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

The start_time in the total time is always the start_time of the read system call #289

Open
xih1919 opened this issue Feb 12, 2025 · 2 comments
Labels
bug Something isn't working

Comments

@xih1919
Copy link

xih1919 commented Feb 12, 2025

Describe the bug
On the server side, the start_time is always the start_time of the read system call, rather than the start_time when the data reaches the network card.

To Reproduce
Steps to reproduce the behavior:

Expected behavior
I hope that the start_time of the total time is mostly the start_time when the data arrives at the network card

Screenshots

Image

Image

Information (please complete the following information):
Linux hcss-ecs-53f2 5.4.0-190-generic #210-Ubuntu SMP Fri Jul 5 17:03:38 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux
kyanos-1.5.0

Copy link

dosubot bot commented Feb 12, 2025

@xih1919 Hello there! I'm here to help you with any bugs, questions, or contributions. Let's work together to solve the issue you're facing.

The issue you're experiencing, where the start_time defaults to the read system call time instead of the network card arrival time, is likely due to the absence of NIC ingress events. In the code, if there are no NIC ingress events, the logic falls back to using the timestamp of the readSyscallEvent as the StartTs [1].

To address this, ensure that NIC ingress events are being captured correctly. This might involve checking the configuration or setup of the event capturing mechanism to ensure that NIC events are not being missed. If NIC events are present, the code should use their timestamps to set the StartTs, aligning with your expected behavior.

To continue talking to Dosu, mention @dosu.


Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other | Bug Report

@dosubot dosubot bot added the bug Something isn't working label Feb 12, 2025
@hengyoush
Copy link
Owner

hengyoush commented Feb 13, 2025

Please start kyanos with --bpf-event-log-level 5 --conntrack-log-level 5, and trigger a http request, then provide the kyanos log files from the /tmp directory. Thanks! @xih1919

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants