Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add envoy dashboard #95

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Add envoy dashboard #95

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9471770
Add envoy dashboard
heydbut May 25, 2023
72e1b02
Merge branch 'main' into feat/envoy
heydbut Sep 13, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,9 +28,11 @@ Each resource has an associated module that will create Lightstep dashboards to
* __OpenTelemetry CouchDB Receiver__ (module: [`otel-collector-couchdbreceiver-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-couchdbreceiver-dashboard))
* __OpenTelemetry Collector__ (module: [`otel-collector-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-dashboard))
* __OpenTelemetry elasticsearchreceiver Integration__ (module: [`otel-collector-elasticsearchreceiver-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-elasticsearchreceiver-dashboard))
* __Envoy - Overview__ (module: [`otel-collector-envoy-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-envoy-dashboard))
* __Flink - Overview__ (module: [`otel-collector-flink-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-flink-dashboard))
* __Fluentd Records__ (module: [`otel-collector-fluentd-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-fluentd-dashboard))
* __Gunicorn - Overview__ (module: [`otel-collector-gunicorn-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-gunicorn-dashboard))
* __OpenTelemetry Hadoop Dashboard__ (module: [`otel-collector-hadoop-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-hadoop-dashboard))
* __OpenTelemetry HBase Dashboard__ (module: [`otel-collector-hbase-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-hbase-dashboard))
* __OpenTelemetry / Host__ (module: [`otel-collector-host-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-host-dashboard))
* __OpenTelemetry / Host Metrics / CPU__ (module: [`otel-collector-hostmetrics-cpu-dashboard`](https://github.com/lightstep/terraform-opentelemetry-dashboards/tree/main/collector-dashboards/otel-collector-hostmetrics-cpu-dashboard))
Expand Down
373 changes: 373 additions & 0 deletions collector-dashboards/otel-collector-envoy-dashboard/main.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,373 @@
terraform {
required_providers {
lightstep = {
source = "lightstep/lightstep"
version = "~> 1.76.0"
}
}
required_version = ">= v1.0.11"
}

resource "lightstep_dashboard" "otel_collector_envoy_dashboard" {
project_name = var.lightstep_project
dashboard_name = "Envoy - Overview"
dashboard_description = "This dashboard provides a high-level overview of your Envoy cluster so you can monitor its performance and resource usage."

group {
rank = 3
title = "Listeners"
visibility_type = "explicit"

chart {
name = "Listeners Success Rate (Excluding Admin Interface)"
type = "timeseries"
rank = 0
x_pos = 0
y_pos = 0
width = 16
height = 8

query {
query_name = "((a/b)*100)"
display = "line"
hidden = false
query_string = "with\n a = metric envoy_http_downstream_rq_xx | filter ((((envoy_response_code_class != \"4\") && (envoy_response_code_class != 4)) && (envoy_response_code_class != 4.0)) && (((envoy_response_code_class != \"5\") && (envoy_response_code_class != 5)) && (envoy_response_code_class != 5.0))) | rate | group_by [], sum;\n b = metric envoy_http_downstream_rq_completed | filter (((envoy_response_code_class == \"2\") || (envoy_response_code_class == 2)) || (envoy_response_code_class == 2.0)) | rate | group_by [], sum;\njoin (((a / b)*100)), a=0, b=0"
}
}
chart {
name = "Listeners Response Time Percentiles"
type = "timeseries"
rank = 1
x_pos = 16
y_pos = 0
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_rq_time | delta | group_by [], sum | point percentile(value, 50.0), percentile(value, 95.0), percentile(value, 99.0), percentile(value, 99.9)"
}
}
chart {
name = "Listener Traffic"
type = "timeseries"
rank = 2
x_pos = 32
y_pos = 0
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_rq_completed | rate | group_by [], sum"
}
query {
query_name = "b"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_cx_tx_bytes_total | rate | group_by [], sum"
}
}
chart {
name = "Requests Rejected By Reason"
type = "timeseries"
rank = 3
x_pos = 0
y_pos = 8
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_http_no_route | rate | group_by [], sum"
}
query {
query_name = "b"
display = "line"
hidden = false
query_string = "metric envoy_http_no_cluster | rate | group_by [], sum"
}
query {
query_name = "c"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_cx_protocol_error | rate | group_by [], sum"
}
}
chart {
name = "Active Connections Per Type and Listener"
type = "timeseries"
rank = 4
x_pos = 16
y_pos = 8
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_cx_active | latest | group_by [], sum"
}
query {
query_name = "b"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_cx_http2_active | latest | group_by [], sum"
}
query {
query_name = "c"
display = "line"
hidden = false
query_string = "metric envoy_http_downstream_cx_http1_active | latest | group_by [], sum"
}
}
}
group {
rank = 0
title = ""
Copy link
Collaborator

@nslaughter nslaughter Sep 13, 2023 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The group without a title seems wrong. Oversight? Trying to accomplish some formatting with the empty group?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should not remove this. It's the default unamed group for some charts

visibility_type = "implicit"
}
group {
rank = 1
title = "Overview"
visibility_type = "explicit"

chart {
name = "Incoming Success Rate (Non-5xx Responses)"
type = "timeseries"
rank = 0
x_pos = 0
y_pos = 0
width = 16
height = 8

query {
query_name = "((1-(a/b))*100)"
display = "big_number"
hidden = false
query_string = "with\n a = metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"5\") || (envoy_response_code_class == 5)) || (envoy_response_code_class == 5.0)) | rate 5m, 5m | group_by [], mean;\n b = metric envoy_cluster_upstream_rq_completed | rate 5m, 5m | group_by [], mean;\njoin (((1-(a / b))*100)), a=0, b=0 | reduce 5m, mean"
Copy link
Collaborator

@nslaughter nslaughter Sep 13, 2023 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I find the name confusing. Looks like the metric is 1 - (5XXs / Completed). In this context I find it confusing to call that a Success Rate, because I think of that as 2XXs. Not sure but also seems like we'd group_by the code class, but hard to tell the intent without the context of the whole dashboard.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's the new version of envoy. I find it confusing but I translated an existing dashboard and it was like this

}
}
chart {
name = "Incoming Requests Volume"
type = "timeseries"
rank = 1
x_pos = 16
y_pos = 0
width = 16
height = 8

query {
query_name = "a"
display = "big_number"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_xx | rate 5m | group_by [], mean"
}
}
chart {
name = "Incoming Requests by Release"
type = "timeseries"
rank = 2
x_pos = 32
y_pos = 0
width = 16
height = 8

query {
query_name = "a"
display = "big_number"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_total | rate 5m | group_by [], mean"
}
}
}
group {
rank = 2
title = "Upstream Clusters"
visibility_type = "explicit"

chart {
name = "Upstream Response 2xx (% Breakdown)"
type = "timeseries"
rank = 0
x_pos = 0
y_pos = 0
width = 16
height = 8

query {
query_name = "((a/b)*100)"
display = "bar"
hidden = false
query_string = "with\n a = metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"2\") || (envoy_response_code_class == 2)) || (envoy_response_code_class == 2.0)) | rate | group_by [], sum;\n b = metric envoy_cluster_upstream_rq_completed | rate | group_by [], sum;\njoin (((a / b)*100)), a=0, b=0"
}
}
chart {
name = "Upstream Response 3xx (% Breakdown)"
type = "timeseries"
rank = 1
x_pos = 16
y_pos = 0
width = 16
height = 8

query {
query_name = "((a/b)*100)"
display = "bar"
hidden = false
query_string = "with\n a = metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"3\") || (envoy_response_code_class == 3)) || (envoy_response_code_class == 3.0)) | rate | group_by [], sum;\n b = metric envoy_cluster_upstream_rq_completed | rate | group_by [], sum;\njoin (((a / b)*100)), a=0, b=0"
}
}
chart {
name = "Upstream Response 4xx (% Breakdown)"
type = "timeseries"
rank = 2
x_pos = 32
y_pos = 0
width = 16
height = 8

query {
query_name = "((a/b)*100)"
display = "bar"
hidden = false
query_string = "with\n a = metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"4\") || (envoy_response_code_class == 4)) || (envoy_response_code_class == 4.0)) | rate | group_by [], sum;\n b = metric envoy_cluster_upstream_rq_completed | rate | group_by [], sum;\njoin (((a / b)*100)), a=0, b=0"
}
}
chart {
name = "Upstream Response 5xx (% Breakdown)"
type = "timeseries"
rank = 3
x_pos = 0
y_pos = 8
width = 16
height = 8

query {
query_name = "((a/b)*100)"
display = "bar"
hidden = false
query_string = "with\n a = metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"5\") || (envoy_response_code_class == 5)) || (envoy_response_code_class == 5.0)) | rate | group_by [], sum;\n b = metric envoy_cluster_upstream_rq_completed | rate | group_by [], sum;\njoin (((a / b)*100)), a=0, b=0"
}
}
chart {
name = "Upstream Response 2xx (Total Breakdown)"
type = "timeseries"
rank = 4
x_pos = 16
y_pos = 8
width = 16
height = 8

query {
query_name = "a"
display = "bar"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"2\") || (envoy_response_code_class == 2)) || (envoy_response_code_class == 2.0)) | rate | group_by [], sum"
}
}
chart {
name = "Upstream Response 3xx (Total Breakdown)"
type = "timeseries"
rank = 5
x_pos = 32
y_pos = 8
width = 16
height = 8

query {
query_name = "a"
display = "bar"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"3\") || (envoy_response_code_class == 3)) || (envoy_response_code_class == 3.0)) | rate | group_by [], sum"
}
}
chart {
name = "Upstream Response 4xx (Total Breakdown)"
type = "timeseries"
rank = 6
x_pos = 0
y_pos = 16
width = 16
height = 8

query {
query_name = "a"
display = "bar"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"4\") || (envoy_response_code_class == 4)) || (envoy_response_code_class == 4.0)) | rate | group_by [], sum"
}
}
chart {
name = "Upstream Response 5xx (Total Breakdown)"
type = "timeseries"
rank = 7
x_pos = 16
y_pos = 16
width = 16
height = 8

query {
query_name = "a"
display = "bar"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_xx | filter (((envoy_response_code_class == \"5\") || (envoy_response_code_class == 5)) || (envoy_response_code_class == 5.0)) | rate | group_by [], sum"
}
}
chart {
name = "Upstream p99 Response Time"
type = "timeseries"
rank = 8
x_pos = 32
y_pos = 16
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_time | delta | group_by [], sum | point percentile(value, 99.0)"
}
}
chart {
name = "Average Upstream Traffic Rate"
type = "timeseries"
rank = 9
x_pos = 0
y_pos = 24
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_cluster_upstream_rq_total | rate | group_by [], sum"
}
}
chart {
name = "Cluster Load Balancer Panics"
type = "timeseries"
rank = 10
x_pos = 16
y_pos = 24
width = 16
height = 8

query {
query_name = "a"
display = "line"
hidden = false
query_string = "metric envoy_cluster_lb_healthy_panic | rate | group_by [], sum"
}
}
}
}
4 changes: 4 additions & 0 deletions collector-dashboards/otel-collector-envoy-dashboard/outputs.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
output "dashboard_url" {
value = "https://app.lightstep.com/${var.lightstep_project}/dashboard/${lightstep_dashboard.otel_collector_envoy_dashboard.id}"
description = "OpenTelemetry Collector Envoy Metrics Dashboard URL"
}
4 changes: 4 additions & 0 deletions collector-dashboards/otel-collector-envoy-dashboard/variables.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
variable "lightstep_project" {
description = "Name of Lightstep project"
type = string
}
Loading