NIkolay Sivko
Nikolay Sivko, Founder & CEO at Coroot, is on a mission to make production troubleshooting easier for developers everywhere. He's deeply enthusiastic about SRE practices, observability, open source solutions. With over a decade of hands-on experience in the Observability field, Nikolay is a seasoned expert who brings practical insights.
Sessions
This talk explores troubleshooting complex systems with hundreds or thousands of services. It addresses key questions on necessary service information and dashboard structuring for rapid comprehension. Effective utilization of observability signals like metrics, logs, traces, and profiles is discussed, along with automation strategies for root cause analysis.
System observability is crucial for issue resolution but demands significant time and resources. This talk focuses on leveraging eBPF to gather telemetry data like metrics, logs, and traces. We'll also explore tracking container activities and using telemetry for troubleshooting effectively.
Understanding CPU time shortages isn't simple. Conventional metrics like CPU usage or Load Average may not suffice. Few realize the Linux kernel tracks task wait times for kernel resources. This talk delves into Linux kernel metrics unveiling CPU time shortages, essential for managing application latency.