If a *virt-launcher* pod's logs contain more than one line with a target string, some log lines are overwritten and lost when data is recorded to files. This situation was not properly replicated in a standard procedure, but it may occur when creating VMs in a custom way.
    

    1. Create a stub pod with custom logs (see below)
    2. Trigger a new gathering from the Insights Operator
    3. Check the archive generated
    

apiVersion: v1
kind: Pod
metadata:
  name: test-gather-custom-pod
  namespace: openshift-insights
  labels:
    kubevirt.io: virt-launcher
spec:
  containers:
    - name: test
      image: quay.io/openshifttest/busybox
      imagePullPolicy: Always
      command:
        - sh
        - -c
        - |
          echo ' 
          {"component":"virt-launcher","level":"info","msg":"LC_ALL=C PATH=/usr/libexec/qemu-kvm -name guest=vm1,debug-threads=on ...","timestamp":"2025-10-30T06:01:20Z"}
          {"component":"virt-launcher","level":"info","msg":"Starting QEMU process: /usr/libexec/qemu-kvm -machine pc-q35-rhel9.4.0,accel=kvm,usb=off ...","timestamp":"2025-10-30T06:01:21Z"}
          {"component":"virt-launcher","level":"info","msg":"QEMU command line: /usr/libexec/qemu-kvm -cpu Cascadelake-Server -drive file=/var/run/kubevirt...","timestamp":"2025-10-30T06:01:22Z"}
          ' && sleep 600;
      securityContext:
        allowPrivilegeEscalation: false
        capabilities:
          drop:
            - ALL
  restartPolicy: Always
  securityContext:
    runAsNonRoot: true
    seccompProfile:
      type: RuntimeDefault

Only the last log line is recorded
    

All the log lines with the target string are gathered
    

This is a fictional scenario, but it could happen in real life.