500 Internal Server error with rclone and "Post request rcat error"

Today I did my first ever restic backup to a onedrive (for business) based repo via rclone. I am aware that the errors that are shown are to do with rclone and not restic. What I want to know is if I need to worry if there are files that have not been backed up despite the “restic -r check” command not showing any errors.
This issue shows the same kind of error, but I also see more kinds of errors.
Here are some snippets of the kinds of errors I am getting:

rclone: 2021/08/10 14:21:49 ERROR : data/c3/c3459b0fbfd5e178bbd0f754ccefae4ea6b49ab26ce7c83d543c6dd4bda8612e: Post request rcat error: Put "https://ulisboa-my.sharepoint.com/personal/ ...


Save(<data/aceacb5496>) returned error, retrying after 328.259627ms: server response unexpected: 500 Internal Server Error (500) ...


rclone: 2021/08/10 17:24:41 ERROR : data/ac/acacb5496f8fe8fb7289407f8eb3106bd309c05eb4922f248ff5bb454ca22: Post request rcat error: itemNotFound: Item not found

I suspect that the backup process ran completely despite the errors because, again, the check command returned no errors, but I could be wrong and some files didn’t go through.

My versions are:

PS C:\> restic version
restic 0.12.1 compiled with go1.16.6 on windows/amd64
PS C:\> rclone version
rclone v1.56.0
- os/version: Microsoft Windows 10 Home 2009 (64 bit)
- os/kernel: 10.0.19043.1110 (x86_64)
- os/type: windows
- os/arch: amd64
- go/version: go1.16.5
- go/linking: dynamic
- go/tags: cmount

What do you think? Do I need to worry?

Thank you.

restic tries to upload files several times. That is unless a Save(xyz) returned error, retrying after [...] repeats ten times, there’s nothing to worry about. However, in that case restic will abort the backup and exit with an error. As check completed successfully, the backup should be fine. To be really sure, you could run check --read-data which downloads and verifies everything.

check --read-data returned that it found errors but wasn’t too informative. I suspect is because of those rclone errors. is there a magic command that fixes everything to do with the repo that contains errors?

It will be quite hard for anyone to suggest magic commands that fixes your errors, if you don’t tell them what those errors actually are… Please paste here the complete check command you ran, and all of its output, so that people can make an assessment.

1 Like

check --read-data shouldn’t report errors due to temporary rclone errors. But without the output from check that’s hard to tell.

Hi @MichaelEischer. Sorry for the delayed reply. Here is the result of running restic check that I ran again.

restic -r rclone:drive:backups check --read-data
using temporary cache in /tmp/restic-check-cache-722370503
enter password for repository: 
repository 3f3ba06e opened successfully, password is correct
created new cache in /tmp/restic-check-cache-722370503
create exclusive lock for repository
load indexes
check all packs
check snapshots, trees and blobs
read all data  2 / 3 snapshots
[0:06] 100.00%  3 / 3 snapshots
Pack ID does not match, want c82e8e7a, got 5b9df361
[2:12:45] 100.00%  6256 / 6256 packs
Fatal: repository contains errors

What does this mean?

Thank You.

Hmm, looks like one of the files in the repository was damaged during the upload. To fix this take a look at route 2 in Recover from broken pack file · Issue #828 · restic/restic · GitHub .