I had an emergency reboot when uploading and when the server came back up, it had clearly lost track of progress and had forgotten a lot of data. No problem; run a repair, right?
Repairs don’t work. It fails with:
Oct 5, 2018 9:03 AM: Operation List with file attempt 4 of 5 failed with message: Error getting response stream (ReadDone1): ReceiveFailure
(I can post a more detailed error message if needed.)
I tried a few times, both with repairs and with delete-and-repair, with no luck. Finally updated to the latest canary build in the hopes that it would fix things, and, nope, no luck.
At this point my backup is basically wrecked. This isn’t a huge problem because I can restart it if I need to, but if the Duplicati backend is this unstable, the program doesn’t seem very usable to me.
The backup is hosted on Google Drive, and is, at this point, about 2tb with 50mb files; so, 80k files, taking into account the indexes as well. It’s occurred to me that this might be too many files - on the other hand, Duplicati knows how much data I was going to upload, and if it can’t handle that filecount then it should probably adjust values automatically to something it’s OK with.
Any suggestions on what I can do to either fix things, or provide information to improve the repair system?