seems not work in my environment about ktls-bpf_redirect-example HOT 12 CLOSED

Hello,
The actual goal of this example was to highlight an issue in the kernel. I attempted to resolve it and submitted a change request. However, due to extensive code modifications, it is now outdated.

from ktls-bpf_redirect-example.

hi julien,
thanks for your reply. so the repo bpf-exp-rp has the same issue, right?
do you know why _prog_parser is not executed when sending data from openssl client?

from ktls-bpf_redirect-example.

Yes, the issue persists with bpf-exp-rp, and the reason for its non-execution remains unclear. This problem is tied to the fact that _prog_parser and ktls both utilize the same mechanism, resulting in a conflict.

from ktls-bpf_redirect-example.

thanks for pointing this out

from ktls-bpf_redirect-example.

Hello, The actual goal of this example was to highlight an issue in the kernel. I attempted to resolve it and submitted a change request. However, due to extensive code modifications, it is now outdated.

@juliens do you have a link or mail list for discussion this issue in the kernel side, I saw this issue also. I want to track when this can be fixed.

from ktls-bpf_redirect-example.

See the thread on the mailing list here : https://lore.kernel.org/bpf/[email protected]/

from ktls-bpf_redirect-example.

See the thread on the mailing list here : https://lore.kernel.org/bpf/[email protected]/

@juliens thanks a lot! It seems the last reply is a year ago, it seems no hope to be fixed, right?

from ktls-bpf_redirect-example.

You're right, it's been a while since nothing happen on this thread.
To be honest, I'm not familiar enough with the kernel source code to fix it. Nevertheless, I maintain hope that resolving a different bug might indirectly lead to the resolution of this one.

from ktls-bpf_redirect-example.

You're right, it's been a while since nothing happen on this thread. To be honest, I'm not familiar enough with the kernel source code to fix it. Nevertheless, I maintain hope that resolving a different bug might indirectly lead to the resolution of this one.

got it, thanks for helping me understand the status!

from ktls-bpf_redirect-example.

You're right, it's been a while since nothing happen on this thread. To be honest, I'm not familiar enough with the kernel source code to fix it. Nevertheless, I maintain hope that resolving a different bug might indirectly lead to the resolution of this one.

got it, thanks for helping me understand the status!

@juliens just want you know, I tested in the 6.x version kernel, it works now :)

from ktls-bpf_redirect-example.

I'm not sure it's completly fixed.

I tried with 6.7.0

Linux juliens 6.7.0 #1-NixOS SMP PREEMPT_DYNAMIC Sun Jan  7 20:18:38 UTC 2024 x86_64 GNU/Linux

When I send:

read R BLOCK
TEST
TEST
TEST

I receive:

❯ nc -l 0.0.0.0 8080
TEST
xTEST
tobject.serialr_202TEST
06_p1%22%2C1%2Cn

from ktls-bpf_redirect-example.

it works in my environment
Linux ubuntu2304.localdomain 6.2.0-33-generic #33-Ubuntu SMP PREEMPT_DYNAMIC Tue Sep 5 14:49:19 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux

But when sending a large message, the server seems not to receive all of them

from ktls-bpf_redirect-example.