Stuck OCR Queue

Troubleshooting Steps For Technician:

Answer the following questions:

1. How many items are waiting in the queue (if you are able to get an estimate)?
2. What is the file size (or average file size) of the files in the queue?
   a. Files that are too large may time out before they are able to be fully processed.
3. Was this a recently migrated account (from EFCO or Desktop), or have they used one of our migration tools recently to mass import items?
   a. Using Pangea or Pandora to migrate items in (either through a full migration or through just a regular data import) puts the files at a trycount of 2 in the system (so the mass import data doesn’t clog up the queue).
4. What is the trycount on the items that are in the queue? Are there any error messages on any of the files?
   a. If the items have a lower try count and have error messages, the error messages should give at least some indication as to why a file/set of files may not be processing.
5. Are there any similarities amongst the files that are in the queue? Were they all generated from the same place (like all scanned in, etc)?
6. Are any of the files in the queue ones that have their file versions updated frequently? How many file versions do the affected files have?
   a. If a user updates a file version on a file before it’s been processed through the OCR queue, it will add that new file version to the queue as the one that it wants to process first (and will not process the previous file versions until later), so if it’s something they’re updating regularly, it may be making it seem like it’s not processing, when in fact the previous versions of the file just haven’t been reached yet because it’s deferring to processing the newest version first.

If all troubleshooting has commenced and the issue is still present, grab the following information for the escalated ticket:

1. Provide answers to all of the above questions.
2. Provide the username of the affected user and provide their account ID (pulled from the Utopia Instances section of their account in the Hub)
3. Provide a sample of exact names of files in the queue (the more distinct the name, the better).
4. Video recording where applicable.