Don't re-initialize an already initialized repository without user interaction

[simonsan]

Hi everybody,

I tested out some new rustic configuration (a new profile file) and mistakenly re-init one of my already used repos (copy & paste errors are saying "hi!"). This repo is actually only the hot repo (there is also a cold one, which is not affected).

First thing first: Why is it possible to simply re-init a repo without any warnings? It will create (at least) a new key (the old one was still there in my case (S3)) and overwrites the config file. Some kind of fail-safe for future dump-asses like me would be great :)

Originally posted by @jfreax in #1140

[aawsome]

@jfreax I tried to reproduce this behavior, but rustic always complained that the config file already exists and doesn't initialize the repository.
Can you give steps how I can reproduce this? Thanks a lot!

[jfreax]

I wish I could. I'm not 100% sure what I did. What I know is, that my original setup was with a hot and cold repository, but my new tests (which overwrote the config) was only using a "normal" (hot) repo without a cold repo configuration.

I will try to reproduce it.

[jfreax]

Unfortunately, I wasn't able to reproduce the behaviour. The config was definitely overwritten. I still have it around. Here are the some facts I still know, it might be helpful to track down what happened:

• whatever I did resulted in an overwritten config-file and a new secondary key in the keys-directory
• the new files appeared in the hot repo
• I originally used a hot+cold setup on S3 (scaleway)
• the new rustic setup I prepared was a simple (hot) repo
• I mixed up the repo endpoint/bucket name by copying over my profile file and forgot to edit some things
• a backup (hot+cold) was running while I did this (not 100% sure about this, but 99%)
  • the backup finished without any problems, restarting it or calling snapshots failed afterwards
• I also experimented with an rclone backend to reach a swift endpoint at (more or less) the same time
  • but this shouldn't interfere here, because I never configured the S3 endpoint on rclone (not even to this day)

[aawsome]

@jfreax I can only imagine that for some reason your backend told there was no config file. But it's hard to guess - and from the code I cannot see where something could go wrong elsewhere...

[aawsome]

closing this as the behavior cannot be reproduced. @jfreax If you encounter the same problem, please don't hesitate to re-open!