Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Edge cases with dattobd #376

Open
woodholly opened this issue May 24, 2024 · 2 comments
Open

Edge cases with dattobd #376

woodholly opened this issue May 24, 2024 · 2 comments

Comments

@woodholly
Copy link

woodholly commented May 24, 2024

Hello, thank you for this awesome project!

Information from https://github.com/datto/dattobd/blob/main/doc/STRUCTURE.md raises several important questions:

  1. If incremental "Writes are tracked in-memory and periodically synced to the index file as needed", what will happen, if system crashes right after write has settled (with sync) to the block device, and before sync to .datto log ? If I understood correctly, .datto log will be in inconsistent state with with dire consequences after reboot. Or block writes and log writes are atomically synced not only in snapshot mode, but in incremental mode too ?

  2. What will happen, if COW storage becomes full ?

@Swistusmen
Copy link
Contributor

Hi
1.AFAIK they are automatically synced, or at least saved pretty fast,so that we never had complains about it,nor expected something like this. I also recommend setting up scripts which loads dattobd to the memory automatically after reboot
2. Next data will not be saved, unfortunately there is no nice way around it, but there are works on that topic

@woodholly
Copy link
Author

Hello @Swistusmen
If "they are automatically synced" it will be consistent, if "at least saved pretty fast" it will be inconsistent, especially in high disk load. These strategies are mutually exclusive :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants