-
Notifications
You must be signed in to change notification settings - Fork 2.4k
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
Greatly trim what netstat collector exposes by default #876
Conversation
Yeah it's annoying that these are not well documented, guess we won't get around tweaking this once we realize something is needed that got disabled. So LGTM for now. |
How about keeping IP/IP6 In/Out Octets? |
My thought was that we already have that from netdev. |
But we don't get IP vs IP6 from netdev. |
Let's also keep Udp Udp6 In/Out Datagrams. |
f5aa239
to
6f315e0
Compare
There a few overall error counters that might be useful:
|
Udp/Udp6 NoPorts is a counter of packets that don't hit a UDP listener. A good indicator of spammy / scanning traffic. Possibly worth including. |
Netstat is 40% of the metrics on my laptop, many of which are highly detailed information about IP internals in the kernel. ~300 such metrics on every machine in your fleet is excessive, so focus on key metrics by default, overridable by the user. Fixes #515 Signed-off-by: Brian Brazil <[email protected]>
We're at 31 now, so that's probably enough. |
6f315e0
to
22e2591
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Does this mean metrics such as the following no longer exist in any capacity ...
|
I don't think so. As Brian noted ("overridable by the user"), this can still be configured:
This PR is just about changing the defaults (node_netstat_TcpExt_ListenOverflows still seems to be part of the defaults, btw). |
Netstat is 40% of the metrics on my laptop, many of which are highly detailed information about IP internals in the kernel. ~300 such metrics on every machine in your fleet is excessive, so focus on key metrics by default, overridable by the user. Fixes prometheus#515 Signed-off-by: Brian Brazil <[email protected]>
Netstat is 40% of the metrics on my laptop, many of which
are highly detailed information about IP internals in the kernel.
~300 such metrics on every machine in your fleet is excessive,
so focus on key metrics by default, overridable by the user.
Fixes #515
@SuperQ I think we should aim for about 20-30 metrics from this collector, I've selected what I think are the main ones but there's undoubtedly more. Unfortunately these metrics have basically no documentation, so I'm working off experience.