mirror of
https://github.com/openfrontio/OpenFrontIO.git
synced 2026-06-21 13:40:18 +00:00
use otel
This commit is contained in:
evan
+14
@@ -19,6 +19,17 @@ RUN mkdir -p /opt/node_exporter && \
|
||||
tar xvz --strip-components=1 -C /opt/node_exporter && \
|
||||
ln -s /opt/node_exporter/node_exporter /usr/local/bin/
|
||||
|
||||
# Install OpenTelemetry Collector
|
||||
ENV OTEL_VERSION=0.97.0
|
||||
RUN curl -sSL https://github.com/open-telemetry/opentelemetry-collector-releases/releases/download/v${OTEL_VERSION}/otelcol-contrib_${OTEL_VERSION}_linux_amd64.tar.gz \
|
||||
-o otelcol.tar.gz && \
|
||||
tar -xzf otelcol.tar.gz && \
|
||||
mv otelcol-contrib /usr/local/bin/ && \
|
||||
rm otelcol.tar.gz
|
||||
|
||||
# Create directory for OpenTelemetry config
|
||||
RUN mkdir -p /etc/otel
|
||||
|
||||
# Install cloudflared
|
||||
RUN curl -L https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64.deb > cloudflared.deb \
|
||||
&& dpkg -i cloudflared.deb \
|
||||
@@ -38,6 +49,9 @@ RUN mkdir -p .git && npm install
|
||||
# Copy the rest of the application code
|
||||
COPY . .
|
||||
|
||||
# Copy OpenTelemetry configuration
|
||||
COPY otel-collector-config.yaml /etc/otel/config.yaml
|
||||
|
||||
# Build the client-side application
|
||||
RUN npm run build-prod
|
||||
|
||||
|
||||
@@ -1,103 +0,0 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Metric Collector for Prometheus Pushgateway
|
||||
# This script collects metrics from Node Exporter and application sources
|
||||
# and pushes them to a Prometheus Pushgateway with custom labels.
|
||||
|
||||
# Configuration
|
||||
NODE_EXPORTER_URL="http://localhost:9100/metrics"
|
||||
APP_METRICS_URL="http://localhost:9090/metrics"
|
||||
PUSHGATEWAY_BASE_URL="https://mon.openfront.io/pushgateway/metrics"
|
||||
AUTH=$MON_USERNAME:$MON_PASSWORD
|
||||
INTERVAL=15 # seconds
|
||||
|
||||
# Function to fetch metrics from Node Exporter
|
||||
fetch_node_exporter_metrics() {
|
||||
curl -s --connect-timeout 5 --max-time 10 "$NODE_EXPORTER_URL" ||
|
||||
echo "# Error fetching Node Exporter metrics"
|
||||
}
|
||||
|
||||
# Function to fetch metrics from your application
|
||||
fetch_app_metrics() {
|
||||
curl -s --connect-timeout 5 --max-time 10 "$APP_METRICS_URL" ||
|
||||
echo "# Error fetching application metrics"
|
||||
}
|
||||
|
||||
# Function to push metrics to Pushgateway
|
||||
push_metrics() {
|
||||
local metrics=$1
|
||||
local job_name=$2
|
||||
|
||||
echo "Pushing $job_name metrics to Pushgateway..."
|
||||
|
||||
# Create a temporary file for the metrics
|
||||
TEMP_FILE=$(mktemp)
|
||||
echo "$metrics" > "$TEMP_FILE"
|
||||
|
||||
# Push to Pushgateway with instance label
|
||||
curl -s -u "$AUTH" --data-binary @"$TEMP_FILE" \
|
||||
"$PUSHGATEWAY_BASE_URL/job/$job_name/instance/$HOST"
|
||||
|
||||
# Check if push was successful
|
||||
if [ $? -eq 0 ]; then
|
||||
echo "$job_name metrics pushed successfully"
|
||||
else
|
||||
echo "Error pushing $job_name metrics"
|
||||
fi
|
||||
|
||||
# Remove temporary file
|
||||
rm "$TEMP_FILE"
|
||||
}
|
||||
|
||||
# Function to add labels to metrics
|
||||
add_labels() {
|
||||
local metrics=$1
|
||||
|
||||
# First, handle metrics with existing labels
|
||||
metrics=$(echo "$metrics" | sed -E 's/(\{[^}]*)\}/\1,env="'$ENV'",host="'$HOST'",subdomain="'$SUBDOMAIN'"}/g')
|
||||
|
||||
# Then, handle metrics with no existing labels
|
||||
metrics=$(echo "$metrics" | sed -E 's/^([a-zA-Z0-9_:]+)[ \t]+([0-9.e+-]+)$/\1{env="'$ENV'",host="'$HOST'",subdomain="'$SUBDOMAIN'"} \2/g')
|
||||
|
||||
echo "$metrics"
|
||||
}
|
||||
|
||||
# Main function to collect and push metrics
|
||||
collect_and_push_metrics() {
|
||||
echo "Starting metrics collection cycle at $(date)"
|
||||
|
||||
# Get metrics from both sources
|
||||
NODE_METRICS=$(fetch_node_exporter_metrics)
|
||||
APP_METRICS=$(fetch_app_metrics)
|
||||
|
||||
# Clean up metrics (remove headers etc.)
|
||||
NODE_METRICS=$(echo "$NODE_METRICS" | grep -v "^Fetching")
|
||||
APP_METRICS=$(echo "$APP_METRICS" | grep -v "^Fetching")
|
||||
|
||||
# Add labels to metrics
|
||||
NODE_METRICS=$(add_labels "$NODE_METRICS")
|
||||
APP_METRICS=$(add_labels "$APP_METRICS")
|
||||
|
||||
# Push to Pushgateway separately
|
||||
push_metrics "$NODE_METRICS" "node_exporter"
|
||||
push_metrics "$APP_METRICS" "app_metrics"
|
||||
|
||||
echo "Metrics collection cycle completed at $(date)"
|
||||
}
|
||||
|
||||
# Main execution
|
||||
echo "===== Starting metrics collector ====="
|
||||
echo "Environment: $ENV, HOST: $HOST, Subdomain: $SUBDOMAIN"
|
||||
echo "Collecting and pushing metrics every $INTERVAL seconds"
|
||||
echo "Node Exporter URL: $NODE_EXPORTER_URL"
|
||||
echo "App Metrics URL: $APP_METRICS_URL"
|
||||
echo "Pushgateway URL: $PUSHGATEWAY_BASE_URL"
|
||||
|
||||
# Wait for app to be ready.
|
||||
sleep 30
|
||||
|
||||
# Then set up interval loop
|
||||
while true; do
|
||||
sleep $INTERVAL
|
||||
collect_and_push_metrics
|
||||
done
|
||||
@@ -0,0 +1,63 @@
|
||||
receivers:
|
||||
prometheus:
|
||||
config:
|
||||
scrape_configs:
|
||||
- job_name: "node_exporter"
|
||||
scrape_interval: 15s
|
||||
static_configs:
|
||||
- targets: ["localhost:9100"]
|
||||
relabel_configs:
|
||||
- target_label: "host"
|
||||
replacement: "${HOST}"
|
||||
- target_label: "env"
|
||||
replacement: "${ENV}"
|
||||
- target_label: "subdomain"
|
||||
replacement: "${SUBDOMAIN}"
|
||||
- target_label: "container_id"
|
||||
replacement: "${HOSTNAME}"
|
||||
|
||||
- job_name: "app_metrics"
|
||||
scrape_interval: 15s
|
||||
static_configs:
|
||||
- targets: ["localhost:9090"]
|
||||
relabel_configs:
|
||||
- target_label: "host"
|
||||
replacement: "${HOST}"
|
||||
- target_label: "env"
|
||||
replacement: "${ENV}"
|
||||
- target_label: "subdomain"
|
||||
replacement: "${SUBDOMAIN}"
|
||||
- target_label: "container_id"
|
||||
replacement: "${HOSTNAME}"
|
||||
- target_label: "service"
|
||||
replacement: "${SERVICE_NAME}"
|
||||
|
||||
processors:
|
||||
batch:
|
||||
timeout: 10s
|
||||
resource:
|
||||
attributes:
|
||||
- key: service.name
|
||||
value: "${SERVICE_NAME}"
|
||||
- key: container.id
|
||||
value: "${HOSTNAME}"
|
||||
|
||||
exporters:
|
||||
prometheuspushgateway:
|
||||
endpoint: "https://mon.openfront.io/pushgateway"
|
||||
namespace: "${SERVICE_NAME}"
|
||||
job: "${SERVICE_NAME}"
|
||||
grouping_key:
|
||||
host: "${HOST}"
|
||||
env: "${ENV}"
|
||||
subdomain: "${SUBDOMAIN}"
|
||||
container_id: "${HOSTNAME}"
|
||||
headers:
|
||||
Authorization: "Basic ${BASIC_AUTH}"
|
||||
|
||||
service:
|
||||
pipelines:
|
||||
metrics:
|
||||
receivers: [prometheus]
|
||||
processors: [batch, resource]
|
||||
exporters: [prometheuspushgateway]
|
||||
+4
-5
@@ -38,10 +38,9 @@ autorestart=true
|
||||
stdout_logfile=/var/log/node_exporter.log
|
||||
stderr_logfile=/var/log/node_exporter-err.log
|
||||
|
||||
|
||||
[program:metrics_exporter]
|
||||
command=/usr/src/app/metric-exporter.sh
|
||||
[program:otelcol]
|
||||
command=/usr/local/bin/otelcol-contrib --config=/etc/otel/config.yaml
|
||||
autostart=true
|
||||
autorestart=true
|
||||
stdout_logfile=/var/log/metrics-exporter.log
|
||||
stderr_logfile=/var/log/metrics-exporter-err.log
|
||||
stderr_logfile=/var/log/supervisor/otelcol-err.log
|
||||
stdout_logfile=/var/log/supervisor/otelcol-out.log
|
||||
|
||||
Reference in New Issue
Block a user