[BUG] postfix.queue.size metric is lost after updating to 7.57.x

[int128]

Agent Environment
datadog-agent 7.57.2

Describe what happened:
When we updated the datadog-agent to 7.57.2, the metric of postfix.queue.size is lost. We reverted the version to 7.55.2 or 7.56.2, the metric is appeared again.

Describe what you expected:
postfix.queue.size metric is sent to Datadog after updating datadog-agent.

Steps to reproduce the issue:
Deploy the postfix container to Kubernetes cluster.

apiVersion: apps/v1
kind: StatefulSet
metadata:
  # ...snip...
spec:
  template:
    metadata:
      annotations:
        ad.datadoghq.com/postfix.checks: |
          {
            "postfix": {
              "init_config": {
                "postqueue": true
              },
              "instances": [
                {
                  "config_directory": "/etc/postfix",
                  "queues": [
                    "incoming",
                    "active",
                    "deferred"
                  ]
                }
              ]
            }
          }
    spec:
      containers:
        - image: smtp-relay-postfix
          imagePullPolicy: Always
          name: postfix
          # ...snip...
          volumeMounts:
            - name: queue
              mountPath: /var/spool/postfix
  volumeClaimTemplates:
    - metadata:
        name: queue
      spec:
        accessModes:
          - ReadWriteOnce
        storageClassName: gp2
        resources:
          requests:
            storage: 50Gi

Here is the list of versions and this problem.

• ❌ public.ecr.aws/datadog/agent:7.57.2 (Datadog Helm Chart 3.73.1)
• ✅ public.ecr.aws/datadog/agent:7.56.2 (Datadog Helm Chart 3.71.1)
• ✅ public.ecr.aws/datadog/agent:7.55.2 (Datadog Helm Chart 3.69.1)

Additional environment details (Operating System, Cloud provider, etc):

• Amazon EKS 1.30
• Datadog Helm Chart https://github.com/DataDog/helm-charts/tree/main/charts/datadog
• Postfix 3.4.14 (Debian)

[GustavoCaso]

Hi @int128

Thank you so much for opening the issue.
I've contacted the teams that own the different services involved in sending metrics from the integrations.

I will keep you posted with any update.

[GustavoCaso]

Hi @int128

After talking with the different teams, we concluded that is best for you to open a support ticket. Our support team would be able to assist you with your issue much better.

Let me know if you need any help opening a support ticket

[aoi1]

Hi! I’m a colleague of int128. I opened a support ticket and confirmed that upgrading the Datadog Agent to 7.58.1 resolved the issue. Thank you very much for your assistance.

[aoi1]

By the way, it would be very helpful if the Helm Chart default image tag would be 7.58.1. Thanks!