Emby Rclone workspace Very long launch

In fact the auto mode of emby decides the quality according to your internet speed measured. Some devices do not carry the container, others yes, so there is often some transcoding, but generally more for a device that judges that your internet is not sufficient. but remember even in direct reading I observe this problem of random slowness

my pi4 would melt if it tried to transcode, sure it never happened yet.

what softwware do you you to stream, vlc or what?

roku, firestick, apple tv, smartphone, comecast tv, chromecast, ipad....

But on single device the unequal problem manifests itself for no reason, so yes I know that some devices do not support some container or the dts audio format, but in my tests the firestick reads everything so the random slowness cannot be explained

so far, this is a just a guessing game.

really need to use a debug log, and when the start of playback is very slow to look into that log.
and look into the emby log.

When I have a minute I'll show you the log :slight_smile:

Hello everything seems to be back in order for the speed of execution of launching of video files. On the other hand I have a question about the vfs cache max. Why once the limit is reached, my disk fills up so that the cache continues in my local disk. What is the point of this limit if it is not respected.

Thank!

rclone mount --allow-other --fast-list --cache-dir=/mnt/aeea3aba-b35e-4f7c-bf59-c3230b580a31/cache --dir-cache-time 1000h --syslog --poll-interval 15s --vfs-cache-mode full --vfs-cache-max-size 3000G --vfs-read-ahead 256M --vfs-cache-max-age 1000h Gcrypt: /mnt/cryption --user-agent="ams"

It's documented on the rclone mount page:

https://rclone.org/commands/rclone_mount/#vfs-file-caching

If using --vfs-cache-max-size note that the cache may exceed this size for two reasons. Firstly because it is only checked every --vfs-cache-poll-interval. Secondly because open files cannot be evicted from the cache.

When we talk about an open file, we are talking about an exact file being read? the pool interval 15s is 15 seconds of interval? If that's right, nothing can prevent the majority of my 4TB drive from being purged....

If you run in debug log, you can see it in the logs as it'll show you. If the file is in use when the poll interval is run, it won't remove it.

Well if only 1 file is in execution it does not delete anything at all?

I'm not sure what that means. If you have a log to look at, I'm happy to help as guessing or making assumptions on what may or may not being happening is not a good use of time.

ok I will activate the debug mode or I can also reduce the dir-cache-time 1000h to 240h.

thank you

what exactly is the line to enter in the command? --vv debug?
I am using this launch command and currently I am at 250gb of cache.

rclone mount --allow-non-empty --allow-other --fast-list --dir-cache-time 1000h --syslog --poll-interval 15s --vfs-cache-mode full --vfs-cache-max-size 200G --vfs-read-ahead 256M --vfs-cache-max-age 1000h Gcrypt: /mnt/cryption --user-agent="ams"

Thanks!

You can add -vv or --log-level DEBUG

felix@gemini:~$ rclone lsd GD: -vv
2021/06/19 08:52:40 DEBUG : Using config file from "/opt/rclone/rclone.conf"
2021/06/19 08:52:40 DEBUG : rclone: Version "v1.55.1" starting with parameters ["rclone" "lsd" "GD:" "-vv"]
2021/06/19 08:52:40 DEBUG : Creating backend with remote "GD:"
2021/06/19 08:52:40 DEBUG : GD: detected overridden config - adding "{TKSWb}" suffix to name
2021/06/19 08:52:40 DEBUG : fs cache: renaming cache item "GD:" to be canonical "GD{TKSWb}:"
          -1 2020-12-27 15:33:11        -1 backups
          -1 2019-09-29 21:57:58        -1 crypt
2021/06/19 08:52:41 DEBUG : 4 go routines active
felix@gemini:~$ rclone lsd GD: --log-level DEBUG
2021/06/19 08:52:50 DEBUG : Using config file from "/opt/rclone/rclone.conf"
2021/06/19 08:52:50 DEBUG : rclone: Version "v1.55.1" starting with parameters ["rclone" "lsd" "GD:" "--log-level" "DEBUG"]
2021/06/19 08:52:50 DEBUG : Creating backend with remote "GD:"
2021/06/19 08:52:50 DEBUG : GD: detected overridden config - adding "{TKSWb}" suffix to name
2021/06/19 08:52:50 DEBUG : fs cache: renaming cache item "GD:" to be canonical "GD{TKSWb}:"
          -1 2020-12-27 15:33:11        -1 backups
          -1 2019-09-29 21:57:58        -1 crypt
2021/06/19 08:52:50 DEBUG : 4 go routines active
felix@gemini:~$

You probably want to add a log file as well.

felix@gemini:~$ rclone lsd GD: --log-level DEBUG --log-file /tmp/rclone.log
          -1 2020-12-27 15:33:11        -1 backups
          -1 2019-09-29 21:57:58        -1 crypt
felix@gemini:~$ cat /tmp/rclone.log
2021/06/19 08:53:34 DEBUG : Using config file from "/opt/rclone/rclone.conf"
2021/06/19 08:53:34 DEBUG : rclone: Version "v1.55.1" starting with parameters ["rclone" "lsd" "GD:" "--log-level" "DEBUG" "--log-file" "/tmp/rclone.log"]
2021/06/19 08:53:34 DEBUG : Creating backend with remote "GD:"
2021/06/19 08:53:34 DEBUG : GD: detected overridden config - adding "{TKSWb}" suffix to name
2021/06/19 08:53:34 DEBUG : fs cache: renaming cache item "GD:" to be canonical "GD{TKSWb}:"
2021/06/19 08:53:34 DEBUG : 4 go routines active
felix@gemini:~$

Alright I'll use --log-level DEBUG
How long do I let it run like this and where will I find this log to send you?

thanks!

You'd want to:

  1. stop rclone
  2. enable debug
  3. recreate the issue
  4. stop rclone
  5. remove debug
  6. share the log

I added --log-level DEBUG and left rclone. I can take it out and start it again. But where do I find the logg?

Scroll up and read my previous post on the log file.

i don't find le logg in /tmp/rclone.log after add --log-level DEBUG, restart rclone, remove --log-level DEBUG and restart rclone