Cache chunk location on an NFS share

#1

Is there anything I should be aware of? I have my current chunk location and DB set on an NFS share like so:

Cache DB path: /nfs/cache/cache-backend/tcache2.db
Cache chunk path: /nfs/cache/cache-backend/tcache2

This is a crypt, wrapping a cache, wrapping a remote:

[tcache2]
type = cache
remote = tdrive2:
chunk_size = 10M
info_age = 2d
chunk_total_size = 1500G
db_path = /nfs/cache/cache-backend
chunk_path = /nfs/cache/cache-backend
workers = 8

[tdrive2]
client_id = xxx
client_secret = xxx
type = drive
token = xxxx
team_drive = xxxx

[tcrypt2]
type = crypt
remote = tcache2:
filename_encryption = obfuscate
directory_name_encryption = true
password = xxx
password2 = xxx-xxxx

The mount command:

/usr/bin/rclone mount tcrypt2:encrypted /mnt/cache --dir-cache-time=160h --cache-info-age=168h --buffer-size 32M --allow-other --allow-non-empty --cache-workers=8 --attr-timeout=1s --syslog --umask 002 --drive-chunk-size 32M --log-level INFO

However I’m not sure the cache is being used properly. The NFS share is mounted with default options (async) and if I start a video and stop it and then restart it I don’t see any network activity from the box serving the chunks. I assume those are still in memory (how often do they get written)? Is there anything I can look for in debug logs that’d tell me that chunks are being committed to disk?

Thanks!

0 Likes

#2

–buffer-size should be 0M with cache as it uses its own mechanism for the chunks so that’s just waste.

You should see the cache reading chunks in the logs if you have debug on.

0 Likes