I have a syncthing share that is stuck on some files that were deleted during syncing (files under the .git directory). Unfortunately, this is causing the puller to complain and fail syncing however which way I manually triggered re-scans on both the source and destination laptops.
The source folder is on a laptop named lenneth
, the destination folder is on a laptop named lezard
. They have a common share Projects
which is my catchall folder for basically all git repos and software I’m working on. Have never had a problem with the Projects folder before, but apparently somewhere during adding this one repo one or the other daemon got disconnected or something and now lezard
fails to sync Projects
, always failing on these files:
[IEEVH] 13:03:53 INFO: Puller (folder "Projects" (Projects), file "madumlao-profile/.git/COMMIT_EDITMSG"): finisher: pull: peers who had this file went away, or the file has changed while syncing. will retry later
[IEEVH] 13:03:53 INFO: Puller (folder "Projects" (Projects), file "madumlao-profile/.git/index"): finisher: pull: peers who had this file went away, or the file has changed while syncing. will retry later
[IEEVH] 13:03:53 INFO: Puller (folder "Projects" (Projects), file "madumlao-profile/install.sh"): finisher: pull: peers who had this file went away, or the file has changed while syncing. will retry later
[IEEVH] 13:03:53 INFO: Folder "Projects" (Projects) isn't making sync progress - retrying in 8m0s.
I’ve force re-scanned the folders, restarted syncthing then force re-scanned, force rsynced from one to the other then rescanned, force deleted and recopied on one then rescanned, force deleted on both then re-scanned then did a fresh copy to one, forced deleted on both then tried to do a fresh git clone, etc, but the puller always seems to think it can’t move forward due to the failed items.
At this point I’m not interested in figuring out why the puller failed. I just want to mark one share authoritative (the laptop I’m working on, lenneth
), and have the other one (lezard
) carry on syncing from there. Is there any way to do something like this, like manipulating indexes or something? – without having to recopy the share from scratch (tens of GBs) that is. Bonus points if it’s only limited to the failed files.