chore(otel): Open Telemetry metrics fixed and provided with demo exam…

…ple (#3038) * Redefine and cleanup Open Telemetry metrics types and usage * Rehauled demo example adding a minimal sample of RCP + Validator Node
gnolang · Nov 21, 2024 · 549da06 · 549da06
1 parent dc65f91
commit 549da06
Show file tree

Hide file tree

Showing 11 changed files with 262 additions and 281 deletions.
diff --git a/gno.land/pkg/sdk/vm/handler.go b/gno.land/pkg/sdk/vm/handler.go
@@ -1,17 +1,12 @@
 package vm
 
 import (
-	"context"
 	"fmt"
 	"strings"
 
 	abci "github.com/gnolang/gno/tm2/pkg/bft/abci/types"
 	"github.com/gnolang/gno/tm2/pkg/sdk"
 	"github.com/gnolang/gno/tm2/pkg/std"
-	"github.com/gnolang/gno/tm2/pkg/telemetry"
-	"github.com/gnolang/gno/tm2/pkg/telemetry/metrics"
-	"go.opentelemetry.io/otel/attribute"
-	"go.opentelemetry.io/otel/metric"
 )
 
 type vmHandler struct {
@@ -107,34 +102,9 @@ func (vh vmHandler) Query(ctx sdk.Context, req abci.RequestQuery) abci.ResponseQ
 				secondPart(req.Path), req.Path)))
 	}
 
-	// Log the telemetry
-	logQueryTelemetry(path, res.IsErr())
-
 	return res
 }
 
-// logQueryTelemetry logs the relevant VM query telemetry
-func logQueryTelemetry(path string, isErr bool) {
-	if !telemetry.MetricsEnabled() {
-		return
-	}
-
-	metrics.VMQueryCalls.Add(
-		context.Background(),
-		1,
-		metric.WithAttributes(
-			attribute.KeyValue{
-				Key:   "path",
-				Value: attribute.StringValue(path),
-			},
-		),
-	)
-
-	if isErr {
-		metrics.VMQueryErrors.Add(context.Background(), 1)
-	}
-}
-
 // queryPackage fetch a package's files.
 func (vh vmHandler) queryPackage(ctx sdk.Context, req abci.RequestQuery) (res abci.ResponseQuery) {
 	res.Data = []byte(fmt.Sprintf("TODO: parse parts get or make fileset..."))

diff --git a/misc/telemetry/README.md b/misc/telemetry/README.md
@@ -1,4 +1,4 @@
-## Overview
+# Open Telemetry overview
 
 The purpose of this Telemetry documentation is to showcase the different node metrics exposed by the Gno node through
 OpenTelemetry, without having to do extraneous setup.
@@ -8,39 +8,44 @@ The containerized setup is the following:
 - Grafana dashboard
 - Prometheus
 - OpenTelemetry collector (separate service that needs to run)
-- Single Gnoland node, with 1s block times and configured telemetry (enabled)
+- 1 RPC Gnoland node, with 1s block times and configured telemetry (enabled)
+- 1 Validator Gnoland node, with 1s block times and configured telemetry (enabled)
 - Supernova process that simulates load periodically (generates network traffic)
 
+## Metrics type
+
+Metrics collected are defined within codebase at `tm2/pkg/telemetry/metrics/metrics.go`.
+They are collected by the OTEL collector who forwards them to Prometheus.
+
+They are of three different types which can be used in Grafana adding different ypt of suffixes to the metrics name :
+
+- Histogram ("_sum", "_count", "_bucket"): Collect variations of values along time
+- Gauge: Measure a single value at the time it is read
+- Counter ("_total"): A value that accumulates over time
+
 ## Starting the containers
 
 ### Step 1: Spinning up Docker
 
 Make sure you have Docker installed and running on your system. After that, within the `misc/telemetry` folder run the
 following command:
 
-```shell
+```bash
 make up
 ```
 
 This will build out the required Docker images for this simulation, and start the services
 
 ### Step 2: Open Grafana
 
-When you've verified that the `telemetry` containers are up and running, head on over to http://localhost:3000 to open
+When you've verified that the `telemetry` containers are up and running, head on over to <http://localhost:3000> to open
 the Grafana dashboard.
 
-Default login details:
-
-```
-username: admin
-password: admin
-```
-
-After you've logged in (you can skip setting a new password), on the left hand side, click on
-`Dashboards -> Gno -> Gno Node Metrics`:
+After you've logged in, on the left hand side, click on
+`Dashboards -> Gno -> Gno Open Telemetry Metrics`:
 ![Grafana](assets/grafana-1.jpeg)
 
-This will open up the predefined Gno Metrics dashboards (added for ease of use) :
+This will open up the predefined Gno Metrics dashboards (added for ease of use):
 ![Metrics Dashboard](assets/grafana-2.jpeg)
 
 Periodically, these metrics will be updated as the `supernova` process is simulating network traffic.
@@ -53,4 +58,4 @@ To stop the cluster, you can run:
 make down
 ```
 
-which will stop the Docker containers. Additionally, you can delete the Docker volumes with `make clean`.
+which will stop the Docker containers. Additionally, you can delete the Docker volumes with `make clean`.
diff --git a/misc/telemetry/docker-compose.yml b/misc/telemetry/docker-compose.yml
@@ -9,6 +9,7 @@ services:
       - ./collector/collector.yaml:/etc/otelcol-contrib/config.yaml
     networks:
       - gnoland-net
+
   prometheus:
     image: prom/prometheus:latest
     command:
@@ -21,34 +22,90 @@ services:
       - ./prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
     networks:
       - gnoland-net
+
   grafana:
-    image: grafana/grafana-enterprise
+    image: grafana/grafana
+    environment:
+      - GF_AUTH_ANONYMOUS_ENABLED=true
+      - GF_AUTH_ANONYMOUS_ORG_ROLE=Admin
     volumes:
       - grafana_data:/var/lib/grafana
-      - ./grafana/datasources.yaml:/etc/grafana/provisioning/datasources/datasources.yaml
-      - ./grafana/dashboards.yaml:/etc/grafana/provisioning/dashboards/dashboards.yaml
-      - ./grafana/gno-dashboards.json:/var/lib/grafana/dashboards/gno-dashboards.json
+      - ./grafana/provisioning:/etc/grafana/provisioning
     ports:
       - "3000:3000"
     networks:
       - gnoland-net
-  gnoland:
-    build:
-      context: ./gnoland
-      dockerfile: Dockerfile
-    ports:
-      - "26657:26657"
+
+  gnoland-val:
+    image: ghcr.io/gnolang/gno/gnoland:master
     networks:
       - gnoland-net
+    volumes:
+      # Shared Volume
+      - gnoland-shared:/gnoroot/shared-data
+    entrypoint:
+      - sh
+      - -c
+      # Recreate gno genesis from git :(
+      - |
+        gnoland secrets init
+        rm -f /gnoroot/shared-data/node_p2p.id
+        apk add git make go linux-headers
+        git clone https://github.com/gnolang/gno.git --single-branch gnoland-src
+        GOPATH='/usr/' make -C gnoland-src/contribs/gnogenesis/
+        gnogenesis generate
+        gnogenesis validator add -name val000 -address $(gnoland secrets get validator_key.address -raw) -pub-key $(gnoland secrets get validator_key.pub_key -raw)
+        gnogenesis balances add -balance-sheet /gnoroot/gno.land/genesis/genesis_balances.txt
+        gnogenesis txs add packages /gnoroot/examples/gno.land
+        gnoland config init
+        gnoland config set consensus.timeout_commit 1s
+        gnoland config set moniker val000
+        gnoland config set telemetry.enabled true
+        gnoland config set telemetry.exporter_endpoint collector:4317
+        gnoland config set telemetry.service_instance_id val0
+        gnoland secrets get node_id.id -raw > /gnoroot/shared-data/node_p2p.id
+        cp /gnoroot/genesis.json /gnoroot/shared-data/genesis.json
+        gnoland start
+    healthcheck:
+      test: ["CMD-SHELL", "test -f /gnoroot/shared-data/node_p2p.id || exit 1"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+      start_period: 60s
+
+  gnoland-rpc:
+    image: ghcr.io/gnolang/gno/gnoland:master
+    networks:
+      - gnoland-net
+    volumes:
+      # Shared Volume
+      - gnoland-shared:/gnoroot/shared-data
+    entrypoint:
+      - sh
+      - -c
+      - |
+        gnoland secrets init
+        gnoland config init
+        gnoland config set consensus.timeout_commit 1s
+        gnoland config set moniker rpc0
+        gnoland config set rpc.laddr tcp://0.0.0.0:26657
+        gnoland config set telemetry.enabled true
+        gnoland config set telemetry.service_instance_id rpc000
+        gnoland config set telemetry.exporter_endpoint collector:4317
+        gnoland config set p2p.persistent_peers "$(cat /gnoroot/shared-data/node_p2p.id)@gnoland-val:26656"
+        gnoland start -genesis /gnoroot/shared-data/genesis.json
+    depends_on:
+      gnoland-val:
+        condition: service_healthy
+        restart: true
+
   supernova:
-    build:
-      dockerfile: supernova.Dockerfile
-      args:
-        supernova_version: v1.2.1
+    image: ghcr.io/gnolang/supernova:1.3.1
     command: >
-      -sub-accounts 10 -transactions 200 -url http://gnoland:26657
+      -sub-accounts 10 -transactions 100 -url http://gnoland-rpc:26657
       -mnemonic "source bonus chronic canvas draft south burst lottery vacant surface solve popular case indicate oppose farm nothing bullet exhibit title speed wink action roast"
-    restart: always
+      -mode PACKAGE_DEPLOYMENT
+    restart: unless-stopped
     networks:
       - gnoland-net
 
@@ -61,5 +118,5 @@ volumes:
     driver: local
   grafana_data:
     driver: local
-  gnoland:
+  gnoland-shared:
     driver: local
diff --git a/misc/telemetry/gnoland/Dockerfile b/misc/telemetry/gnoland/Dockerfile
diff --git a/misc/telemetry/gnoland/setup.sh b/misc/telemetry/gnoland/setup.sh
diff --git a/misc/telemetry/grafana/dashboards.yaml → ...a/provisioning/dashboards/dashboards.yaml b/misc/telemetry/grafana/dashboards.yaml → ...a/provisioning/dashboards/dashboards.yaml
@@ -5,4 +5,4 @@ providers:
     folder: Gno
     type: file
     options:
-      path: /var/lib/grafana/dashboards
+      path: /etc/grafana/provisioning/dashboards