Files
cephrgw-operator/cmd/operator/main.go
T
unkinben 466514063a
ci/woodpecker/pr/build Pipeline was successful
ci/woodpecker/pr/pre-commit Pipeline was successful
ci/woodpecker/pr/test Pipeline was successful
Talk to radosgw directly via go-ceph + aws-sdk-go-v2
The operator drove the Ceph manager dashboard REST API to manage RGW users,
buckets and policies. That coupled it to a dashboard login, the dashboard's RGW
wiring, and the dashboard's bucket API surface. Rebuild the Ceph integration to
talk directly to radosgw the way the CLI does, using native Go libraries, while
keeping every operator capability identical.

The exported surface of internal/ceph is unchanged, so the three controllers
and cmd/operator's structure are untouched (bar the CEPH_RGW_* config plumbing).

- replace the internal/ceph client internals with github.com/ceph/go-ceph
  rgw/admin (Admin Ops API) for users, keys, quotas and bucket info/removal
- add github.com/aws/aws-sdk-go-v2 S3 client for bucket create, versioning,
  policy, tagging and object lock, signed as the bucket owner
- map go-ceph admin.ErrNoSuch*/ErrUserExists and smithy APIError codes into
  IsNotFound/IsConflict so controller create-vs-update branching is preserved
- set S3 path-style addressing and WhenRequired checksum modes for RGW
- delete the hand-rolled dashboard client, token auth and JSON plumbing
- keep policy.go/BuildBucketPolicy/BuildTagJSON as pure builders
- replace the client tests with NewClient validation and error-classifier tests
- keep CGO_ENABLED=0 distroless: only go-ceph's pure-Go rgw/admin is imported
- switch env/config to CEPH_RGW_* (endpoint, admin endpoint, access/secret key,
  region, CA, insecure) and update the deployment manifest
- rewrite README and docs/ceph-setup.md for the single RGW admin user
  (caps users=*;buckets=*), keeping Vault/VSO as the primary credential source

Claude-Session: https://claude.ai/code/session_016CEncETbf8cvy1PhsHfFHM
2026-07-24 22:36:10 +10:00

123 lines
3.8 KiB
Go

package main
import (
"context"
"flag"
"os"
"time"
"k8s.io/apimachinery/pkg/runtime"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
clientgoscheme "k8s.io/client-go/kubernetes/scheme"
ctrl "sigs.k8s.io/controller-runtime"
"sigs.k8s.io/controller-runtime/pkg/healthz"
"sigs.k8s.io/controller-runtime/pkg/log/zap"
metricsserver "sigs.k8s.io/controller-runtime/pkg/metrics/server"
cephv1alpha1 "git.unkin.net/unkin/cephrgw-operator/api/v1alpha1"
"git.unkin.net/unkin/cephrgw-operator/internal/ceph"
"git.unkin.net/unkin/cephrgw-operator/internal/controller"
)
var scheme = runtime.NewScheme()
func init() {
utilruntime.Must(clientgoscheme.AddToScheme(scheme))
utilruntime.Must(cephv1alpha1.AddToScheme(scheme))
}
func main() {
var metricsAddr, probeAddr string
var leaderElect bool
flag.StringVar(&metricsAddr, "metrics-bind-address", ":8080", "metrics endpoint address")
flag.StringVar(&probeAddr, "health-probe-bind-address", ":8081", "health probe address")
flag.BoolVar(&leaderElect, "leader-elect", false, "enable leader election")
flag.Parse()
ctrl.SetLogger(zap.New(zap.UseDevMode(false)))
logger := ctrl.Log.WithName("setup")
cephCfg, endpoint, err := cephConfigFromEnv()
if err != nil {
logger.Error(err, "invalid radosgw configuration")
os.Exit(1)
}
cephClient, err := ceph.NewClient(cephCfg)
if err != nil {
logger.Error(err, "unable to build radosgw client")
os.Exit(1)
}
// Fail fast on obviously-broken credentials, but do not block startup on a
// transiently unreachable radosgw.
pingCtx, cancel := context.WithTimeout(context.Background(), 15*time.Second)
if err := cephClient.Ping(pingCtx); err != nil {
logger.Error(err, "initial radosgw authentication failed; continuing and will retry per-reconcile")
}
cancel()
mgr, err := ctrl.NewManager(ctrl.GetConfigOrDie(), ctrl.Options{
Scheme: scheme,
Metrics: metricsserver.Options{BindAddress: metricsAddr},
HealthProbeBindAddress: probeAddr,
LeaderElection: leaderElect,
LeaderElectionID: "cephrgw-operator",
})
if err != nil {
logger.Error(err, "unable to create manager")
os.Exit(1)
}
if err := controller.SetupAll(mgr, cephClient, endpoint); err != nil {
logger.Error(err, "unable to set up controllers")
os.Exit(1)
}
if err := mgr.AddHealthzCheck("healthz", healthz.Ping); err != nil {
logger.Error(err, "unable to set up health check")
os.Exit(1)
}
if err := mgr.AddReadyzCheck("readyz", healthz.Ping); err != nil {
logger.Error(err, "unable to set up ready check")
os.Exit(1)
}
logger.Info("starting cephrgw-operator")
if err := mgr.Start(ctrl.SetupSignalHandler()); err != nil {
logger.Error(err, "manager exited with error")
os.Exit(1)
}
}
// cephConfigFromEnv reads radosgw connection settings from the environment,
// which the deployment sources from the cephrgw-credentials Secret.
//
// The operator talks to the radosgw Admin Ops and S3 APIs at
// CEPH_RGW_ADMIN_ENDPOINT (falling back to CEPH_RGW_ENDPOINT). The returned
// endpoint string is the S3 endpoint written into consumer credential Secrets,
// which may differ (e.g. a public S3 name) from the API endpoint.
func cephConfigFromEnv() (ceph.Config, string, error) {
consumerEndpoint := os.Getenv("CEPH_RGW_ENDPOINT")
apiEndpoint := os.Getenv("CEPH_RGW_ADMIN_ENDPOINT")
if apiEndpoint == "" {
apiEndpoint = consumerEndpoint
}
cfg := ceph.Config{
Endpoint: apiEndpoint,
AccessKey: os.Getenv("CEPH_RGW_ACCESS_KEY"),
SecretKey: os.Getenv("CEPH_RGW_SECRET_KEY"),
Region: os.Getenv("CEPH_RGW_REGION"),
Insecure: os.Getenv("CEPH_RGW_INSECURE") == "true",
}
if f := os.Getenv("CEPH_RGW_CA_FILE"); f != "" {
b, err := os.ReadFile(f)
if err != nil {
return cfg, "", err
}
cfg.CACert = b
} else if inline := os.Getenv("CEPH_RGW_CA"); inline != "" {
cfg.CACert = []byte(inline)
}
return cfg, consumerEndpoint, nil
}