Disappointed Cat

I have a home server running Linux. All shares are added on linux locally and on two win8 PCs also. The laptop is usually off. If I'm working on a web project I open the files with Netbeans from Windows via a Samba share so I can instantly check it in the browser. (The webserver is on the linux peer.) So the win8 client just recieves files in the background, like backup. But somehow 3 times a file was synced back overwriting a newer one on the linux peer. Just in case it's not clear: All shares are added to a local path and I only work on one peer's files at a time. Everything is also available via Samba but not everything is synced with BTSync.

Sent one from Linux and one from Win8. Happy debugging.

Update I found some log entries about all 3 occasions. Sorry, I didn't check before. These are logs from a linux peer, where I edited the files via a Samba share. The .3 IP belongs to a Win8 client, where these files weren't even opened. [20130530 23:42:07.401] Incoming connection from 172.30.20.3:18553 [20130530 23:46:52.214] 172.30.20.3:49072: did not pick any blocks. blocking peer temporarily [20130530 23:46:52.266] ImporterBase.cls.php: Piece 0 complete [20130601 03:17:23.134] ReadFile error: CatLab.cls.php:0:22680:22680:3 [20130601 03:17:24.286] 172.30.20.3:49072: did not pick any blocks. blocking peer temporarily [20130601 03:17:24.432] CatLab.cls.php: Piece 0 complete [20130601 04:04:17.988] Incoming connection from 172.30.20.3:54037 [20130601 04:07:28.557] 172.30.20.3:49072: did not pick any blocks. blocking peer temporarily [20130601 04:07:28.610] CatLab.cls.php: Piece 0 complete Also there are hundres of lines like these but they are way earlier, going back days.. [20130530 18:33:11.011] Failed read from file error = 9 [20130530 18:33:11.011] File changed during hashing

Browsing the logs I noticed that all paths are now prepended with \\?\. Is it safe to assume that UNC is now used by default?

It happened with files I was editing in Netbeans. I was often saving the second time before it was even synced and maybe it confused the client somehow. Or maybe Netbeans' code scanner (for auto completion, etc) locked the file and that caused some problems. This only happened 3 times in a few hours so I don't know what really caused this. What I do know is that I worked on projects before and earlier versions didn't have this problem, or I was very lucky. Maybe there should be a limit on how often can a file be synced. Say once in 2 minutes. I'm guessing this would eliminate these kind of conflicts and also reduce the over syncing.

It is a problem on low quality internet services. Opening like a hundred connections for DHT can drop the internet connection or render it unusable for the time being. It also disturbing that my PC connects to random people all over the world.

The 1.1.22 build randomly overwrites a few of my source files. I think the time diff is still buggy. There's only 1-2 second diff between the machines, one uses the other as an NTP server automatically. I can't manually reproduce it.

Is at least one of those ports opened on the respective computer? In other words: are they allowed through the firewalls?

Why not just use file ACLs? It also has the advantage of being independent from any program.

It seems the android app causes other peers to touch (unix touch) directories that are partially synced. In other words every time I start the droid app, the (c|m)time of directories that contain synced files (on the phone) changes on my linux server. This confuses the heck out of my backup scripts.. Edit: This seems to be an isolated issue or not directly related to BTSync. ^^

Here's another note: Mobil-to-mobil shares only show QR codes so if one device doesn't have a camera it becomes a one-way feature. Also the battery usage needs the same attention as CPU/RAM usage on the PC. I can't run this non-stop.

I dug up my old My Audio tablet (4.0) and it crashed before I could even install the app. Anyway after dealing with the same old shit, BTSync installed and works without a problem. I also tested the mobil-to-mobil feature. Works seamlessly. I can't wait for this to be widely used.

So this basically creates a regular share, right?

Not yet, but I will with my tablet soon. I imagine it will generate a QR code for my selected files, the other device scans it and downloads them. This is really awesome. Will it sync if multiple devices scan it? We could spread files like drugs.

I tried it out too. It works fine on rooted HTC Desire, Android 4.2, CyanogenMOD. It didn't require root however. My first impressions: * I like the UI and the QR code approach. * The folder view needs alphabetical ordering - it's chaotic now. * I also like the on-demand solution but I have a different idea how it should work. * Cellular data should be a per share setting. I think there should be only one file browser in which the user can select which files should be always offline via long pressing or something. This also should work with complete folders, that almost no other programs support. By simply opening something it should be temporarly synced. Also creating the entire directory structure makes things inconvenient. Keep up the good work.

There is already a topic about Apache rev. proxy - just for reference. At first glance it's not working because the webUI redirects to /gui and your proxy operates under /btsync.

I'd use a simple bash script to manage all kind of trashes. Here's mine: #!/bin/bash trashes="/mnt/Volume/.Trash*/ /mnt/Volume/.Dropbox/.dropbox.cache/* /mnt/Volume/*/.SyncTrash/" case "$1" in rm) for dir in $trashes; do if [ -d "$dir" ]; then rm -rf "$dir" fi done ;; gc) for dir in $trashes; do if [ -d "$dir" ]; then find "$dir" -depth -type f -ctime +7 -exec rm -rf "{}" \; find "$dir" -depth -type d -empty -exec rmdir "{}" \; fi done ;; ls) for dir in $trashes; do if [ -d "$dir" ]; then find "$dir" -type f -printf "[%Cm.%Cd %CH:%CM] %p\n" fi done ;; *) echo "Usage: $0 {rm|gc|ls}" exit 1 esac exit 0 You can set up "/path/to/script gc" as a daily cronjob and it deletes files that have been trashed 7 days ago.

I was thinking it may be necessary for [faster] migration. Well I guess not. It's also nice that the sync.dat file is now a few KB and not 20+ MB so it doesn't take minutes to load into an editor.

Great news! I noticed the metacache directory is not deleted automatically. Can we get rid of it? My music library will be re-indexing many-many hours and it still blocks every other share that comes after it in configuration order. I think indexing should run in two threads - one thread for one share at a time. It seems the DHT floods (when it's disabled) are gone. Thanks. And great job on the optimization.

Really? Please show us a way to wire it together with BTSync without the risk of exposing the key of the original data. Should I create a virtualized system with an OS that even encrypts the RAM? Implementing it in the client is a good idea indeed, but keep in mind that this shuts out Windows [8] as a host. I could settle for a Unix-only encrypted nodes feature nevertheless.

I can also confirm the issue on IE 10. Meh ...

It's not about mountable/usable encrypted storage. The use case would be seed servers and backup locations. BTSync wouldn't have to decrypt the AES-256 encrypted transfered data, just keep it as is. The problem with the implementation is the file indexing.

The webUI is rather rudimentary and it's low priority as it should be in this early stage. If you really want to reorder the shares (like me) then use a bencode editor and reorder the blocks in sync.dat. At the next start, the file guard hash will be wrong but it won't cause problems except there's gonna be a sync.[number].bad copy of the file which you can delete. PS: When you edit any config files, always stop the client.

There are a couple of problems with EncFS (and ecryptfs): - It's horribly slow. - Dokan does not work on Win8 which is a dependency. - On Unix systems if you also want file permissions and ACLs you have to give up even more performance. And for last: BTSyc would waste resources to re-encrypt everything for transfers. If there won't be official encrypted nodes feature someone will probably figure something out and implement it using the API.

BTSync doesn't follow symlinks. Accessing the symlink would also keep the disk awake, would it not?