More updates in telegraf

Makefile

@@ -35,6 +35,64 @@ start:
 	echo "View logs with: make logs"; \
 	exit 1
 
+start-alloy-only:
+	@echo "Starting Grafana Alloy only..."
+	@$(DOCKER_COMPOSE) up grafana-alloy -d --build --force-recreate
+	@echo "Waiting for Grafana Alloy to become healthy..."
+	@timeout=60; \
+	while [ $$timeout -gt 0 ]; do \
+		status=$$(docker compose ps grafana-alloy --format json | jq -r '.Health // "starting"'); \
+		if [ "$$status" = "healthy" ]; then \
+			echo ""; \
+			echo "✅ Grafana Alloy is healthy!"; \
+			echo ""; \
+			echo "🚀 Grafana Alloy is ready:"; \
+			echo "   - gRPC endpoint: localhost:$${ALLOY_GRPC_PORT:-4317}"; \
+			echo "   - HTTP endpoint: localhost:$${ALLOY_HTTP_PORT:-4318}"; \
+			echo ""; \
+			exit 0; \
+		else \
+			printf "\r⏳ Waiting for Grafana Alloy... ($$timeout seconds remaining)    "; \
+			sleep 1; \
+			timeout=$$((timeout - 1)); \
+		fi; \
+	done; \
+	echo ""; \
+	echo "⚠️  Timeout waiting for Grafana Alloy to become healthy"; \
+	echo "Check service status with: docker compose ps grafana-alloy"; \
+	echo "View logs with: docker compose logs grafana-alloy"; \
+	exit 1
+
+start-telegraf-only:
+	@echo "Starting Telegraf only..."
+	@$(DOCKER_COMPOSE) up telegraf -d --build --force-recreate
+	@echo "Waiting for Telegraf to become healthy..."
+	@timeout=60; \
+	while [ $$timeout -gt 0 ]; do \
+		status=$$(docker compose ps telegraf --format json | jq -r '.Health // "starting"'); \
+		if [ "$$status" = "healthy" ]; then \
+			echo ""; \
+			echo "✅ Telegraf is healthy!"; \
+			echo ""; \
+			echo "🚀 Telegraf is ready:"; \
+			echo "   - UDP endpoint: localhost:$${TELEGRAF_PORT:-8094}"; \
+			echo "   - TCP endpoint: localhost:$${TELEGRAF_PORT:-8094}"; \
+			echo ""; \
+			exit 0; \
+		else \
+			printf "\r⏳ Waiting for Telegraf... ($$timeout seconds remaining)    "; \
+			sleep 1; \
+			timeout=$$((timeout - 1)); \
+		fi; \
+	done; \
+	echo ""; \
+	echo "⚠️  Timeout waiting for Telegraf to become healthy"; \
+	echo "Check service status with: docker compose ps telegraf"; \
+	echo "View logs with: docker compose logs telegraf"; \
+	exit 1
+
+make start-agents-only: start-telegraf-only start-alloy-only
+
 ## Stop all services
 stop:
 	@echo "Shutting down services..."

docker-compose.yml

@@ -86,7 +86,7 @@ services:
       timeout: 30s
       retries: 30
   telegraf:
-    image: telegraf:1.27
+    image: telegraf:1.34
     depends_on:
       - influxdb
     volumes:

telegraf/telegraf.conf

@@ -1,47 +1,93 @@
+[global_tags]
+  # This will be used for checking eligibility for alerting
+  # Currently there will be 3: sov-dev, sov-int-staging, sov-testnet. Only sov-testnet will have alerts
+  environment = "sov-dev"
+  # Optional, can be helpful in alerts: primary-mock-da, secondary-celestia
+  # role = "mock-da"
 [agent]
-  interval = "5s"
+  interval = "30s"
   round_interval = true
-  debug = true
-  quiet = false
-  metric_batch_size = 1000
-  metric_buffer_limit = 10000
-  flush_interval = "5s"
+  metric_batch_size = 5000
+  metric_buffer_limit = 50000
+  collection_jitter = "0s"
+  flush_interval = "30s"
+  flush_jitter = "0s"
+  precision = "0s"
   hostname = "dev-docker"
-  precision = ""
 
 [[outputs.influxdb_v2]]
+  # Adjust for staging/testnet environment
   urls = ["http://influxdb:8086"]
   organization = "sovereign"
   bucket = "sov-rollup"
   token = "sovereign"
   insecure_skip_verify = true
 
+# Collectors
 [[inputs.socket_listener]]
   service_address = "udp://:8094"
   data_format = "influx"
 
 [[inputs.socket_listener]]
-    service_address = "tcp://:8094"
-    data_format = "influx"
+  service_address = "tcp://:8094"
+  data_format = "influx"
 
+# Basic
 [[inputs.cpu]]
   percpu = true
   totalcpu = true
-
-[[inputs.system]]
-
+  collect_cpu_time = false
+  report_active = falsecore_tags = false
 [[inputs.mem]]
 
+# Storage
 [[inputs.disk]]
-  ignore_fs = ["tmpfs", "devtmpfs"]
+  ignore_fs = ["tmpfs", "devtmpfs", "devfs", "iso9660", "overlay", "aufs", "squashfs"]
+  # Adjust per server
   mount_points = ["/", "/mnt/rollup"]
-
 [[inputs.diskio]]
-
 [[inputs.filecount]]
-   directories = ["/mnt", "/mnt/rollup", "/mnt/rollup/state", "/mnt/rollup/accessory", "/mnt/rollup/ledger", "/mnt/rollup/preferred_sequencer"]
+  recursive = true
+  size = "100B"
+  # Adjust per server
+  directories = [ "/mnt", "/mnt/rollup/**" ]
 
+# System
+[[inputs.system]]
+[[inputs.swap]]
+[[inputs.kernel]]
+[[inputs.interrupts]]
+[[inputs.processes]]
+
+# Network
 [[inputs.net]]
-    ignore_protocol_stats = true
+  interfaces = ["*"]
+  ignore_protocol_stats = true
+[[inputs.netstat]]
 
-[[inputs.processes]]
+# Misc
+[[inputs.procstat]]
+  pattern = "rollup"
+  pid_finder = "native"
+  # Collect all available metrics
+  fieldpass = ["*"]
+# Telegraf self-monitoring
+[[inputs.internal]]
+[[inputs.http_response]]
+  interval = "2s"
+  urls = ["http://127.0.0.1:12346/sequencer/ready"]
+
+#[[inputs.postgresql]]
+#  address = "host=localhost user=rollup password=hunter2 sslmode=disable"
+# Rollup, legacy
+#[[inputs.prometheus]]
+#  urls = ["http://127.0.0.1:9845/metrics"]
+#   metric_version = 1
+#   [inputs.prometheus.tags]
+#     source = "sov-rollup"
+# Alloy
+#[[inputs.prometheus]]
+#  urls = ["http://127.0.0.1:12345/metrics"]
+#  metric_version = 2
+#  [inputs.prometheus.tags]
+#    source = "alloy"