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
I am not sure if this feature is reasonable for this project, may be more proper when migrate this profiler into openTelemetry collector as profiling agent.
We plan to adopt continuous profiling for internal use. And the common use case is to select some specific pods to obtain profiling data instead of all kinds of processes in K8s node.
Invested several agents like Grafana Alloy/Parca/Pixie, all of them support the feature that only profiling specific pods. And as we move eBPF profiler into collector as agent, this should be a very reasonable requirement right?
The text was updated successfully, but these errors were encountered:
As a sampling based profiler, eBPF profiler is designed and architected to profile the whole system. Filtering, e.g. on a label, can be applied on post processing stages, the backend or UI.
With the ongoing integration into OTel collector, more processors/filters should become available.
I am not sure if this feature is reasonable for this project, may be more proper when migrate this profiler into openTelemetry collector as profiling agent.
We plan to adopt continuous profiling for internal use. And the common use case is to select some specific pods to obtain profiling data instead of all kinds of processes in K8s node.
Invested several agents like Grafana Alloy/Parca/Pixie, all of them support the feature that only profiling specific pods. And as we move eBPF profiler into collector as agent, this should be a very reasonable requirement right?
The text was updated successfully, but these errors were encountered: