feat: add observability tooling for Flashbox images

bob-common/mkosi.build

@@ -52,3 +52,11 @@ build_rust_package \
     "input-only-proxy" \
     "v0.0.2" \
     "https://github.com/flashbots/input-only-proxy"
+
+# Build gomplate (template engine for Prometheus config)
+make_git_package \
+    "gomplate" \
+    "v4.3.3" \
+    "https://github.com/hairyhenderson/gomplate" \
+    'go build -trimpath -ldflags "-s -w -buildid=" -o ./build/gomplate ./cmd/gomplate' \
+    "build/gomplate:/usr/bin/gomplate"

bob-common/mkosi.conf

@@ -29,6 +29,9 @@ Packages=podman
          openssh-sftp-server
          udev
          libsnappy1v5
+         prometheus
+         prometheus-node-exporter
+         prometheus-process-exporter
 
 BuildPackages=build-essential
               git

bob-common/mkosi.extra/etc/prometheus/process-exporter.yml

@@ -0,0 +1,5 @@
+process_names:
+  # Monitor the searcher container (conmon + all children via --children flag)
+  - name: "searcher-container"
+    cmdline:
+    - 'conmon.*searcher-container'

bob-common/mkosi.extra/etc/prometheus/prometheus.yml.tmpl

@@ -0,0 +1,43 @@
+global:
+  scrape_interval: 15s
+  evaluation_interval: 15s
+
+# Recording rules for aggregated metrics
+rule_files:
+  - /etc/prometheus/recording_rules.yml
+
+# Scrape configurations
+scrape_configs:
+  # Node exporter on localhost
+  - job_name: 'node'
+    static_configs:
+      - targets: ['localhost:9100']
+    metric_relabel_configs:
+      # Only keep aggregated metrics for remote write
+      - source_labels: [__name__]
+        regex: 'node_(cpu|memory|disk|filesystem|network)_.*'
+        action: keep
+
+  # Process exporter for container monitoring
+  - job_name: 'process'
+    static_configs:
+      - targets: ['localhost:9256']
+
+{{- $config := (datasource "config") }}
+{{- if $config.remote_write_flashbots_url }}
+
+# Remote write configuration (dynamically configured)
+remote_write:
+  # Flashbots endpoint
+  - url: {{ $config.remote_write_flashbots_url }}
+    write_relabel_configs:
+      # Only send aggregated metrics
+      - source_labels: [__name__]
+        regex: 'flashbox:.*'
+        action: keep
+    {{- if $config.remote_write_flashbots_auth }}
+    basic_auth:
+      username: {{ $config.remote_write_flashbots_username }}
+      password: {{ $config.remote_write_flashbots_password }}
+    {{- end }}
+{{- end }}

bob-common/mkosi.extra/etc/prometheus/recording_rules.yml

@@ -0,0 +1,49 @@
+groups:
+  - name: flashbox_aggregated_metrics
+    interval: 30s  # How often to evaluate rules
+    rules:
+      # CPU aggregated metrics
+      - record: flashbox:cpu_usage_percent
+        expr: 100 - (avg(rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
+
+      - record: flashbox:cpu_usage_percent_by_mode
+        expr: avg(rate(node_cpu_seconds_total[5m])) by (mode) * 100
+
+      # Memory aggregated metrics
+      - record: flashbox:memory_usage_percent
+        expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes)) * 100
+
+      - record: flashbox:memory_available_gb
+        expr: node_memory_MemAvailable_bytes / 1024 / 1024 / 1024
+
+      # Disk aggregated metrics - both root and persistent
+      # Root filesystem (always available)
+      - record: flashbox:disk_usage_percent_root
+        expr: 100 - (node_filesystem_avail_bytes{mountpoint="/"} / node_filesystem_size_bytes{mountpoint="/"} * 100)
+
+      # Persistent storage (available after mount) - returns no data if not mounted
+      - record: flashbox:disk_usage_percent_persistent  
+        expr: 100 - (node_filesystem_avail_bytes{mountpoint="/persistent"} / node_filesystem_size_bytes{mountpoint="/persistent"} * 100)
+
+      - record: flashbox:disk_io_read_mb_per_sec
+        expr: rate(node_disk_read_bytes_total[5m]) / 1024 / 1024
+
+      - record: flashbox:disk_io_write_mb_per_sec
+        expr: rate(node_disk_written_bytes_total[5m]) / 1024 / 1024
+
+      # Container health metrics (using process exporter)
+      - record: flashbox:container_alive
+        expr: up{job="process"} * on(instance) group_left(cgroup) namedprocess_namegroup_num_procs{groupname=~".*searcher-container.*"}
+
+      - record: flashbox:container_cpu_percent
+        expr: rate(namedprocess_namegroup_cpu_seconds_total{groupname=~".*searcher-container.*"}[5m]) * 100
+
+      - record: flashbox:container_memory_mb
+        expr: namedprocess_namegroup_memory_bytes{groupname=~".*searcher-container.*"} / 1024 / 1024
+
+      # Network metrics (only counters, no detailed info)
+      - record: flashbox:network_receive_mb_total
+        expr: sum(node_network_receive_bytes_total) / 1024 / 1024
+
+      - record: flashbox:network_transmit_mb_total
+        expr: sum(node_network_transmit_bytes_total) / 1024 / 1024

bob-common/mkosi.extra/etc/systemd/system/node-exporter.service

@@ -0,0 +1,56 @@
+[Unit]
+Description=Prometheus Node Exporter
+Documentation=https://github.com/prometheus/node_exporter
+After=network-online.target
+Wants=network-online.target
+
+[Service]
+Type=simple
+User=prometheus
+Group=prometheus
+ExecStart=/usr/bin/prometheus-node-exporter \
+  --web.listen-address=127.0.0.1:9100 \
+  --collector.cpu \
+  --collector.meminfo \
+  --collector.diskstats \
+  --collector.filesystem \
+  --collector.netdev \
+  --collector.loadavg \
+  --no-collector.arp \
+  --no-collector.bcache \
+  --no-collector.bonding \
+  --no-collector.conntrack \
+  --no-collector.cpufreq \
+  --no-collector.edac \
+  --no-collector.entropy \
+  --no-collector.filefd \
+  --no-collector.hwmon \
+  --no-collector.infiniband \
+  --no-collector.ipvs \
+  --no-collector.mdadm \
+  --no-collector.netclass \
+  --no-collector.netstat \
+  --no-collector.nfs \
+  --no-collector.nfsd \
+  --no-collector.pressure \
+  --no-collector.rapl \
+  --no-collector.schedstat \
+  --no-collector.sockstat \
+  --no-collector.softnet \
+  --no-collector.stat \
+  --no-collector.textfile \
+  --no-collector.thermal_zone \
+  --no-collector.time \
+  --no-collector.timex \
+  --no-collector.udp_queues \
+  --no-collector.uname \
+  --no-collector.vmstat \
+  --no-collector.xfs \
+  --no-collector.zfs \
+  --no-collector.systemd \
+  --collector.filesystem.mount-points-exclude=^/(dev|proc|sys|run|var/lib/docker)($|/)
+Restart=on-failure
+RestartSec=5s
+
+[Install]
+WantedBy=multi-user.target

bob-common/mkosi.extra/etc/systemd/system/process-exporter.service

@@ -0,0 +1,19 @@
+[Unit]
+Description=Prometheus Process Exporter
+Documentation=https://github.com/ncabatoff/process-exporter
+After=network-online.target searcher-container.service
+Wants=network-online.target
+
+[Service]
+Type=simple
+User=prometheus
+Group=prometheus
+ExecStart=/usr/bin/prometheus-process-exporter \
+  --web.listen-address=127.0.0.1:9256 \
+  --config.path=/etc/prometheus/process-exporter.yml \
+  --children
+Restart=on-failure
+RestartSec=5s
+
+[Install]
+WantedBy=multi-user.target

bob-common/mkosi.extra/etc/systemd/system/prometheus.service

@@ -0,0 +1,25 @@
+[Unit]
+Description=Prometheus Monitoring System
+Documentation=https://prometheus.io/docs/introduction/overview/
+After=network-online.target fetch-config.service
+Wants=network-online.target
+Requires=fetch-config.service
+
+[Service]
+Type=simple
+User=prometheus
+Group=prometheus
+ExecStartPre=+/usr/bin/gomplate -f /etc/prometheus/prometheus.yml.tmpl -o /etc/prometheus/prometheus.yml -d config=/etc/flashbox/observability-config.json
+ExecStart=/usr/bin/prometheus \
+  --config.file=/etc/prometheus/prometheus.yml \
+  --storage.tsdb.path=/var/lib/prometheus/ \
+  --storage.tsdb.retention.time=24h \
+  --web.console.templates=/usr/share/prometheus/consoles \
+  --web.console.libraries=/usr/share/prometheus/console_libraries \
+  --web.listen-address=127.0.0.1:9090
+ExecReload=/bin/kill -HUP $MAINPID
+Restart=on-failure
+RestartSec=5s
+
+[Install]
+WantedBy=multi-user.target

bob-common/mkosi.extra/etc/systemd/system/searcher-firewall.service

@@ -1,7 +1,7 @@
 [Unit]
 Description=Searcher Network and Firewall Rules
-After=network.target network-setup.service
-Requires=network-setup.service
+After=network.target network-setup.service fetch-config.service
+Requires=network-setup.service fetch-config.service
 
 [Service]
 Type=oneshot

bob-common/mkosi.extra/usr/bin/fetch-config.sh

@@ -0,0 +1,157 @@
+#!/bin/sh
+set -eu -o pipefail
+
+# Common configuration fetching script for FlashBox (bob-l1 and bob-l2)
+# This script provides shared functionality for configuration management
+# Project-specific configuration should be done via /etc/bob/dynamic-config.sh
+
+CONFIG_PATH=/etc/bob/config.env
+OBSERVABILITY_CONFIG_PATH=/etc/flashbox/observability-config.json
+
+# Don't override if config already exists
+if [ -f "$CONFIG_PATH" ]; then
+    echo "Config already exists at $CONFIG_PATH, skipping"
+    exit 0
+fi
+
+# Helper functions
+fetch_metadata_value() {
+    curl -s \
+        --header "Metadata-Flavor: Google" \
+        "http://metadata/computeMetadata/v1/instance/attributes/$1"
+}
+
+get_ips_from_uris() {
+    # Extract IP addresses from URIs
+    echo "$1" | grep -oE '[0-9]{1,3}(\.[0-9]{1,3}){3}' || echo ""
+}
+
+write_observability_config() {
+    local metrics_flashbots_url="$1"
+    local metrics_flashbots_username="$2"
+    local metrics_flashbots_password="$3"
+
+    # Extract IP for firewall rules
+    local metrics_endpoint=""
+
+    if [ -n "$metrics_flashbots_url" ]; then
+        metrics_endpoint=$(get_ips_from_uris "$metrics_flashbots_url" | head -1)
+    fi
+
+    # Append observability config to main config
+    cat <<EOF >> "$CONFIG_PATH"
+CONFIG_METRICS_FLASHBOTS_URL='${metrics_flashbots_url}'
+CONFIG_METRICS_FLASHBOTS_USERNAME='${metrics_flashbots_username}'
+CONFIG_METRICS_FLASHBOTS_PASSWORD='${metrics_flashbots_password}'
+METRICS_ENDPOINT='${metrics_endpoint}'
+EOF
+
+    # Create observability config for Prometheus (always needed for gomplate templating)
+    mkdir -p /etc/flashbox
+    cat <<EOF > "$OBSERVABILITY_CONFIG_PATH"
+{
+  "remote_write_flashbots_url": "${metrics_flashbots_url}",
+  "remote_write_flashbots_username": "${metrics_flashbots_username}",
+  "remote_write_flashbots_password": "${metrics_flashbots_password}",
+  "remote_write_flashbots_auth": $([ -n "${metrics_flashbots_username}" ] && echo '"true"' || echo '""')
+}
+EOF
+    echo "Observability configuration written to $OBSERVABILITY_CONFIG_PATH"
+}
+
+# Check for local QEMU development environment
+if dmidecode -s system-manufacturer 2>/dev/null | grep -q "QEMU" && \
+   [ -f /etc/systemd/system/serial-console.service ]; then
+    echo "Running in local QEMU dev image, using default test values"
+
+    # Get default gateway (host in QEMU user-mode networking)
+    GATEWAY=$(ip route | awk '/default/ {print $3}')
+    if [ -z "$GATEWAY" ]; then
+        echo "Warning: Could not detect gateway, falling back to 10.0.2.2"
+        GATEWAY="10.0.2.2"
+    fi
+
+    # Export gateway for custom script
+    export GATEWAY
+
+    # Call project-specific configuration if it exists
+    if [ -x /etc/bob/dynamic-config.sh ]; then
+        echo "Running project-specific configuration..."
+        /etc/bob/dynamic-config.sh qemu "$CONFIG_PATH"
+    else
+        echo "Warning: No project-specific configuration found at /etc/bob/dynamic-config.sh"
+    fi
+
+    # Add empty observability config for local dev
+    write_observability_config "" "" ""
+
+    exit 0
+fi
+
+# Production configuration using Vault
+echo "Fetching configuration from Vault..."
+
+# Get instance metadata
+instance_name=$(fetch_metadata_value "name")
+vault_addr=$(fetch_metadata_value "vault_addr")
+vault_auth_mount=$(fetch_metadata_value "vault_auth_mount_gcp")
+vault_kv_path=$(fetch_metadata_value "vault_kv_path")
+vault_kv_common_suffix=$(fetch_metadata_value "vault_kv_common_suffix")
+
+# Authenticate with Vault using GCP identity
+gcp_token=$(curl \
+  --header "Metadata-Flavor: Google" \
+  --data-urlencode "audience=http://vault/$instance_name" \
+  --data-urlencode "format=full" \
+  "http://metadata/computeMetadata/v1/instance/service-accounts/default/identity")
+
+vault_token=$(curl \
+    --data "$(printf '{"role":"%s","jwt":"%s"}' "$instance_name" "$gcp_token")" \
+    "${vault_addr}/v1/${vault_auth_mount}/login" | \
+    jq -r .auth.client_token)
+
+# Fetch common and instance-specific data
+common_data=$(curl \
+    --header "X-Vault-Token: ${vault_token}" \
+    "${vault_addr}/v1/${vault_kv_path}/node/${vault_kv_common_suffix}" |
+    jq -c .data.data)
+
+secret_data=$(curl \
+    --header "X-Vault-Token: ${vault_token}" \
+    "${vault_addr}/v1/${vault_kv_path}/node/${instance_name}" |
+    jq -c .data.data)
+
+# Merge data objects
+data=$(echo "$common_data $secret_data" | jq -s 'add')
+
+# Helper to get values from merged data
+get_data_value() {
+    echo "$data" | jq -rc --arg key "$1" '.[$key] // ""'
+}
+
+# Export data for project-specific script
+export VAULT_DATA="$data"
+export -f get_data_value
+export -f get_ips_from_uris
+
+# Call project-specific configuration
+if [ -x /etc/bob/dynamic-config.sh ]; then
+    echo "Running project-specific configuration..."
+    /etc/bob/dynamic-config.sh vault "$CONFIG_PATH"
+else
+    echo "Error: No project-specific configuration found at /etc/bob/dynamic-config.sh"
+    exit 1
+fi
+
+# Fetch observability configuration
+metrics_flashbots_url=$(get_data_value metrics_flashbots_url)
+metrics_flashbots_username=$(get_data_value metrics_flashbots_username)
+metrics_flashbots_password=$(get_data_value metrics_flashbots_password)
+
+# Write observability configuration
+write_observability_config \
+    "$metrics_flashbots_url" \
+    "$metrics_flashbots_username" \
+    "$metrics_flashbots_password"
+
+echo "Configuration successfully fetched and written to $CONFIG_PATH"