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
Improve PID monitoring (step 2) #13530
Conversation
|
@vlvkobal about these errors, they have relationship with internal A new |
Summary
Fixes #12343
This PR is reducing CPU and memory usage for eBPF.plugin.
It is also disabling integration by default integration for legacy kernels, and some kernels that are LTS, but they do not have
trampoline
.This PR is also bringing
co-re
codes for more threads.Test Plan
ebpf.plugin
and verify that is your kernel was compiled withCONFIG_DEBUG_INFO_BTF
(zgrep BTF /proc/config.gz) the integration withapps
andcgroup
will be enabled by default, on the other hand, you should not have them enabled.libbpf
errors (grep libbpf /var/log/netdata/error.log) for threads different ofsocket
. Forsocket
this is expected after kernel5.19
to be released. This will be fixed in next PR.ebpf.programs
were reduced, they should have10992
instead the traditional32768
by default when you runbpftool map show
.Additional Information
This PR was tested on:
For users: How does this change affect me?
Describe the PR affects users: - Which area of Netdata is affected by the change? ebpf.plugin - Can they see the change or is it an under the hood? If they can see it, where? The plugin will use less resources from host. - How is the user impacted by the change? A better performance from plugin; - What are there any benefits of the change? plugin improvement.