Use a more flexible health probe bind address to enable both IPv4 and…

[ndebuhr]

Description

This PR proposes a more flexible health probe bind address to enable both IPv4 and IPv6 environments.

Currently, the sample deployment and default Helm chart installation fail in IPv4 environments - the fdb-operator pod crash loops with the following error:

kubectl logs fdb-operator-869cb4684d-8f5vz -n fdb-operator
Defaulted container "manager" out of: manager, foundationdb-kubernetes-init-7-1 (init), foundationdb-kubernetes-init-7-3 (init), foundationdb-kubernetes-init-7-4 (init)
{"level":"info","ts":"2025-10-30T01:03:58Z","logger":"setup","msg":"Operator starting in single namespace mode","namespace":"fdb-operator"}
{"level":"error","ts":"2025-10-30T01:03:58Z","logger":"setup","msg":"unable to start manager","error":"error listening on [::1]:9443: listen tcp [::1]:9443: bind: cannot assign requested address","stacktrace":"github.com/FoundationDB/fdb-kubernetes-operator/v2/setup.StartManager\n\t/workspace/setup/setup.go:429\nmain.main\n\t/workspace/main.go:58\nruntime.main\n\t/usr/local/go/src/runtime/proc.go:283"}

This PR replaces the hard-coded IPv6 localhost address with a localhost spec that binds to whichever address family the system resolver prefers.

Type of change

Bug fix (non-breaking change which fixes an issue).

Discussion

We could alternatively update the Helm chart and/or deployment sample to override via the --health-probe-bind-address flag, but I think an IP-stack-agnostic default value probably makes the most sense.

Testing

Tested manually on IPv4 (Google Kubernetes Engine standard cluster), and the change fixes the crash looping.

kubectl logs fdb-operator-558dbbc867-
l4527 -n fdb-operator --follow
Defaulted container "manager" out of: manager, foundationdb-kubernetes-init-7-1 (init), foundationdb-kubernetes-init-7-3 (init), foundationdb-kubernetes-init-7-4 (init)
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"Operator starting in single namespace mode","namespace":"fdb-operator"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"Could not parse version from directory name","name":"7.1"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"Could not parse version from directory name","name":"7.4"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"Could not parse version from directory name","name":"7.3"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"Updating pod update method","podUpdateMethod":"update"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"setup","msg":"setup manager"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"controller-runtime.metrics","msg":"Starting metrics server"}
{"level":"info","ts":"2025-10-30T01:43:04Z","logger":"controller-runtime.metrics","msg":"Serving metrics server","bindAddress":":8080","secure":false}
{"level":"info","ts":"2025-10-30T01:43:04Z","msg":"starting server","name":"health probe","addr":"127.0.0.1:9443"}
{"level":"info","ts":"2025-10-30T01:43:04Z","msg":"attempting to acquire leader lease fdb-operator/fdb-kubernetes-operator..."}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"successfully acquired lease fdb-operator/fdb-kubernetes-operator"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","source":"kind source: *v1.Service"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbrestore","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBRestore","source":"kind source: *v1beta2.FoundationDBRestore"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","source":"kind source: *v1beta2.FoundationDBCluster"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","source":"kind source: *v1.Pod"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbbackup","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBBackup","source":"kind source: *v1.Deployment"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","source":"kind source: *v1.PersistentVolumeClaim"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","source":"kind source: *v1.ConfigMap"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting EventSource","controller":"foundationdbbackup","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBBackup","source":"kind source: *v1beta2.FoundationDBBackup"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting Controller","controller":"foundationdbrestore","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBRestore"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting workers","controller":"foundationdbrestore","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBRestore","worker count":1}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting Controller","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting workers","controller":"foundationdbcluster","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBCluster","worker count":1}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting Controller","controller":"foundationdbbackup","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBBackup"}
{"level":"info","ts":"2025-10-30T01:43:22Z","msg":"Starting workers","controller":"foundationdbbackup","controllerGroup":"apps.foundationdb.org","controllerKind":"FoundationDBBackup","worker count":1}

Documentation

I don't believe we need to add/update documentation for this one.

[johscheuer]

We had a discussion around the default in the PR that added this flag: #2340 (comment). We decided to keep the current default value to reduce the risk of breaking things for customers. I would propose to change the default in the helm chart and creating a GitHub issue to change the default before cutting the 3.0 release of the operator (probably next year).