Add liveness/readiness probe to operator deployment

[duhminick]

Issue #, if available: N/A

Description of changes:

• Add readiness probe
  • Uses /readyz
• Add liveness probe
  • Uses /healthz

The operator used/uses kubebuilder to create the operator. Kubebuilder did not include liveness/readiness probes for the deployment at the time the operator was created. In more recent versions, it is included: kubernetes-sigs/kubebuilder#1856. Just following the same pattern, i.e. path, port, initial delay seconds, period seconds.

Testing

Happy path

$ helm upgrade \
    --wait \
    --create-namespace \
    --namespace amazon-cloudwatch \
    amazon-cloudwatch-observability \
    ./charts/amazon-cloudwatch-observability \
    -f custom-values.yaml
Release "amazon-cloudwatch-observability" has been upgraded. Happy Helming!
NAME: amazon-cloudwatch-observability
LAST DEPLOYED: Thu Oct  2 21:25:34 2025
NAMESPACE: amazon-cloudwatch
STATUS: deployed
REVISION: 49
TEST SUITE: None

The deployment is in a ready state after the probe succeeds.

Unhappy path

If you change the port/path to something incorrect, the Helm upgrade will fail with

Error: UPGRADE FAILED: context deadline exceeded

you will also get this in the events for the deployment as well:

Warning  Unhealthy  2s (x4 over 32s)  kubelet            Readiness probe failed: Get "http://192.168.18.209:808/re │
│ adyz": dial tcp 192.168.18.209:808: connect: connection refused    

The deployment stays in a not ready state.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.