Licencing question and feature request

[EelcoHoogendoorn]

I have a question about licencing; it is not quite clear to me what licence would be required for my situation, os let me describe that first.

I have a home and work computer that I would like to keep synchronized using some server (ftp, cloud, whatever). Furthermore, both home and work computers have an SSD and a HDD; the HDD contains a core of data that I would like to keep perfectly mirrored over the server, and I want to be able to 'check out' whatever active projects I am working on to my SSD.

After a lot of searching, it seems that SFFS is the best tool to archieve all these goals. But would I require two licences for this simple setup, just because it involves my home and work computer? My work and home computer shouldnt be booted at the same time, but can I install SFFS on both at the same time using a single licence? What if I forget to turn off my home computer; will SFFS refuse to start at work? The webpage in not very clear on this.

With regards to the checkout-type synching I want between my SDDs and HDDs, there is a feature that I would like to suggest: I want a simple way to stop mirroring a given folder. Right now, I suppose the best way to disable checkout of a project folder is to stop synching it, then manually delete it from my SSD. But the robust thing to do would be for the checkout-folder on my SSD to be automatically swept clean of everything not included in the list of checked folders, so all youd have to do is manage the folder selection from within SFFS. All the behaviors that are expected of such a scenario could be bundled under one new type of synching mode.

There are probably more small things that could be done to facilitate this scenario that I cant think of right now; it could be another nice selling point for SFFS, because I dont see any of the competetion doing this.

[superflexible]

Hello,

you can use a Single User license for this. The program at work will not refuse to run even if you still have it open at home.

Deletion of deselected folders is possible, but only in a one-way synchronization. Probably you have to sync both ways, so right now we don't have a feature for that. I will keep it in mind!

Cheers,
Tobias Giesen

[EelcoHoogendoorn]

Thank you for your prompt reply, and good to hear about the licencing.

Indeed I have to sync both ways, but your suggestion does seem to raise the possibility of chaining two one-way profiles to archieve the intended effect of a 'checkout sync'; but perhaps that would create conflicts that I am unaware of?

Let me think; so I sync a subset of HDD projects one-way to my SSD. Files I add/modify on my SSD are not written back to the HDD. If I add another undiscriminating one-way sync from SSD to HDD, what would happen if I delete a file from SSD? Would it get stuck in some sort of endless file-deletion-creation loop? As long as these sync profiles never run at the same time, I think it should all be ok? Or would that still confuse the database system? Actually I dont need the versioning and whatnot at the HDD->SSD level, so a 'dumber' and less likely to break sync would be fine too I think..

I dont want continuous crawling of my disks anyway, since I find it interferes with compilation processes; so I want to pull from the server and then HDD->SSD upon booting and do the reverse at shutdown, and then perhaps periodically when my computer is idle to prevent unnecessary hanging at shutdown. Either way there is not a lot of traffic, and its only for a single-user setup, with data only going in one direction at a time, so I think the two one-way sync profiles 'shouldnt' have to get in eachothers way without compromising the desired functionality.

I could figure all this out by trail and error of course, but id rather not discover afterwards ive been corrupting my data with some unsupported synching voodoo, so if you have any more thoughts on the subject, they are very welcome!

[superflexible]

I don't see any problems there. If your jobs are defined correctly to achieve your goal, then there are no hidden issues with the database or anything.

A database is only used for two-way SmartTracking syncs anyway. Most one-way jobs don't need it.

[EelcoHoogendoorn]

A minor update; ive rigorously tried the various options, and it seems the desired behavior can not be produced.

That is, if I make the SSD drive a mirror of a subset of folders on the HDD, and add another profile to copy results back from the SSD to HDD, I run into the problem that this second syncjob can not distinguish between deletions on my SSD due to wanting to get rid of a file, and deletions on my SSD due to reducing the subset of selected folders; they get propagated back to my HDD. Not propagating deletions is not desirable; then id rather manage the synchronicity between deselelecting and deletion by hand.

A two-way mirror does not exist, and a smart-sync does not delete files from the target upon reducing the set of selected folders.

If that functionality becomes available, youd be my hero; especially if it comes with explorer integration where i can check/uncheck wether or not any given subfolder is 'checked out'.