-
Notifications
You must be signed in to change notification settings - Fork 466
[GCP] Add loadbalancing_metrics distribution fields #4004
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
[GCP] Add loadbalancing_metrics distribution fields #4004
Conversation
🌐 Coverage report
|
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.
👍 please add a changelog entry and bump package version to 2.6.0
@@ -59,3 +59,43 @@ | |||
- name: tcp_ssl_proxy.open_connections.value | |||
type: long | |||
description: Current number of outstanding connections through the TCP/SSL proxy. | |||
- name: https.backend_latencies.value | |||
type: object | |||
object_type: histogram |
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.
Why does this need to use object_type
? Could it simply use type: histogram
?
In what stack versions is object_type: histogram
supported? That seems like something new.
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.
I used the prometheus integration as a reference. Should I just use type: histogram
instead?
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.
From my understanding object_type
should only be used for type: array
fields, see https://github.com/elastic/package-spec/blob/34c28eedce0d1ae7c140c998dd1e1bfdccdfc578/versions/1/integration/data_stream/fields/fields.spec.yml#L236-L239
So I think object_type: histogram
is supported, but should only be used if the field is of type: array
.
@andrewkroh if you confirm this I think this should be changed.
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.
Is there an example document somewhere that I can review to see what the fields look like?
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.
By looking at this documentation I think you are right type: histogram
is a valid type.
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.
After researching this topic for a bit, my conclusion is that: object_type: histogram
is supported but is not clear to me if type: histogram
would produce the same result.
@gpop63 could you try using type: histogram
to see if the behaviour is the same?
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.
I tested the way @endorama suggested and using type: histogram
in the integration field type does not work (reason":"error parsing field [gcp.loadbalancing.l3.external.rtt_latencies.value], with unknown parameter [histogram]"
) maybe because in metricbeat they were added as type: object
. I think we should either leave it like this or try changing metric types in metricbeat to type: histogram
.
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.
I also raised 2 issues with the team behind package-spec for missing documentation and they fixed it super fast with this PR, confirming support for object_type
.
@andrewkroh Let us know if you still have doubts around this, otherwise I think we should proceed by using object_type
.
…istribution_metrics
🚀 Benchmarks reportPackage
|
Data stream | Previous EPS | New EPS | Diff (%) | Result |
---|---|---|---|---|
dns |
1808.32 | 1524.39 | -283.93 (-15.7%) | 💔 |
loadbalancing_logs |
3937.01 | 3333.33 | -603.68 (-15.33%) | 💔 |
To see the full report comment with /test benchmark fullreport
What does this PR do?
Adds
loadbalancing_metrics
missing distribution fields.Checklist
changelog.yml
file.Author's Checklist
How to test this PR locally
Related issues
Screenshots