Creates multiple snapshots on first run

[peterablehmann]

After running the backup for the first time, I found 8 snapshots in my repo.

[nix-shell:~]# restic -r <repo> snapshots -H sleipnir
enter password for repository:
repository ec66015b opened (version 2, compression level auto)
ID        Time                 Host        Tags        Paths
------------------------------------------------------------
3e19c67e  2024-02-03 21:46:26  sleipnir    C:\         C:\
fa76b203  2024-02-03 23:22:48  sleipnir    C:\         C:\
75aecaaf  2024-02-03 23:42:47  sleipnir    C:\         C:\
a93eda23  2024-02-04 00:02:28  sleipnir    C:\         C:\
eb401342  2024-02-04 02:38:05  sleipnir    C:\         C:\
024dfc8a  2024-02-04 02:58:01  sleipnir    C:\         C:\
d2d3e522  2024-02-04 03:17:48  sleipnir    C:\         C:\
abec3cb6  2024-02-04 03:37:55  sleipnir    C:\         C:\
------------------------------------------------------------
8 snapshots

Is this expected behavior?
None of this was triggered by task planning as that is set to 20:00.

[kmwoley]

No, that's not the expected behavior.
Without any additional information, it's impossible to tell what happened here.
Did you look in the scripts logging directory to see what the scripts logs produced?
Can you reproduce the issue?

[peterablehmann]

It did happen twice after I decided to delete the repository and start again.
But I don't have that install anymore and did not try again on other devices after that experience.

[kmwoley]

I suspect the retry logic saw errors in the first few runs and chose to re-run the backup until there were no errors, even though a snapshot was created. The error detection logic is very sensitive and will re-run until there are no errors, even if a snapshot is saved. If this happens to anyone else, please report a bug with the script logs so it can be investigated.

…

On Sun, Feb 9, 2025 at 12:06 PM Peter Lehmann ***@***.***> wrote: It did happen twice after I decided to delete the repository and start again. But I don't have that install anymore and did not try again on other devices after that experience. — Reply to this email directly, view it on GitHub <#93 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABOWG6E2NHWC23ENUFJXBNT2O6YMTAVCNFSM6AAAAABWNV33R6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBWGU3DGNJVGM> . You are receiving this because you commented.Message ID: ***@***.***>