We have a situation that has occurred for quite a few years now. I've been in contact with Support a couple of times and tried to work through the problem with our IT dept and haven't found a solution, so I wanted to reach out to the DataTransfer Community to see if anybody else has experienced this. When we have an auto-job/Recurring job configured to pick up XML files from a network share and process the job within DataTransfer, the job will go to an INCOMPLETE status if multiple files have been sent at the same time. However, the job has actually processed successfully in our SIS (Banner). There is a connection back and forth between DataTransfer and Banner where it's sending the information, Banner is accepting it and then should send back a token to complete the transfer. What is happening, or what has been explained, is that the network connection to complete the process is being dropped. When a lot of jobs are being sent through at the same time and DataTransfer is processing those or one right after the other, what we think is happening is that Banner is going into a protective mode (like a DoS attack of sorts?) and stops responding which means it doesn't send back the response to 'complete' the job process. Then those jobs sit in the Active Dashboard as Incomplete. We have to manually hit the run button on the Dashboard to get them to complete the process and clear out. I've run into dead ends on our IT side with our DBA team and Networking to find a solution. Has anybody else experienced this or have any suggestions on what to check? This only happens with Recurring jobs and not every time, just when a lot of files have been sent - or DataTransfer picks up multiple jobs to run at the same time.
Hyland Support: "PDT logs suggested that connection pools that were in use were dropping or becoming stale, and wireshark logs showed packet loss during higher traffic."
**Edited: We are currently on version Hyland DataTransfer version 7.0.1_b1555**
