-
-
Notifications
You must be signed in to change notification settings - Fork 2.2k
Cannot share keys after re-login #29714
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
Comments
Even if they were, the request would be rejected. This is a feature called PFS https://www.vmware.com/topics/perfect-forward-secrecy If a user has 0 devices logged in at the time a message is sent then no encryption keys are shared to that user for a given megolm session. The solution to this is something called device dehydration which is already supported in Labs but requires specific support from your server. Closing in favour of device dehydration |
Perhaps I wasn't clear enough. There were two users in the room: main and test. Only the test user logged out, while main remained in the room the entire time. So there was always at least one user (main) present in the room continuously. But when test returned, it still had no option to request keys from main. |
@hitriyvalenok perhaps I wasn't clear, I said "devices" not "users". A user has 0-N devices. If a user has 0 devices then PFS means they cannot decrypt anything that happens during the period of time when they have 0 devices. PFS allows you to protect against situations where devices are lost/stolen or passwords leaked so that impacted messages are minimised to just that which happened during the breach, any history before the breach, and any messages after the breach would be undecryptable.
Because the Matrix protocol does not allow this. I mean you could request, but the other side would reject, due to PFS, which is a feature of the Megolm protocol. |
Thanks for your reply!
Can I use this feature on the default matrix.org server? |
I don't think so, matrix.org tends to not enable features until they are stable in the Matrix spec and matrix-org/matrix-spec-proposals#3814 is still in draft |
Is device dehydration safe to use? |
I'm not involved with that project. So the worst case scenario would be that the dehydrated device becomes unrecoverable and it does not grant you access to history whilst you had 0 (other) devices. Thus, depends on what you mean safe. I think the implementation may still have some bugs: https://github.com/element-hq/element-web/issues?q=sort%3Aupdated-desc%20is%3Aissue%20is%3Aopen%20%20label%3AA-E2EE-Dehydration%20 |
Uh oh!
There was an error while loading. Please reload this page.
Steps to reproduce
I created a second account (wentel-test) and set up a test room. I logged out from wentel-test, sent a message to the room from my main account, then logged back as wentel-test (using the recovery key). The message expectedly couldn't be decrypted.
Outcome
The keys aren't being requested automatically (even though notifications are enabled in my main account's browser), and I can't request them manually (only have three options: Share, Report, View Source). Key storage is enabled, all settings are at default.
What did you expect?
wentel-test to be able to request sharing the keys OR keys to be shared automatically
What happened instead?
nothing
Operating system
MacOS (Catalinf)
Browser information
128.0.6613.138
URL for webapp
https://app.element.io/
Application version
1.11.97
Homeserver
No response
Will you send logs?
wentel (main).zip
wentel-test.zip
The text was updated successfully, but these errors were encountered: