3.0.0-rc4: TCP connection error: i/o timeout about traefik HOT 8 CLOSED

Watchtower finally blew up in my face.

from traefik.

On another issue someone mentioned the lingeringtimeout setting:

#10594 (comment)

And it seems like they introduced a new default value in 2.11.1 and 3.0.0-rc4

v2.11.0...v2.11.1 v3.0.0-rc3...v3.0.0-rc4

I am not sure if it solves the problem but maybe you want to try it

Thanks for pointing out the workaround, however IMO this is a regression - this behaviour should not change between RC3 and RC4. If it was a beta, maybe that would be acceptable.

from traefik.

I am facing the same issue on v2.11.1. Downgrade to v2.11.0 fixed it for me.

from traefik.

I'll second this issue. I'm using image traefik:v3.0 (rc4) and seeing TCP connections drop after a few seconds, in my case Websocket connections. The following lines appearing in traefik logs

2024-04-10 21:04:27 2024-04-11T04:04:27Z ERR Error while Peeking first byte error="read tcp 172.26.0.12:80->172.26.0.1:53492: i/o timeout"
2024-04-10 21:04:34 2024-04-11T04:04:34Z ERR Error while Peeking first byte error="read tcp 172.26.0.12:80->172.26.0.1:57684: i/o timeout"

Rolling back to rc3 resolves the issue.

from traefik.

We also encounter some weird issues this morning after upgrading to the rc4.
Sometimes it ends the connection with 499 - Client Closed Request which is not true,
we also rolled back to rc3 which works perfectly ;)

from traefik.

Same issue after updating to rc4. Can confirm older version works fine.

from traefik.

On another issue someone mentioned the lingeringtimeout setting:

#10594 (comment)

And it seems like they introduced a new default value in 2.11.1 and 3.0.0-rc4

v2.11.0...v2.11.1
v3.0.0-rc3...v3.0.0-rc4

I am not sure if it solves the problem but maybe you want to try it

from traefik.

On behalf of the maintainer team, sorry for the inconvenience.
We opened a global issue to cover this bug: #10598
I'm closing this one, as we will update the global issue while we make progress on this.

from traefik.