I ran into this today: I have one master system where I store all my photos (~72GB). This is a Debian Linux box. I have 3 other systems with read-only keys to that folder, one Windows 7 and two OS X 10.9.
I upgraded all to 1.3 last night and found several folders with IMG_2000.Conflict.jpg files on the Windows machine. After reading through this thread and some others I figured out it was due to the case-sensitivity issue. Some editor or utility I used saved the files as *.jpg instead of *.JPG so I had "dupes". This was easy to fix (or so I thought) by eliminating the duplicate files on the source (Linux) machine.
The problem was that even though I removed the duplicates on the source and eventually on every one of the systems (while BTSync was stopped) they kept reappearing on the Windows machine no matter what when I would start BTSync back up again. It's like they were still somehow in the database even though they no longer existed on the filesystems of any of the machines.
The way I ultimately fixed it was to remove the sync folder on the Windows machine, delete the *.Conflict.* files, and re-add the sync folder. This somehow blew them away, probably by cleaning out whatever queue or database still had the files in it.
Now that I know what was wrong it makes sense but the fact that I wasn't able to fix it without removing and re-adding the sync folder isn't optimal. Even on a very fast machine, it took around a half hour for the databases to rebuild and start syncing again. This problem would be greatly exaggerated if this were hundreds of gigabytes or more of files.
It would be really nice if this could be handled in some kind of informative message in the clients when you have differing case handling. If I hadn't found this thread I'd probably still be banging my head against the wall as just renaming the files isn't very communicative. For someone that understands case sensitivity and runs multiple operating systems it's arguably discoverable but for people with a Linux-based NAS or something on their network who don't crack open the filesystem it could be a real pickle.
By the way, please keep it up. BTSync is a utility I've wanted for YEARS. I cobbled together similar things using rsync and other utilities but BTSync is what I was always trying to get to.
Dealing With '.conflict.' Folders
in Sync Troubleshooting
Posted
I ran into this today: I have one master system where I store all my photos (~72GB). This is a Debian Linux box. I have 3 other systems with read-only keys to that folder, one Windows 7 and two OS X 10.9.
I upgraded all to 1.3 last night and found several folders with IMG_2000.Conflict.jpg files on the Windows machine. After reading through this thread and some others I figured out it was due to the case-sensitivity issue. Some editor or utility I used saved the files as *.jpg instead of *.JPG so I had "dupes". This was easy to fix (or so I thought) by eliminating the duplicate files on the source (Linux) machine.
The problem was that even though I removed the duplicates on the source and eventually on every one of the systems (while BTSync was stopped) they kept reappearing on the Windows machine no matter what when I would start BTSync back up again. It's like they were still somehow in the database even though they no longer existed on the filesystems of any of the machines.
The way I ultimately fixed it was to remove the sync folder on the Windows machine, delete the *.Conflict.* files, and re-add the sync folder. This somehow blew them away, probably by cleaning out whatever queue or database still had the files in it.
Now that I know what was wrong it makes sense but the fact that I wasn't able to fix it without removing and re-adding the sync folder isn't optimal. Even on a very fast machine, it took around a half hour for the databases to rebuild and start syncing again. This problem would be greatly exaggerated if this were hundreds of gigabytes or more of files.
It would be really nice if this could be handled in some kind of informative message in the clients when you have differing case handling. If I hadn't found this thread I'd probably still be banging my head against the wall as just renaming the files isn't very communicative. For someone that understands case sensitivity and runs multiple operating systems it's arguably discoverable but for people with a Linux-based NAS or something on their network who don't crack open the filesystem it could be a real pickle.
By the way, please keep it up. BTSync is a utility I've wanted for YEARS. I cobbled together similar things using rsync and other utilities but BTSync is what I was always trying to get to.