforce.com/services/data/v45.0/jobs/ingest/jobID. This matches the documentation and all samples I've found.

Here is the JSON returned from this call, job id, created by, object have been removed for this post.

{

    "operation": "insert",

    "createdDate": "2019-03-27T19:08:46.000+0000",

    "systemModstamp": "2019-03-27T19:08:52.000+0000",

    "state": "Failed",

    "concurrencyMode": "Parallel",

    "contentType": "CSV",

    "apiVersion": 45.0,

    "jobType": "V2Ingest",

    "lineEnding": "CRLF",

    "columnDelimiter": "COMMA",

    "numberRecordsProcessed": 0,

    "numberRecordsFailed": 0,

    "retries": 0,

    "totalProcessingTime": 0,

    "apiActiveProcessingTime": 0,

    "apexProcessingTime": 0,

    "errorMessage": "InvalidBatch : InvalidBatch : Field name not found : --e52c6b94-9d37-4eda-b096-59d96b7e0cb5"

}

I've double, triple checked the column names I'm sending in the CSV file header record, they all match. Plus if they didn't, then Postman would also fail.

Our resident internet guru is also at a loss as to why this isn't working. Here is the code I'm using to return the job status.

client.DefaultRequestHeaders.Accept.Clear();

client.DefaultRequestHeaders.Accept.Add(new MediaTypeWithQualityHeaderValue("application/json"));

HttpRequestMessage request = new HttpRequestMessage(HttpMethod.Get, BulkEndPoint + ContentUrl.Substring(0, ContentUrl.IndexOf("/batches")));

Task<HttpResponseMessage> responseMessage = client.SendAsync(request);

Task<string> response = responseMessage.GetAwaiter().GetResult().Content.ReadAsStringAsync();

I've been through the Saleforce documents, tutorials, and many other C# examples. I haven't been able to determine the issue.

I've searched the internet high and low and I can't find anything resembling this issue.

The only thing I can think of is the Salesforce consultant wanted me to embed a CRLF into our address lines so the complete street address would be in one field. But I don't know why that would be ok when using Postman but not ok when running from code.

If any one has any ideas, I'd really appreciate it.

Edit: I can confirm is it not the CRLF embedded in the street address. I've swapped between CR, LF, CRLF, or no character at all, no difference.

Edit: To be perfectly clear the job is being flagged as failed as soon as I close the job. To do that call a PATCH with a body of { "state" : "UploadComplete" }.

Problem resolved!

{

    "operation": "insert",

    "createdDate": "2019-03-28T19:40:38.000+0000",

    "systemModstamp": "2019-03-28T19:41:34.000+0000",

    "state": "JobComplete",

    "concurrencyMode": "Parallel",

    "contentType": "CSV",

    "apiVersion": 45.0,

    "jobType": "V2Ingest",

    "lineEnding": "CRLF",

    "columnDelimiter": "COMMA",

    "numberRecordsProcessed": 883,

    "numberRecordsFailed": 809,

    "retries": 0,

    "totalProcessingTime": 11210,

    "apiActiveProcessingTime": 10847,

    "apexProcessingTime": 0

}

Just in case anyone else runs into this same issue. I didn't find this information anywhere, I just kept modifying the code until I got it working. The issue for me was how the CSV file was being uploaded to Salesforce.

This was my original code.

MultipartFormDataContent content = new MultipartFormDataContent()

{

    Headers =

    {

        ContentType = new MediaTypeHeaderValue("text/csv")

    }

};



client.DefaultRequestHeaders.Accept.Clear();



using (FileStream fs = new FileStream(UploadFileType.FilePath, FileMode.Open, FileAccess.Read))

{

    StreamContent streamContent = new StreamContent(fs);

    streamContent.Headers.Add("Content-Type", "application/octet-stream");

    streamContent.Headers.Add("Content-Disposition", "form-data; name="file"; filename="" + Path.GetFileName(UploadFileType.FilePath) + """);

    content.Add(streamContent);

    Task<HttpResponseMessage> responseMessage = client.PutAsync(BulkEndPoint + ContentUrl, content);

    var result = responseMessage.Result.Content.ReadAsStringAsync();

    string resultString = result.Result;

}

I'm guessing it didn't like the headers for the content, streamContent variables.

The new code, this still uploads the file, but doesn't fail when the job is closed. And as shown above, the return JSON from the job info no longer has an error message. This contradicts the examples I was able to find about uploading files to Salesforce using the bulk API v2.0.

// We don't want any weird headers hanging around

client.DefaultRequestHeaders.Accept.Clear();

client.DefaultRequestHeaders.Add("Accept", "application/json");



using (FileStream fs = new FileStream(UploadFileType.FilePath, FileMode.Open, FileAccess.Read))

{

    StreamContent streamContent = new StreamContent(fs);

    streamContent.Headers.Add("Content-Type", "text/csv");

    //streamContent.Headers.Add("Content-Disposition", "form-data; name="file"; filename="" + Path.GetFileName(UploadFileType.FilePath) + """);

    Task<HttpResponseMessage> responseMessage = client.PutAsync(BulkEndPoint + ContentUrl, streamContent);

    var result = responseMessage.Result.Content.ReadAsStringAsync();

    string resultString = result.Result;

}