I didn't find info about which layer implements the options like --vfs-cache* ... that are passed to mount, if you run in a layered crypt > cache > remote (B2/S3) mode.
The question is:
which layer actually uses the "--vfs-cache" etc.. options, if you mount the crypt wrapping layer which wraps the cache, which in turn connects to the main remote gateway layer.
But if you mount the crypt, which remote actually uses the option "--vfs-cache" for example ? The crypt, which you mount or the remote it self (GD in your case) which is mounted/wrapped by the crypt?
I did answer your plain question as it is whatever remote you mount. You can pass any options to the mount. You mount remotes. In your config, you can point a crypt to any remote regardless if it's B2 or Google Drive. If you have a specific parameter, you can use it on the mount command.
fast-list doesn't work on a mount.
If you want to use --vfs-cache-mode writes as that's a generic one, just pass it on the mount.
allow-non-emtpy is an awful parameter as I'd never use that as it allows hiding things and over mounts.
Please, say yes or no.
Do options like "--vfs-cache-mode writes" for rclone mount command are used by the crypt remote or by the backblaze remote, if mounted in crypt >> cache >> backblaze scenario.
In other words if I mount the crypt remote, does it performs the vfs caching, or it passes this option further to the cache remote, which passes it to the backblaze remote?
Thank you.
Whatever you mount is doing the "caching". If you run a mount command such as:
rclone mount cryptremote: /Test --vfs-cache-mode-writes
That is doing the vfs-cache-mode writes. The caching is mount relevant and not relevant to the remote as you are telling the /Test mount to cache. It doesn't matter what the remote is.