ETL filling up OCR queue?

[mosea3]

In my company we just need full text search on PDFs that were already scanned and converted into Text-PDFs - so no OCR needed.
And OCR was disabled in /etc/opensemantic/etl and the ETL service was restarted

Still, something is filling the OCR queue and converting PDFs into images (connected to issue #343 )

Where can I backtrace this activity?

etl.txt

[Mandalka]

Is OCR yet enabled in the web admin / config ui?

This ui will write /etc/opensemanticsearch/etl-webadmin which overwrites settings in /etc/opensemanticsearch/etl

[schneipk]

I've got a similar issue. But for me I have OCR turned on. Running enrich later causes an error (seems to be deprecated) and some files/images of Websites get OCRd while others don't. Thank you for this cool project & the good work you're doing.

[phretor]

Same issue here. I tried the Desktop VM as well as the latest Docker Compose file. Worst is that I don't see any errors being thrown. @Mandalka do you see the same issue with the latest build?

[bmnnit]

i have the debian paket install:
ii open-semantic-search 21.12.25 all Search engine
and the problem that im unable to disable ocr, whatever i do there are always things added to the ocr queue..
im adding files with i.e:
opensemanticsearch-index-dir /home/opensemanticetl/mnt/Projekte/archiv/aktuelle_Projekte -v