Overview:
File stuck in Data Import app when importing large amount of records (f.e., 10k MARC Bib records)

Steps to reproduce:

1. Login to evrk2

2. Go to “Data import” app

3. Click on “or choose file“ button

4. Select a file with thousands of MARC Bib records (you can use the attached .mrc file with 10k records)

5. Select “Default - Create instance and SRS MARC Bib“ job profile

6. Click on “Actions” in the third pane → Select “Run“ option

7. Click on “Run” button in appeared modal

Expected result:

• After some time, several log rows for chunks (parts) of the total amount of the uploaded records appear in the second pane.

• All rows have “File name” value equal to name of the uploaded file with digit postfixes (“-1“, “-2”, etc.)

• All rows have “Completed“ status.

• The sum of “Records“ values for all rows equals to the total amount of uploaded records.

Actual result:

• Some of the rows have “Completed with error“ status.

• When such logs opened, if you click on the record titles in the log, you can see on of the the following error messages on “Incoming record“, “SRS MARC“, “Instance“ tabs:

  • io.vertx.core.impl.NoStackTraceThrowable: {"errors":[{"type":"UnauthorizedException","code":"authorization_error","message":"Unauthorized"}],"total_records":1}

  • No action

• The import job remains shown in the first pane in “Running“ accordion until it is manually stopped. The progress is stuck on a values < 100%, and some job parts are not finished.

• Sometimes, further import jobs (with lesser amount of records) are then also stuck in progress. This is NOT consistently reproducible

Additional information:

• Was able to import 10k MARC bib records on Snapshot (Note: file split is not turned on there, but turned on on evrk2)

• Also reproducible on FSE Dry run 1

• Related test case - https://foliotest.testrail.io/index.php?/tests/view/3034890

• See example: