piotrnik

Do you have DHT enabled for any of your shares? I've seen odd peer connections when I have it enabled on some of mine.

Just to clarify, can you check my understanding of this following scenario? Say I currently have 3 devices with the same r/w key. I create a new approval request link on device A. Only device A will see the approval request once new device D adds the folder, and then has to approve it before new device D gains access to the folder and the files start syncing. Devices B and C would never know about the approval, and would just see the new device D once A approves it. Is this correct?

There are invisible column lines that you can use to expand/contract the columns. They're hard to find though - you just have to hover your mouse until you find them.

I would also appreciate this, though it's not critical for my use case.

Yay for 1.4 - http://blog.bittorrent.com/2014/08/26/introducing-bittorrent-sync-1-4-an-easier-way-to-share-large-files/ Looking forward to trying it out!

There are encrypted secrets that can be used with a hosting provider or such - the BTSync installation with an encrypted secret has only the encrypted files (I believe with or without the file tree, depending on secret type). To retrieve the data, you must connect with another computer having the normal read-only or read-write secrets (ie they will download and decrypt it). I don't know all the details of this, or the types of encrypted keys, but I do know that it's possible and used by quite a few people with untrusted hosts (be it a hosting provider or a friend/family member that they don't want able to see/change their files).

I could see something like this being useful, though it's really not that hard to convert KB/s to Kb/s (multiply by 8). Additionally, depending on your ISP, there is another reason that this could be of little use - at least in the US, the vast majority of ISPs don't provide what users have signed up for (most of the time it's not even close). It's not uncommon for a 20 Mb/s tier to only get 1.5-2 MB/s download speeds rather than the 2.5 MB/s that should be coming from a 20Mb/s connection (I have several friends with different ISPs that have this problem, some worse than others). "Real-world" speeds seem to vary by day and time, and can be considerably slower than the advertised speed. Dishonest or unethical? Sure, but it's also how it is. The problem resulting from this is that if you set the speed limits based on your ISPs advertised speed, you'd often set the speed limits higher than the total bandwidth you really have available. The most reliable option is to do a number of speedtests at various times of the day (most of which provide your speeds in both Mb/S and MB/s), and then set your limits accordingly. I've found that leaving 30-50 KB/s of headroom (ie .24 to .4 Mb/s) is enough for general web browsing (though not streaming anything), and that leaving enough room on the upload side is just as important as leaving room on download. I do like the idea of "xx% of whatever's available" though, as that would be very nice for connections (like shared cable ones) that slow down in the evenings.

you could try changing the folder structure, so the affected folder is a subfolder rather than the main folder.

How does BTSync handle if a file moves to a new folder?From what I've noticed, it will delete the file in the old location and then make a new file in the new location (transferring the full content of the file to do this). Not ideal by any means, but over a lan connection it's not that big of a problem to me. I believe this is because of the way that BTSync works - it focuses on the folders and keeps them synced, rather than the individual files like synctoy. How does BTSync handle if a file gets its name changes?Sometimes the name change is done without a full transfer, and sometimes it seems that BTSync recognizes it as a delete and new file, thus it transfers the whole file again. I haven't found a way to reproduce one of them specifically; it seems pretty random to me. Again, over lan this doesn't bother me much. How are file deletions handled?If a file is deleted from computer A, on computer B (and c, d, etc) it is moved to the .syncarchive folder by default (keeping the folder structure). The files remain there for a determined length of time (I believe by default it's 30 days, but this is changable via an advanced option). If needed, they can be restored by a simple copy/paste on computer B (provided computer B has a full-access key, which sounds like your use case; this restore will trigger a full transfer to other machines, as with any 'new' file). How would this best work using the 3 computers or would I be better off to just use 2 and manually still use Synctoy as a 1-way sync to update the server as a backup every once in a whileI would personally just have BTSync on all three computers, and keep the server as a backup (it should keep deletions in its .syncarchive from both machines, so it would be a single source for restores, rather than having to check both computer A and B's .sycnarchives depending on who changed/deleted the file). Additionally, if one of the computers had a hdd crash, a restore could be quite a bit faster (depending on network setup, as well as lan/internet connections), as BTSync would be getting the files from both other sources instead of just one other computer. On the other hand, you could just set Synctoy as a scheduled task to run 1x/day or the like, which would accomplish something similar (though without real-time updating). ---- Overall, I do think this would be something useful to you. I use it for something similar with documents, and it works quite well thanks to the real-time updating (and with less hassle than synctoy, which I used previously). Before you use it on your main folders, I would test it with a temporary folder, just to make sure it all works right and as you were expecting. Also, when I switched over from synctoy to BTSync, I just made sure the folders were all synced and then added them to BTSync with the same full-access or read-only secret. It then indexed them (which took a while, due to the TBs of information I had to sync), and was ready to go - it didnt transfer anything across during the indexing process (though it looked like it was, it was really just scanning, as there was no network activity). Good luck!

Aside from adding them individually to .syncignore (or via a wildcard/formula if possible), I don't think it's possible. There isn't really a way that I've found to ignore or select files based on filesize (BTSync seems to be more folder-oriented than single file-oriented).

I've noticed that it shows files as transferring that are currently indexing (and also shows the remaining amount to index as a remaining amount to transfer). I described what I saw in more detail in a previous post below, perhaps that will help a bit. http://forum.bittorrent.com/topic/28256-pre-seeding-files/?=p82103

sync makes the .syncarchive folder, a .syncid, and the .syncignore files in the root of every shared folder. as far as I've found, there's no option to move them or put them in an alternate place. It doesn't look like it's possible - http://forum.bittorrent.com/topic/31151-syncid/

Updates don't normally show via the check now button for a week or two after each new version. I think this is to allow any regressions to be ironed out before the majority of users update to the new version, and also has those who do update use the forums, where they are more able to see any issues or bugs that have arisen.

I also think this would be a really cool feature, especially since many of the items I've synced are exactly that - photos, movies, music, disk images, etc - that rarely change. Due to cpu usage, I'd prefer a user-settable range, and I'd probably choose to do re-check every 2-3 months (probably up to 6 months, depending on the exact content). I would also agree that it would be best per-folder, so that you could re-check important files more often, as well as not having to suddenly re-hash several TB of data at once.

If you turn off the tracker, dht, relay, etc, and only have a predefined host of the linux server for the salesmen computers, then yes, they should only connect to the linux server.

can you find it in the .syncarchive folder? something else may be causing it to be deleted.

I believe GreatMarko meant "limited" as in "not able to install non-windows store (ie MS approved) programs."

Excellent, thanks! I just hadn't seen any forums posts on it and wanted to make sure the issue was known, so thanks for the update.

You could submit for an api key, I believe that officially allows commercial use.

Over the past few months of using BTSync, I've noticed a bit of a bug that I've been able to replicate on all the computers attempted. It's not really big or much of a hindrance, but it is occasionally annoying. I'm using win7 x64 (as are the other computers I've replicated this issue on, with one additional one that is x86). On the devices tab, it shows the various devices connected to, the folders on each, and the sync status. If you hover the mouse over the folder names, it will bring up a tooltip of the full file path on the local computer. If you alt-tab to another program, this tooltip remains, and will not go away until you return to the BTSync window and move the mouse back over the same folder name. I've also had it happen sometimes if I move the mouse out of the BTSync window and click to a window behind it, especially if I'm moving my mouse quickly back and forth. In like manner, if I go back to the BTSync window and hover over the folder name, it goes away. It's much harder to replicate if I'm RDPed into a computer, but I can replicate it on the local setup every time. Like I said, not a big deal, but it would be a nice fix, especially with 1.4 coming out soon.

That's very useful to know. I always thought it worked as GreatMarko said - a folder having a 50 peer connection limit per computer (ie each computer only being able to connect to a max of 50 others at any one time), not that there was a hard 50 online user limit for the entire folder/secret.

+1

If you disable trackers/dht and relays, then yes, it is a direct transfer with no other communication than between the two devices. For this to work though, you will need to know the hostname/ip:port for the computers, to manually input for them under the folder options. I know this is possible for computer-to-computer transfers, but I'm not sure if the mobile versions allow you to disable trackers or setup manual connections.

Usually this means they are in progress of syncing, iirc. Once the sync finishes, they are renamed to the correct filename/extension.

It will have to rescan the folder, which may take some time depending on the size of it, but it should realize that you have the full folder content and not need to re-download anything.