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

Slow first connect after reopening app #3432

Closed
nisbet-hubbard opened this issue Oct 20, 2024 · 6 comments
Closed

Slow first connect after reopening app #3432

nisbet-hubbard opened this issue Oct 20, 2024 · 6 comments
Labels

Comments

@nisbet-hubbard
Copy link

Steps to reproduce

Closing and reopening app without using VPN

Outcome

What did you expect?

Syncing appears briefly

What happened instead?

Stuck at Syncing for 8 minutes. Potentially related to #3316.

Your phone model

No response

Operating system version

No response

Application version

1.9.1

Homeserver

conduwuit

Will you send logs?

Yes

@pixlwave
Copy link
Member

Heads up that #3316 has zero effect when a user is already logged in.

@nisbet-hubbard
Copy link
Author

You're absolutely right.

I mentioned it only because the kind of behaviour I'm observing (fast connect within hours, long wait when connecting after a day or so) seems to suggest Element is making a similar blocked request each time it connects to the homeserver after a long hiatus.

@pixlwave
Copy link
Member

Looking at your logs, they're filled with TimedOut errors on /_matrix/client/unstable/org.matrix.msc3575/sync and other endpoints like /_matrix/client/v3/room_keys/keys/…, /_matrix/client/v3/keys/query etc. Closing this as a backend issue.

@nisbet-hubbard
Copy link
Author

Glad the issue lies elsewhere. How should I go about debugging this?

Since /org.matrix.msc3575/ is involved here, would it be a good idea to just wait for conduwuit to implement the new sliding sync? Or should I report the issue now?

@pixlwave
Copy link
Member

pixlwave commented Nov 4, 2024

There's a chance this is in some way related to #3478

@nisbet-hubbard
Copy link
Author

Thanks for the pointer!

As killing the app in this case doesn't seem to improve the timeouts, we can probably rule out that possibility.

Having had more time to debug this, I've finally identified Cloudflare as the blocker (even though the requests aren't blocked as such). Removing Cloudflare from the chain also enables the use of mobile-optimised algorithms like ChaCha20-Poly1305, as the homeserver no longer has to talk to a server.

So, these might be other reasons for not using CF for Matrix.

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

No branches or pull requests

2 participants