mirror of
https://github.com/dani-garcia/vaultwarden.git
synced 2025-12-11 09:13:02 +03:00
v2024.5.0 doesn't allow one to enroll in account recovery #476
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @realkinetix on GitHub.
Running vaultwarden 1.30.5-f05398a6 with bw_web build 2024.5.0 produces the following when trying to enroll in account recovery:
Switched to bw_web build v2024.3.1 and it works fine.
@BlackDex commented on GitHub:
Not sure if we should fix this in some way. Almost all those kind of actions send the hash to be verified server side too. And then someone removed that check. I wonder how it works if someone logged-in via device and tries to enroll them self.
I'm also not sure when that PR will be merged and if we can wait for that for example.
@stefan0xC commented on GitHub:
Seems like the client (since
web-v2024.4.2cf.bf11b90c43) only sends the "ResetPasswordKey" when enrolling manually into account recovery because they did not validate it server-side.Should be fixed by https://github.com/bitwarden/clients/pull/8770 (but I have not looked if this requires further changes to our own validation logic).
@addisonbeck commented on GitHub:
I'm not all that familiar with this issue in the context of
vaultwarden, but I agree that https://github.com/bitwarden/clients/pull/8770 and it's sibling PRs will at least help resolve this since the primary end goal of that work is to do a server side hash verification during account recovery enrollment. Bitwarden clients will be sending a populatedMasterPasswordHashto the API again, which seems to be what is expected for this to work.Mostly chiming in because I can help with this:
I'm expecting to start having this work QA'd in the next week or so. The changes should be in
mainin the next couple of weeks, and will mostly likely land in production in mid July.You could fix this a little faster in
vaultwardenif you're willing to temporarily remove the server side hash check and put it back later. This is a small security issue, but the amount of things that would already need to have gone wrong for it to be exploited is high.