Rclone deletes with filter hangs

I am executing a delete using a filter-from on a cache with crypt, e.g.

rclone --filter-from /mnt/disks/media/format.txt delete .

Some times this will delete the files successfully, othertimes it will hang part way through and there will be no messages in the log. This may even lock up the whole file system so I need to reboot the box.

Is this to be expected when other applications are connecting to the crypt so should I be doing these deletes using a duplicate crypt?

I am trying to find a way to automate deletes using a pattern rather than through laborious manual deletes which appear to be working successfully (the regular linux rm seems to be working successfully).


Are you deleting things in an rclone mount?

Is it that which is freezing?

Yes, it is a mount I am deleting from and is freezing

You can delete straight from the remote. Using the mount just adds latency and an extra layer which isn’t needed.

Ah, okay because with rmdirs I found deleting straight from remote resolved the issue.

I just get a little confused because some things you can do direct on the remote but others like copying or moving will give an error that the db is in use and obviously its more convenient to a delete . (local path) than explicitly enter the full remote path but at the end of the day its the better approach.


For me, I mainly do everything with commands except playing and maybe renaming files.

Adding new files or if I want to do mass deletion, I find that using the proper commands to be faster.

I don’t think anything hangs per se, but it will take some time to look down the full path and build up the cache so that looks like a hang while the command processes.

That gets worse if you have a lot of files in a particular area or something along those lines as it just takes time.