Index Management Policy is not executed - why?

[jst]

Versions:
OpenSearch(-Dashboards): 2.12.0, Debian: 12

Describe the issue:
I created an Index State Management Policy, which should delete all indices, that are older than 30 days.
All the jobs begin normally, but they never end - they keep the status “Running” / “Evaluating transition conditions” (see screenshot below).

Any ideas? Thank you very much!

Configuration:

{
    "policy": {
        "policy_id": "delete_after_30_days",
        "description": "Deletes indices older than 30 days",
        "last_updated_time": 1717575252051,
        "schema_version": 21,
        "error_notification": {
            "channel": {
                "id": "RS-qs4EBQs81ddEaD87U"
            },
            "message_template": {
                "source": "",
                "lang": "mustache"
            }
        },
        "default_state": "init_state",
        "states": [
            {
                "name": "init_state",
                "actions": [],
                "transitions": [
                    {
                        "state_name": "delete_state",
                        "conditions": {
                            "min_index_age": "30d"
                        }
                    }
                ]
            },
            {
                "name": "delete_state",
                "actions": [
                    {
                        "retry": {
                            "count": 3,
                            "backoff": "exponential",
                            "delay": "1m"
                        },
                        "delete": {}
                    }
                ],
                "transitions": []
            }
        ],
        "ism_template": [
            {
                "index_patterns": [
                    "[clients]-winlogbeat-*"
                ],
                "priority": 1,
                "last_updated_time": 1717505488316
            },
            {
                "index_patterns": [
                    "security-auditlog-*"
                ],
                "priority": 1,
                "last_updated_time": 1717510460570
            },
            {
                "index_patterns": [
                    "logstash-*"
                ],
                "priority": 1,
                "last_updated_time": 1717513238407
            },
            {
                "index_patterns": [
                    "winlogbeat-*"
                ],
                "priority": 1,
                "last_updated_time": 1717575252051
            },
            {
                "index_patterns": [
                    "filebeat-*"
                ],
                "priority": 1,
                "last_updated_time": 1717575252051
            }
        ]
    }
}

Relevant Logs or Screenshots:

[gaobinlong]

Is there any index older than 30 days in your cluster? What’s the status of the policy execution on these indexes?

[jst]

Yes, definitely.