Comment 6 for bug 1770200

I've stripped out all incrementals until it didn't make sense to strip
out more (since I'd be redoing >90% of the backup), did an incremental
backup (no errors), and ran a verify operation. I'm still getting
several dozens of errors, which I've uniq'd after anonimizing the filenames:

Error 'First patch in sequence [<duplicity.path.ROPath instance at
0x7ff2843fde60>] was a diff' patching <anonimized>
Error 'librsync error 103 while in patch cycle' patching <anonimized>
Error 'Patch sequence isn't regular, but has 2 entries' patching
<anonimized>
python: ERROR: (rs_file_copy_cb) unexpected eof on fd12
python: ERROR: (rs_job_complete) patch job failed: unexpected end of input

My conclusion is I'll have to re-do the backup. Disconnects are a fact
of life on my connection, and since a full backup is >3 months upload
time, my concern is I will have another corrupted backup at the end. Any
suggestions are welcome; thanks for your help so far.

On 06/02/2018 09:30 AM, Kenneth Loafman wrote:
> Good idea about stripping out incremental sets until you find the
> problem. That may be the best way to proceed. I would move them to
> another directory instead of deleting just to be safe. Let me know how
> it goes.
>