You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is a parent issue of issue #78. This issue aims to expand the current metrics landscape by adding a packet loss tracer for the internal TCP connections.
Solution:
Trace the tcp_identify_packet_loss kprobe and log relevant insights.
Insights needed:
Packet loss
% of total packet loss
total size of data loss
comparison between loss data and total transmitted data
Activity
[#117]: added metrics_tracer crate (kernel space BPF program)
[#117]: added metrics crate (user space metrics collector)
[#117]: added kubernetes manifest for metrics collector service
[CortexFlow#117]: packet loss tracer working along with chaos mesh
[CortexFlow#117]: packet loss tracer working along with chaos mesh
[CortexFlow#117]: packet loss tracer working along with chaos mesh
[CortexFlow#117]: packet loss tracer working along with chaos mesh
[CortexFlow#117]: packet loss tracer working along with chaos mesh
[CortexFlow#117]: renaming fields of net_metrics
[CortexFlow#117]: Move display events for metric tracer to helpers
[CortexFlow#117]: Move display events for metric tracer to helpers