Vfs-chunk-size on initial Plex scan question

What is the problem you are having with rclone?

Not a problem, just a usage question

What is your rclone version (output from rclone version)

rclone v1.50.2-090-g0ecb8bc2-beta

  • os/arch: linux/amd64
  • go version: go1.13.5

Which OS you are using and how many bits (eg Windows 7, 64 bit)

Centos 7 minimal vm running in esx 6.5u1 hypervisor

Which cloud storage system are you using? (eg Google Drive)

Google Drive

The command you were trying to run (eg rclone copy /tmp remote:tmp)

N/A

A log from the command with the -vv flag (eg output from rclone -vv copy /tmp remote:tmp)

N/A

Hey all. Quick question. My understanding is vfs-chunk-size determines the size of the initial chunk downloaded upon file requests, then scales up to the chunk max value. The general consensus is 128M is a good value, and that's what mine is set to and works well for streaming.

My question is, if I'm loading in a library for an initial scan by Plex, does this mean it will grab 128M for every imported item since Plex has to read a portion of the file on import? If the answer is yes, is a lower value better for the initial scan? I'm noticing massive amounts of data usage on my cloud host for the scan. If my thought process is correct, does anyone have a recommended value for the initial scan, just enough to get the required data from the file in one chunk, not for streaming, and would this increase the initial scan performance and/or lower data consumption for the initial scan?

I can test various scenarios but didn't want to reinvent the wheel if the answers are already known.
Thanks in advance.

That's actually a common misconception. It's the size of the range requested from the backend. So if you use the default 128M and say you run a mediainfo on a file, it only downloads very small parts of the file as it requests, reads what it needs and closes it before it continues.

All in all, I'd just leave it at default as that's how I've always scanned with both Plex and Emby. It's more important in Plex to turn off the in depth analysis and such as that's the bandwidth consuming one.

If you want to make it less wasteful, you can definitely start at 32M or something along those lines and it ramps up fairly quick anyway if you are streaming so no harm, no foul.

We landed on 128M as the default as it's a good balance for everything but there is some inherit waste if you are only reading a few bits of the file.

If you want to test the various settings, you can always stop everything, isolate so only you are using the mount and tests various values and use mediainfo and see how running that impacts the bandwidth usage.

1 Like

Thanks very much! I do have Plex tuned for cloud mount (deep analysis off, thumbnails off, repeated full scans off, etc) and the scan is progressing just fine, I guess all the data is the metadata pulls since everything is new for this Plex instance. I guess next time I will just clone the VM, delete preferences.xml for Plex and add as new server.

Thanks again for the quick response, very insightful!

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.