OCR queue gives no result

[Rolf Schneider]

Dear Community,

I am looking for an EMDS solution, and Mayan appears to be what I need.
However, the main reason why EDMS are so practical is OCR. Which is
what does not works in my case (Ubuntu 16.04, Mayan 2.2).
I have installed tesseract, and it works nicely in the terminal:

tesseract filename.png stdout

gives me the extracted text to the terminal.

But when I submit documents to the OCR queue there is no response at all:
no error message,
no output,
no OCR activity (checked with top in terminal)

I would really like to fix this issue. Can anybody help? Maybe post his or her wrking tesseract.py backend?

Thanks

[Roberto Rosario]

Do you have the background task queue workers working? Otherwise background tasks like OCR won't run.

Did you installed by hand or by using the Docker image?

To see if tesseract of scanning, send a big document to OCR and do the following on the command line:

ps x a | grep 'tesseract'

you should see a process running tesseract if not check the "OCR errors" log in the "Tools" section to see if an error message is being logged.

Example: http://demo.mayan-edms.com/ocr/all/