Failed to get metrics when install security profiles operator to a non-default namespace. TargetDown alert was fired for the security-profiles-operator. The alert details showed: "100% of the metrics/metrics targets in NS security-profiles-operator namespace have been unreachable for more than 15 minutes. This may be a symptom of network connectivity issues, down nodes, or failures within these components. Assess the health of the infrastructure and nodes running these targets and then contact support."

4.12.0-rc.6 + security-profiles-operator-bundle-container-0.5.0-62

Alwayso

1.Install security profiles operator to a namespace called security-profiles-operator

2. Create a seccompprofile: 
$ oc apply -f -<<EOF

apiVersion: security-profiles-operator.x-k8s.io/v1beta1
kind: SeccompProfile
metadata:
  name: sleep-sh-pod
  Namespace: security-profiles-operator
spec:
  defaultAction: SCMP_ACT_ERRNO
  architectures:
  - SCMP_ARCH_X86_64
  syscalls:
  - action: SCMP_ACT_ALLOW
    names:
    - arch_prctl
    - brk
    - capget
    - capset
    - chdir
    - clone
    - close
    - dup3
    - epoll_ctl
    - epoll_pwait
    - execve
    - exit_group
    - fchdir
    - fchown
    - fcntl
    - fstat
    - fstatfs
    - futex
    - getcwd
    - getdents64
    - getpid
    - getppid
    - getuid
    - ioctl
    - lseek
    - mmap
    - mount
    - mprotect
    - nanosleep
    - newfstatat
    - open
    - openat
    - pivot_root
    - prctl
    - read
    - rt_sigaction
    - rt_sigprocmask
    - rt_sigreturn
    - set_tid_address
    - setgid
    - setgroups
    - sethostname
    - setuid
    - stat
    - statfs
    - tgkill
    - time
    - umask
    - umount2
    - wait4
    - write
    - mkdir
    - mkdirat
EOF

3. Check the metrics through CMDLine and GUI 

The metrics couldn't show for CMLine nor GUI.
$ oc run --rm -i --restart=Never --image=registry.fedoraproject.org/fedora-minimal:latest -n security-profiles-operator metrics-test -- bash -c 'curl -ks -H "Authorization: Bearer $(cat /var/run/secrets/kubernetes.io/serviceaccount/token)" https://metrics.security-profiles-operator/metrics-spod'
Forbidden (user=system:serviceaccount:security-profiles-operator:default, verb=get, resource=, subresource=)
pod "metrics-test" deleted
In GUI, click Observe > alert, the targetDown alert was fired for the security-profiles-operator namespace. The alert details showed: "100% of the metrics/metrics targets in NS security-profiles-operator namespace have been unreachable for more than 15 minutes. This may be a symptom of network connectivity issues, down nodes, or failures within these components. Assess the health of the infrastructure and nodes running these targets and then contact support."
In GUI, click Observe > Metrics, run query "security_profiles_operator_seccomp_profile_total", it will prompt "No datapoints found".

The metrics could show both for CMLine and GUI.

The metrics works when inatll security profiles operator into the default openshift-security-profiles operator namespace