Some DI have completed with errors (some of the srs records are not linked to instance records) on Kiwi

Cornell is reporting an issue experienced in their production environment that they believe is the same as (or very similar to) https://issues.folio.org/browse/MODSOURCE-480

2 Data Import Jobs today 06/01/2022 (id = 51768 ended at 11:26AM EDT and id = 51709 ended at 9:40AM EDT)

have completed with errors (some of the srs records are not linked to instance records)

https://cornell.folio.ebsco.com/data-import/job-summary/06d2db9b-3713-4122-a286-8d3204f0b587?sort=recordNumber

https://cornell.folio.ebsco.com/data-import/job-summary/65c7162d-2d62-43b4-9fa1-b436cab0a4a8?sort=recordNumber

For the first job. (id = 51768) – a717336d-51e7-4192-b41e-9a93106bc9f4 is a problematic uuid

For the second job (id = 51709)
8b492fc1-6126-4e69-93ab-9f644f3af2bb and
c6e058e7-c115-429d-914d-9361deb3b2d9 are problematic uuids
 
Attaching – logs from mod-srs (~11:25AM where we observe Timeout errors) and mod-srm from same time (where we observe response code 500, Internal Server Errors)
 
Also attaching output from records_lb table for 1 off the. associated jobs in case it is helpful (record_state is OLD for 1 of the problematic UUIDs a717336d-51e7-4192-b41e-9a93106bc9f4)
SELECT * FROM fs00001034_mod_source_record_storage.records_lb where snapshot_id='06d2db9b-3713-4122-a286-8d3204f0b587'

Cornell is on kiwi –

• Source Record Manager Module (mod-source-record-manager-3.2.9)
• Source Record Storage Module (mod-source-record-storage-5.2.8)
  ​
  At the time of these DI import errors I did not see spikes in cpu/memory for DI. related modules
  Could you please assist in investigating why this failure is happening
  Thank You