[Feature] Failure log / info stored per failed download and accessible from web

[KashifZA]

Is it possible to capture a log of the download and store in history if the download fails?

So i tried a few downloads and worked.. then i tried a few that failed (not youtube) but couldn't tell why. Would be great for troubleshooting as not all video sites work with Youtube-dl and you where the default fails, you need to supply add a extension.

Additionally if the site requires a username/password, you won't know that the login fails as there is no log to indicate it.

[renamedquery]

Sure thing! I can try to add this, I may implement a momentary fix which is something like running the app as gunicorn3 --workers 4 --bind 0.0.0.0:80 wsgi:app >> log_x_x_x.txt but I do realize that a proper log would be important. Will add this soon.

[KashifZA]

Great, will mod my service for now so I can figure out why it currently seems to fail for any non YouTube link e.g Vimeo. I looked through the code but think the log would help figure what’s happening.

On Thu, 30 Jul 2020 at 21:21, Harrison [email protected] wrote:

Sure thing! I can try to add this, I may implement a momentary fix which is something like running the app as gunicorn3 --workers 4 --bind 0.0.0.0:80 wsgi:app >> log_x_x_x.txt but I do realize that a proper log would be important. Will add this soon.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/katznboyz1/youtube-dl-server/issues/3#issuecomment-666627284, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADHVVM467VTPML67T4B2AM3R6HB3TANCNFSM4PN45STA .

-- Sent from my iPhone.

[renamedquery]

Hm for some reason github didn't alert me to the new message on this thread, sorry for the late response. I was able to fix the issue with non-youtube links, it had to do with an issue where the program was fetching metadata that didnt exist, like the author of the video, as this is not available for most non-youtube sites. Unfortunately since there are a bunch of try/catch statements to prevent errors, the issue wasn't that obvious until they were removed. The issue with non-youtube links is currently fixed, and there is also a warning sent to stdout for failed logins. I will try and come up with a way to handle the logging soon.