feat: add more host metrics

iac/provider-gcp/nomad/configs/otel-collector.yaml

@@ -14,11 +14,11 @@ receivers:
         - job_name: nomad
           scrape_interval: 15s
           scrape_timeout: 5s
-          metrics_path: '/v1/metrics'
+          metrics_path: "/v1/metrics"
           static_configs:
-            - targets: ['localhost:4646']
+            - targets: ["localhost:4646"]
           params:
-            format: ['prometheus']
+            format: ["prometheus"]
 
   hostmetrics:
     root_path: /hostfs
@@ -30,15 +30,15 @@ receivers:
     scrapers:
       cpu:
         metrics:
-          system.cpu.time:  # counter of seconds spent
+          system.cpu.time: # counter of seconds spent
             enabled: true
           system.cpu.frequency:
             enabled: false
           system.cpu.logical.count:
             enabled: true
           system.cpu.physical.count:
             enabled: true
-          system.cpu.utilization:  # percentage utilized in the moment
+          system.cpu.utilization: # percentage utilized in the moment
             enabled: false
 
       # entirely disabled, so we omit the module
@@ -89,7 +89,7 @@ receivers:
             enabled: true
           system.linux.memory.available: # free memory
             enabled: true
-          system.linux.memory.dirty:  # data waiting to be written to disk
+          system.linux.memory.dirty: # data waiting to be written to disk
             enabled: true
           system.memory.limit: # total memory
             enabled: true
@@ -156,36 +156,37 @@ receivers:
             enabled: true
 
       # entirely disabled, so we omit the module
-      #process:
-      #  metrics:
-      #    process.cpu.time:
-      #      enabled: false
-      #    process.disk.io:
-      #      enabled: false
-      #    process.memory.usage:
-      #      enabled: false
-      #    process.memory.virtual:
-      #      enabled: false
-      #    process.context_switches:
-      #      enabled: false
-      #    process.cpu.utilization:
-      #      enabled: false
-      #    process.disk.operations:
-      #      enabled: false
-      #    process.handles:
-      #      enabled: false
-      #    process.memory.utilization:
-      #      enabled: false
-      #    process.open_file_descriptors:
-      #      enabled: false
-      #    process.paging.faults:
-      #      enabled: false
-      #    process.signals_pending:
-      #      enabled: false
-      #    process.threads:
-      #      enabled: false
-      #    process.uptime:
-      #      enabled: false
+      process:
+        mute_process_all_errors: true
+        metrics:
+          process.cpu.time:
+            enabled: false
+          process.disk.io:
+            enabled: false
+          process.memory.usage:
+            enabled: false
+          process.memory.virtual:
+            enabled: false
+          process.context_switches:
+            enabled: true
+          process.cpu.utilization:
+            enabled: false
+          process.disk.operations:
+            enabled: false
+          process.handles:
+            enabled: false
+          process.memory.utilization:
+            enabled: false
+          process.open_file_descriptors:
+            enabled: true
+          process.paging.faults:
+            enabled: false
+          process.signals_pending:
+            enabled: false
+          process.threads:
+            enabled: true
+          process.uptime:
+            enabled: false
 
       # entirely disabled, so we omit the module
       #system:
@@ -253,6 +254,36 @@ processors:
               - state
             aggregation_type: sum
 
+  metricstransform/aggregate_process_metrics:
+    transforms:
+      - include: "process.open_file_descriptors"
+        match_type: strict
+        action: update
+        new_name: "system.open_file_descriptors"
+        operations:
+          - action: aggregate_labels
+            label_set:
+              - node.id
+            aggregation_type: sum
+      - include: "process.context_switches"
+        match_type: strict
+        action: update
+        new_name: "system.context_switches"
+        operations:
+          - action: aggregate_labels
+            label_set:
+              - node.id
+            aggregation_type: sum
+      - include: "process.threads"
+        match_type: strict
+        action: update
+        new_name: "system.threads"
+        operations:
+          - action: aggregate_labels
+            label_set:
+              - node.id
+            aggregation_type: sum
+
   resourcedetection:
     # https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/processor/resourcedetectionprocessor
     detectors: [gcp]
@@ -415,6 +446,7 @@ service:
       processors:
         - attributes/host_metrics_node
         - metricstransform/single_cpu
+        - metricstransform/aggregate_process_metrics
         - resourcedetection
         - transform/set-name
         - batch