I have been using samba to stream and like you said I totally agree with your assessment of network delay.
But it was has been dead stable for me even streaming from rclone, with never a dropped connection, but because of how there is no buffer nothing is smooth.
What about nfs? I used this script btw: GitHub - angristan/wireguard-install: WireGuard VPN installer for Linux servers
Well thanks for that info. Guess I'm out of luck. I guess setting a very low or 0 dir cache time should resolve this?
For me it wasn't buffering as I simply hadn't gone to that stage yet.
mergerFS recursing the mount was taking over 70s instead of 6s it normally takes, and the entire mount slowed to a crawl. I could see in htop that it would wait on IO (D) very frequently.
Apart from that, systemd behaved weird with mounts, so I just wanted to ditch this ancient protocol.
not sure your use case but running rclone vfs/refresh will pre-cache the dir cache.
for example, if i add some media files to my mount and now i need emby to scan that.
i run that command first and then have emby rescan the library.
I enabled that bit. My use case for this mount is simple. *arr sends stuff to the remote client, imports it (moves server-side), then a script moves it from the server to the cloud and behind the scenes it simply changes mounts but since all of them are in a single mergerFS mount, it makes no difference to Plex.
yeah, i see that in the systemd file...
try a low value for --dir-cache-time
cannot help with *rr, mergerfs, as i run a very simple media server.
i do not know,
as i mentioned, @Animosity022 is the expert at linux, merger, *arr, plex and so on..
perhaps he will choose to stop by here in this topic...
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.