Universal Media Server | Forum

Hello all

I just installed UMS (9.8.2) in /opt/ums/ums , as a service on my headless Debian server, on its own user, ums, whose home directory is /opt/ums.

As I was watching debug.log, I noticed UMS was scanning its user folder, including UMS folders

All my videos are on /srv/video, so I edited UMS.conf and wrote hoping that UMS would scan that folder and that one only

Restart UMS

And nothing shows on the debug.log, still scanning the user folder (even after I deleted .cache, .config, .mplayer folders)

I tried without the spaces around the equal sign, same thing

I ended up creating a symlink to /srv/video in /opt/ums, and then my movies started appearing (in the log first, and then as available on DLNA), alongside all the other files in /opt/ums.

What did I do wrong ?

Thanks

It could be many things, but there's something called "folders_monitored" too AFAICR which might interfere. BUt, one tip is to stop UMS before editing the config file, because UMS might overwrite your changes otherwise.

Whitespace doesn't matter, it will be trimmed anyway.

Thanks

Of course, UMS was stopped before editing the UMS.conf

Another common cause is that you're not editing the config file in use. The location of the "active" configuration file can be found relatively early in the log.

Nadahar wrote: ↑Tue Nov 24, 2020 4:32 am Another common cause is that you're not editing the config file in use. The location of the "active" configuration file can be found relatively early in the log.

YES!

That was it. I was editing the main UMS.conf file, I somehow missed the info saying I should edit the file in the ~/.config/UMS folder!

It works well now, thanks, except for a charset handling in folder names (/srv/vidéo does not work!).
I’ll look this forum up for this issue.

The configuration file is in "Java properties" format - which is encoded with Latin 1. To use "non-standard" characters, you must encode them with "\uxxxx" where "xxxx" is the hexadecimal unicode code point for the character. This means that you'd have to write: ..because 00e9 is the unicode code point value for é: https://www.compart.com/en/unicode/U+00E9

Thanks
(By non-standard you mean non-ASCII, because é is perfectly standard to me )

Yes, sort of. It's supposed to support Latin-1, which I think should include é, but assuming ASCII is absolutely safest.