Need to remove blank "instance" label in exporter metrics

[Vaibhav-1995]

What did you do

Getting one "instance" label which has no value in exporter metrics as below -

# HELP jvm_memory_pool_allocated_bytes_total Total bytes allocated in a given JVM memory pool. Only updated after GC, not continuously.
# TYPE jvm_memory_pool_allocated_bytes_total counter
jvm_memory_pool_allocated_bytes_total{pool="Eden Space",} 2.398098168E10
jvm_memory_pool_allocated_bytes_total{pool="CodeHeap 'profiled nmethods'",} 2.7943424E7
jvm_memory_pool_allocated_bytes_total{pool="CodeHeap 'non-profiled nmethods'",} 1.5028992E7
jvm_memory_pool_allocated_bytes_total{pool="Compressed Class Space",} 3559888.0
jvm_memory_pool_allocated_bytes_total{pool="Metaspace",} 3.3488968E7
jvm_memory_pool_allocated_bytes_total{pool="Tenured Gen",} 2.40252464E8
jvm_memory_pool_allocated_bytes_total{pool="Survivor Space",} 2.89199688E8
jvm_memory_pool_allocated_bytes_total{pool="CodeHeap 'non-nmethods'",} 1534848.0
# HELP jvm_info VM version info
# TYPE jvm_info gauge
jvm_info{runtime="OpenJDK Runtime Environment",vendor="Eclipse Adoptium",version="21.0.4+7-LTS",} 1.0
# HELP cwagent_logical_disk_free_space_average CloudWatch metric CWAgent LogicalDisk % Free Space Dimensions: [ImageId, InstanceId, InstanceType, instance, objectname] Statistic: Average Unit: None
# TYPE cwagent_logical_disk_free_space_average gauge
cwagent_logical_disk_free_space_average{job="cwagent",instance="",instance="C:",instance_id="i-xxxxx",image_id="ami-xxxxx",objectname="LogicalDisk",instance_type="t3.medium",} 78.53746032714844 1758697320000

What did you expect to see?

I don't want that blank "instance" label to come in exporter metrics as it is causing issue when prometheus scraping this metric it is not taking "instance" label in which actual value for drive name is present as "C" next to this blank "instance" label.

See below when prometheus scape this metric it drops "instance" label in which drive name is present -

{Account="xxx", Region="ap-south-1", __name__="cwagent_logical_disk_free_space_average", cluster_name="xxx", exported_job="cwagent", image_id="ami-xxx", instance_id="i-xxx", instance_type="m5a.xlarge", job="cloudwatch-exporter", objectname="LogicalDisk", prometheus="monitoring/xxx-prometheus-kube-prometheus", prometheus_replica="prometheus-low-ss-05-prometheus-kube-prometheus-0"}

Environment

• Exporter version: 0.16.0
• Operating system & architecture: EKS
• Running in containers? y
• Using the official image? y

Exporter configuration file

expand

image:
  repository: xxx/prometheus-cloudwatch-exporter
  tag: 0.16.0
  pullSecrets:
  - name: "xxx"

service:
  labels:
    logging: "true"
    app: as1-cloudwatch-exporter
    bu: corp
    svc: qa-prometheus-cloudwatch-exporter
    env: qa

pod:
  annotations:
    monitoring: "true"
    email: 'vaibhav.ingulkar@xxxx.com'
  labels:
    app: as1-cloudwatch-exporter 
    logging: "true"
    bu: corp
    svc: qa-prometheus-cloudwatch-exporter
    env: qa

resources:
  limits:
    cpu: 500m
    memory: 500Mi
  requests:
    cpu: 450m
    memory: 450Mi

serviceAccount:
  annotations: 
    eks.amazonaws.com/role-arn: "arn:aws:iam::xxxxx:role/qa-prom-cloudwatch-exporter-role"

config: |-
  # This is the default configuration for prometheus-cloudwatch-exporter
  region: ap-south-1
  delay_seconds: 60
  period_seconds: 3600
  metrics:
  - aws_namespace: CWAgent
    aws_metric_name: LogicalDisk % Free Space 
    aws_dimensions: [ImageId, InstanceId, InstanceType, instance, objectname ]
    aws_statistics: [Average]
    aws_tag_select:
      resource_type_selection: ec2:instance
      resource_id_dimension: InstanceId

tolerations:
- effect: NoSchedule
  key: type
  operator: Equal
  value: system

affinity:
  nodeAffinity:
    requiredDuringSchedulingIgnoredDuringExecution:
      nodeSelectorTerms:
        - matchExpressions:
            - key: type
              operator: In
              values:
                - system

priorityClassName: "system-cluster-critical"