feat: Add o11y supports

all-in-one/Dockerfile

@@ -3,13 +3,21 @@ ARG BASE_VERSION=2022-10-27T19-02-22
 ARG CORE_VERSION=2.0.4
 ARG CONSOLE_VERSION=1.4.6
 ARG APISERVER_VERSION=0.0.16
+ARG PROMETHEUS_VERSION=v2.40.7
+ARG PROMTAIL_VERSION=2.9.4
+ARG LOKI_VERSION=2.9.4
+ARG GRAFANA_VERSION=9.3.6
 
-FROM ${HUB}/api-server:${APISERVER_VERSION} as apiserver
-FROM ${HUB}/higress:${CORE_VERSION} as controller
-FROM ${HUB}/pilot:${CORE_VERSION} as pilot
-FROM ${HUB}/gateway:${CORE_VERSION} as gateway
-FROM ${HUB}/console:${CONSOLE_VERSION} as console
-FROM eclipse-temurin:21-jre as jdk
+FROM ${HUB}/api-server:${APISERVER_VERSION} AS apiserver
+FROM ${HUB}/higress:${CORE_VERSION} AS controller
+FROM ${HUB}/pilot:${CORE_VERSION} AS pilot
+FROM ${HUB}/gateway:${CORE_VERSION} AS gateway
+FROM ${HUB}/console:${CONSOLE_VERSION} AS console
+FROM ${HUB}/prometheus:${PROMETHEUS_VERSION} AS prometheus
+FROM ${HUB}/promtail:${PROMTAIL_VERSION} AS promtail
+FROM ${HUB}/loki:${LOKI_VERSION} AS loki
+FROM ${HUB}/grafana:${GRAFANA_VERSION} AS grafana
+FROM ${HUB}/eclipse-temurin:21-jre AS jdk
 
 FROM ${HUB}/base:${BASE_VERSION}
 
@@ -21,12 +29,17 @@ COPY --from=controller /usr/local/bin/higress /usr/local/bin/higress
 
 # Install pilot
 COPY --from=pilot /usr/local/bin/pilot-discovery /usr/local/bin/pilot-discovery
-COPY --from=pilot /var/lib/istio/envoy/*.json /var/lib/istio/envoy/
 COPY --from=pilot /usr/local/bin/higress-pilot-start.sh /usr/local/bin/higress-pilot-start.sh
 
 # Install gateway
+COPY --from=gateway /var/lib/istio/envoy/*.json /var/lib/istio/envoy/
 COPY --from=gateway /usr/local/bin/pilot-agent /usr/local/bin/pilot-agent
 COPY --from=gateway /usr/local/bin/envoy /usr/local/bin/envoy
+COPY --from=gateway /usr/local/bin/higress-proxy-*.sh /usr/local/bin/
+RUN chmod a+x /usr/local/bin/higress-proxy-container-init.sh; \
+  sed -i 's/1337/0/g' /usr/local/bin/higress-proxy-container-init.sh; \
+  /usr/local/bin/higress-proxy-container-init.sh
+COPY --from=gateway /usr/local/bin/supercronic* /usr/local/bin/
 
 # Install console
 COPY --from=console /app /app
@@ -36,16 +49,28 @@ ENV JAVA_HOME=/opt/java/openjdk
 COPY --from=jdk $JAVA_HOME $JAVA_HOME
 ENV PATH="${JAVA_HOME}/bin:${PATH}"
 
-# Install supervisord and initialize related folders
-RUN apt-get update && \
-  apt-get install --no-install-recommends -y \
-  supervisor \
+# Install Prometheus
+COPY --from=prometheus /bin/prometheus /usr/local/bin/prometheus
+
+# Install Promtail
+COPY --from=promtail /usr/bin/promtail /usr/local/bin/promtail
+
+# Install Loki
+COPY --from=loki /usr/bin/loki /usr/local/bin/loki
+
+# Install Grafana
+COPY --from=grafana /usr/share/grafana /usr/share/grafana
+COPY --from=grafana /run.sh /usr/local/bin/grafana.sh
+
+# Install supervisord, logrotate, cron and initialize related folders
+RUN apt-get update --allow-unauthenticated && \
+  apt-get install --no-install-recommends -y --allow-unauthenticated \
+  supervisor logrotate cron \
+  && apt-get upgrade -y --allow-unauthenticated \
   && apt-get clean \
   && rm -rf /var/log/*log /var/lib/apt/lists/* /var/log/apt/* /var/lib/dpkg/*-old /var/cache/debconf/*-old \
   && mkdir -p /var/log/higress \
-  && chown 1337:1337 -R /var/log/higress \
-  && mkdir /data \
-  && chown 1337:1337 -R /data
+  && mkdir /data
 COPY ./supervisord/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
 
 # Initialize configurations
@@ -54,6 +79,10 @@ COPY ./gateway/podinfo /etc/istio/pod
 COPY ./scripts /usr/local/bin
 COPY ./apiserver/config /app/kubeconfig
 COPY ./config /opt/data/defaultConfig
+COPY ./prometheus /etc/prometheus
+COPY ./promtail /etc/promtail
+COPY ./loki /etc/loki
+COPY ./grafana /etc/grafana
 
 EXPOSE 8080 8443 8001
 

all-in-one/config/configmaps/higress-config.yaml

@@ -0,0 +1,21 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: higress-config
+  namespace: higress-system
+  creationTimestamp: "2000-01-01T00:00:00Z"
+  resourceVersion: "1"
+data:
+  higress: |-
+    downstream:
+      connectionBufferLimits: 32768
+      http2:
+        initialConnectionWindowSize: 1048576
+        initialStreamWindowSize: 65535
+        maxConcurrentStreams: 100
+      idleTimeout: 180
+      maxRequestHeadersKb: 60
+      routeTimeout: 0
+    upstream:
+      connectionBufferLimits: 10485760
+      idleTimeout: 10

all-in-one/get-ai-gateway.sh

@@ -156,7 +156,7 @@ runConfigWizard() {
     "Zhipu AI|ZHIPUAI"
     "Ollama|OLLAMA|configureOllamaProvider"
     "Claude|CLAUDE|configureClaudeProvider"
-    "Baidu AI Cloud|BAIDU"
+    # "Baidu AI Cloud|BAIDU"
     # "Tencent Hunyuan|HUNYUAN"
     "Stepfun|STEPFUN"
     "Minimax|MINIMAX|configureMinimaxProvider"
@@ -324,6 +324,7 @@ ${env}=${!env}"
   done
   cat <<EOF >$DATA_FOLDER/$CONFIG_FILENAME
 MODE=full
+o11y=on
 CONFIG_TEMPLATE=ai-gateway
 GATEWAY_HTTP_PORT=${GATEWAY_HTTP_PORT}
 GATEWAY_HTTPS_PORT=${GATEWAY_HTTPS_PORT}

all-in-one/grafana/grafana.ini

@@ -0,0 +1,21 @@
+[server]
+protocol=http
+domain=localhost
+root_url="%(protocol)s://%(domain)s/grafana"
+serve_from_sub_path=true
+
+[auth]
+disable_login_form=true
+disable_signout_menu=true
+
+[auth.anonymous]
+enabled=true
+org_name=Main Org.
+org_role=Viewer
+
+[users]
+default_theme=light
+viewers_can_edit=true
+
+[security]
+allow_embedding=true

all-in-one/loki/config.yaml

@@ -0,0 +1,49 @@
+auth_enabled: false
+common:
+  compactor_address: 'loki'
+  path_prefix: /var/loki
+  replication_factor: 1
+  storage:
+    filesystem:
+      chunks_directory: /var/loki/chunks
+      rules_directory: /var/loki/rules
+frontend:
+  scheduler_address: ""
+frontend_worker:
+  scheduler_address: ""
+index_gateway:
+  mode: simple
+limits_config:
+  max_cache_freshness_per_query: 10m
+  reject_old_samples: true
+  reject_old_samples_max_age: 168h
+  split_queries_by_interval: 15m
+memberlist:
+  join_members:
+  - localhost
+query_range:
+  align_queries_with_step: true
+ruler:
+  storage:
+    type: local
+runtime_config:
+  file: /etc/loki/runtime-config.yaml
+schema_config:
+  configs:
+  - from: "2022-01-11"
+    index:
+      period: 24h
+      prefix: loki_index_
+    object_store: filesystem
+    schema: v12
+    store: boltdb-shipper
+server:
+  http_listen_port: 3100
+  grpc_listen_port: 9095
+storage_config:
+  hedging:
+    at: 250ms
+    max_per_second: 20
+    up_to: 3
+tracing:
+  enabled: false

all-in-one/loki/runtime-config.yaml

@@ -0,0 +1 @@
+{}

all-in-one/meshConfig/mesh

@@ -16,6 +16,9 @@ dnsRefreshRate: 200s
 enableAutoMtls: false
 enablePrometheusMerge: true
 ingressControllerMode: "OFF"
+mseIngressGlobalConfig:
+  enableH3: false
+  enableProxyProtocol: false
 protocolDetectionTimeout: 100ms
 rootNamespace: higress-system
 trustDomain: cluster.local

all-in-one/prometheus/prometheus.yaml

@@ -0,0 +1,17 @@
+global:
+  scrape_interval:     15s 
+  evaluation_interval: 15s
+scrape_configs:
+  - job_name: 'prometheus'
+    metrics_path: /prometheus/metrics
+    static_configs:
+    - targets: ['localhost:9090']
+  - job_name: 'gateway'
+    metrics_path: /stats/prometheus
+    static_configs:
+    - targets: ['localhost:15020']
+      labels:
+        container: 'higress-gateway'
+        namespace: 'higress-system'
+        higress: 'higress-system-higress-gateway'
+        pod: 'higress'

all-in-one/promtail/promtail.yaml

@@ -0,0 +1,54 @@
+server:
+  log_level: info
+  http_listen_port: 3101
+  grpc_listen_port: 9096
+
+clients:
+- url: http://localhost:3100/loki/api/v1/push
+
+positions:
+  filename: /var/promtail/promtail-positions.yaml
+target_config:
+  sync_period: 10s
+scrape_configs:
+- job_name: access-logs
+  static_configs:
+  - targets:
+    - localhost
+    labels:
+      __path__: /var/log/proxy/access.log
+  pipeline_stages:
+  - json:
+      expressions:
+        authority:
+        method:
+        path:
+        protocol:
+        request_id:
+        response_code:
+        response_flags:
+        route_name:
+        trace_id:
+        upstream_cluster:
+        upstream_host:
+        upstream_transport_failure_reason:
+        user_agent:
+        x_forwarded_for:
+  - labels:
+      authority:
+      method:
+      path:
+      protocol:
+      request_id:
+      response_code:
+      response_flags:
+      route_name:
+      trace_id:
+      upstream_cluster:
+      upstream_host:
+      upstream_transport_failure_reason:
+      user_agent:
+      x_forwarded_for:
+  - timestamp:
+      source: timestamp
+      format: RFC3339Nano

all-in-one/scripts/base.sh

@@ -27,7 +27,6 @@ function readinessCheck() {
 
 function createDir() {
     sudo mkdir -p "$1"
-    sudo chown 1337:1337 "$1"
 }
 
 case $MODE in
@@ -39,9 +38,19 @@ case $MODE in
         MODE=full
         ;;
 esac
-
 echo "Mode=$MODE"
 
+case $O11Y in
+    true|TRUE|on|ON|yes|YES)
+        O11Y=on
+        ;;
+    *)
+        # Default to full mode
+        O11Y=off
+        ;;
+esac
+echo "O11Y=$O11Y"
+
 CONSOLE_USED_MARKER='/data/.console-used'
 CONSOLE_USED='false'
 if [ -f "$CONSOLE_USED_MARKER" ]; then

all-in-one/scripts/config-template/ai-gateway.sh

@@ -21,7 +21,7 @@ function initializeLlmProviderConfigs() {
   initializeLlmProviderConfig yi yi YI api.lingyiwanwu.com
   initializeLlmProviderConfig deepseek deepseek DEEPSEEK api.deepseek.com
   initializeLlmProviderConfig zhipuai zhipuai ZHIPUAI open.bigmodel.cn
-  initializeLlmProviderConfig baidu baidu BAIDU aip.baidubce.com
+  # initializeLlmProviderConfig baidu baidu BAIDU aip.baidubce.com
   # initializeLlmProviderConfig hunyuan hunyuan HUNYUAN hunyuan.tencentcloudapi.com 443 "https" "" "${EXTRA_CONFIGS[@]}"
   initializeLlmProviderConfig stepfun stepfun STEPFUN api.stepfun.com
   # initializeLlmProviderConfig cloudflare cloudflare CLOUDFLARE api.cloudflare.com 443 "https" "" "${EXTRA_CONFIGS[@]}"

all-in-one/scripts/start-console.sh

@@ -24,5 +24,12 @@ touch "$CONSOLE_USED_MARKER"
 
 set -e
 
-HIGRESS_CONSOLE_KUBE_CONFIG="/app/kubeconfig" SERVER_PORT="$CONSOLE_PORT" \
+if [ "$O11Y" == "on" ]; then
+    export HIGRESS_CONSOLE_DASHBOARD_BASE_URL="http://localhost:3000/grafana"
+    export HIGRESS_CONSOLE_DASHBOARD_DATASOURCE_PROM_URL="http://localhost:9090/prometheus"
+    export HIGRESS_CONSOLE_DASHBOARD_DATASOURCE_LOKI_URL="http://localhost:3100"
+fi
+
+HIGRESS_CONSOLE_KUBE_CONFIG="/app/kubeconfig" \
+    SERVER_PORT="$CONSOLE_PORT" \
     bash /app/start.sh

all-in-one/scripts/start-gateway.sh

@@ -17,8 +17,16 @@ set -e
 
 createDir /etc/istio/proxy
 createDir /var/lib/istio/data
+createDir /var/log/proxy
+touch /var/log/proxy/access.log
 
-/usr/local/bin/pilot-agent proxy router \
+if [ "$O11Y" == "on" ]; then
+    sed -i -E 's/^accessLogFile: .+$/accessLogFile: \/var\/log\/proxy\/access.log/' /etc/istio/config/mesh
+else
+    sed -i -E 's/^accessLogFile: .+$/accessLogFile: \/dev\/stdout/' /etc/istio/config/mesh
+fi
+
+/usr/local/bin/higress-proxy-start.sh proxy router \
     --domain=higress-system.svc.cluster.local \
     --proxyLogLevel=${GATEWAY_LOG_LEVEL:-warning} \
     --proxyComponentLogLevel=${GATEWAY_COMPONENT_LOG_LEVEL:-misc:error} \

all-in-one/scripts/start-grafana.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+cd "$(dirname -- "$0")"
+ROOT=$(pwd)
+cd - >/dev/null
+source $ROOT/base.sh
+
+if [ "$O11Y" != "on" ]; then
+    echo "Grafana won't run when o11y is not turned on."
+    sleep 2
+    exit 0
+fi
+
+set -e
+
+createDir /var/lib/grafana
+createDir /var/log/grafana
+
+GF_PATHS_CONFIG="/etc/grafana/grafana.ini" \
+    GF_PATHS_DATA="/var/lib/grafana" \
+    GF_PATHS_HOME="/usr/share/grafana" \
+    GF_PATHS_LOGS="/var/log/grafana" \
+    GF_PATHS_PLUGINS="/var/lib/grafana/plugins" \
+    GF_PATHS_PROVISIONING="/etc/grafana/provisioning" \
+    PATH="/usr/share/grafana/bin:$PATH" \
+    bash /usr/local/bin/grafana.sh