Stuck Background processing of Media imports...

brucek

I recently compiled and installed and setup MediaInfo, pdftotext, ffmpeg (with mp3), and dcraw.
I also setup and enabled background processing of media.

For the most part, things are working OK, but every now and then a PDF will get stuck, resulting in a stuck task listed. Sometimes the media is indeed not imported....sometimes the media seems fully imported, but still shows up as a stuck task.

I am NOT seeing anything in the php error logs, or /app/logs that helps.
Any idea where/how to find what/where is stalling the jobs?

Also, is there anyway to kill/remove stuck jobs without directly deleting them from the database?

Thanks

This is related to trying to get metadata extracted from imported media...where I am also very stuck

monica

Any errors on media processing would be in app/log or in PHP/web service logs if it's a php error. Depending on your system, there might be something in syslog too.

If you have version 2, then there is a way to resume a stuck process with the task queue dashboard widget, otherwise you'll need to manually run the process-task-queue caUtils with the force option on.

To extract file metadata from things
photos or videos, then you create an import mapping for it and set the options in embedded metadata extraction section of app.conf

See https://manual.collectiveaccess.org/providence/user/media/media_embedded_metadata.html
And
https://manual.collectiveaccess.org/providence/user/import/mediaembeddedmetadata.html

brucek

OK, thank you. I was not finding those pages when I was searching for extracting metadata.
This makes much more sense. I'll give it a try.

brucek

Update - The mapping is mostly working....in that the metadata is mapping into the ca_object_representations' record correctly. But in the Media representations browser, where you select and upload media, the [Metadata] always says "No embeded metadata was extracted from the media"
even though it obviously was extracted.