Snapshotting not working when using `spec.snapshot.enableOnMasterOnly`

[applike-ss]

I am running the most recent version of the operator (v1.3.0 as of today).

When setting spec.snapshot.enableOnMasterOnly and all pods are rolling due to some other config change, I do see this in the log stating that the snapshot_cron setting will be disabled:

clearing snapshot cron schedule on replica

However I do never see this:

setting snapshot cron schedule on master

Which I should when the new master is proposed (if i'm not mistaken?) as per: link 1 and link 2

I did also not see either of these in my logs:

response of `SLAVE OF NO ONE` on master is not OK
error running SLAVE OF NO ONE command

So it doesn't seem that there was an obvious redis client error.

When asking the currently as master labeled pod for its snapshot_cron config, it also replies with it being empty:

$ redis-cli
127.0.0.1:6379> config get snapshot*
1) "snapshot_cron"
2) ""

Demo object I use:

apiVersion: dragonflydb.io/v1alpha1
kind: Dragonfly
metadata:
  name: dragonfly
  namespace: xyz
spec:
  annotations:
  args:
  - "--cache_mode"
  - "--cluster_mode=emulated"
  - "--version_check=false"
  - "--proactor_threads=1"
  - "--maxmemory=256Mi"
  - "--dbfilename=dragonfly"
  - "--max_eviction_per_heartbeat=100"
  - "--dbnum=16"
  - "--eviction_memory_budget_threshold=0.1"
  - "--max_segment_to_consider=4"
  - "--shard_repl_backlog_len=1"
  image: ghcr.io/dragonflydb/dragonfly:v1.34.1
  imagePullPolicy: Always
  replicas: 3
  resources:
    limits:
      cpu: 1
      memory: 320Mi
    requests:
      cpu: 100m
      ephemeral-storage: 1Gi
      memory: 320Mi
  serviceAccountName: dragonfly
  snapshot:
    cron: "*/5 * * * *"
    dir: s3://bucket/xyz
    enableOnMasterOnly: true
  topologySpreadConstraints:
  - labelSelector:
      matchLabels:
        app: dragonfly
    maxSkew: 1
    topologyKey: kubernetes.io/hostname
    whenUnsatisfiable: DoNotSchedule
  - labelSelector:
      matchLabels:
        app: dragonfly
    maxSkew: 1
    topologyKey: topology.kubernetes.io/zone
    whenUnsatisfiable: ScheduleAnyway

As a result snapshotting doesn't seem to work at all for me anymore.

Is there anything that i'm obviously doing wrong here or is this maybe a bug in the implementation?