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

Panic on startup #60

Open
pvonmoradi opened this issue Aug 21, 2021 · 8 comments
Open

Panic on startup #60

pvonmoradi opened this issue Aug 21, 2021 · 8 comments

Comments

@pvonmoradi
Copy link
Contributor

Panic in the main Weechat thread: panicked at 'called `Result::unwrap()` on an `Err` value: MatrixError(StateStore(Sled(Corruption { at: None, bt: () })))', src/server.rs:758:72

Restarting doesn't fix it. Using master build c4d0f1e

@poljar
Copy link
Owner

poljar commented Aug 21, 2021

This seems to indicate that the DB got corrupted, it's not quite clear why at would be None, this is coming from sled.

We should catch that error and not panic, but the bigger problem is that this happened at all.

What kind of hardware are you running this on? What OS/filesystem?

@pvonmoradi
Copy link
Contributor Author

pvonmoradi commented Aug 21, 2021

What kind of hardware are you running this on? What OS/filesystem?

OS: Ubuntu 18.04.5 LTS x86_64
Kernel: 5.4.0-81-generic
Host: Thinkpad T450s
Toshiba 2.5" HDD

FS is ext4

@poljar
Copy link
Owner

poljar commented Aug 21, 2021

Nothing really unusual, removing the database under ~/.weechat/matrix-rust/$server-name should get you going again.

Perhaps it would be wise to keep a backup of the DB and report the issue upstream on the Sled github issue tracker.

@pvonmoradi
Copy link
Contributor Author

Nothing really unusual, removing the database under ~/.weechat/matrix-rust/$server-name should get you going again.

Perhaps it would be wise to keep a backup of the DB and report the issue upstream on the Sled github issue tracker.

Up until recently, I always used /quit to exit weechat but very recently I just use a keybinding to close the parent terminal (Alacritty). Maybe this has something to do with it?

@pvonmoradi
Copy link
Contributor Author

By the way, db is 357M which is very unusual for texts...

@poljar
Copy link
Owner

poljar commented Aug 21, 2021

Possibly, depending on which signal Alacritty sends out, though I would have expected Sled to handle getting killed without getting corrupted, SQLite certainly handles this.

@aiobofh
Copy link

aiobofh commented Feb 6, 2022

I noticed this too today...

Note: I'm running my own home-server (conduit.rs)

Reproduce like this: Be logged in with plugin. Use another client (in my case Element on desktop), join a federated room. In my case I joined the FOSDEM room on matrix.org.

Suddenly weechat start to hog around 50% constant, and disk accesses of 500kB/s.... for infinite time. Until disk is full. :(

Sadly... unloading and loading the plugin again still is stuck in the same state. Had to remove the matrix-rust folder and re-set-up everything. I kept the .conf-file but removed the folder. Then all joined rooms appear as normal and CPU usage is low again.

@aiobofh
Copy link

aiobofh commented Feb 6, 2022

Addendum: unloading the plugin, and then removing the matrix-rust-sdk folder in .weechat/matrix-rust//, then load the plugin again also did the trick, (as a temporary work-around)

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

3 participants