Well, regardless, it’s there and it makes sense in this context.
Thanks, I managed to make it work, took a bit of time figuring out how to set up parameters. I’d also say that it works good enough for me, any other improvements can be easily made on top of this work if necessary. One last thing that I noticed is that though now unmount seems to work fine on exit, I still get this in the console:
/bin/fusermount: failed to unmount /mnt/OD: Invalid argument
I am running Debian if that matters.
Wow, I am kind of shocked by how quickly you managed to write up an example implementation of a mount command @david.rios.gomes . Really impressive stuff. I’m going to try to download the build and test it out later this weekend if I have time but this is obviously an excellent start to a complete, official mount implementation. Given that you have familiarized yourself with the Duplicacy code and are familiar with implementing a FUSE file system for it, perhaps you could give a bit of insight into a couple of questions I have? If not, no big deal.
How practical would it be to use something like this (either in its current form or a future, optimized form) in a semi-permanent fashion? For example, many people currently use rclone’s mount command for mounting cloud storage to a local file system for the purposes of running a Plex server. Do you foresee something like that being practical using this instead of rclone? That is, mounting a media folder found in a specific snapshot/revision and using that mount for Plex or something similar?
How difficult and/or practical do you think it would be to implement some type of write support into a future implementation of this mount command? For example, mount a specific revision and then any writes on top of that revision are written back to the online storage every x number of min, possibly with each accumulated write session having its own revision #? That doesn’t sound like too elegant of a solution but I’m not sure how else you would do it… Perhaps there is another approach I’m not considering?
Anyways, thanks for all the hard work and effort you put into creating this FUSE/mount implementation. I believe this alone is a huge step forward for Duplicacy!
Yeah, I tried but couldn’t get rid of that message. Seems like it either tries to umount twice or not at all, so twice it is.
Hey.
I think that’s probably already OK with the current implementation. It just doesn’t automatically refresh the repository, so if new revisions are added you need to stop and start the mount command again, but overall I think the current implementation is already stable enough for that.
Write support is very unlikely given the nature of the tool itself. Probably never going to happen.
1 Like
I see this wasn’t included in the newest release, though everything looks good at the pull request, as far as I can tell. Do we know anthing about the status?
IMO, the implementation is far from stable and properly tested, and the command line usage needs a lot of rework to make it consistent with current syntax. The memory usage was insane in my testing, when compared to Rclone mount, for example.
Hi,
Has this been worked upon since last reply? I’m using the webUi and cannot see this option.
Hello, no, still not implemented in the GUI, and not in the latest CLI version either.