Datastream rollover ISM policy - backing indices inherite the policy and want to roll over themselves

OpenSearch Dashboards
1

Versions (relevant - OpenSearch/Dashboard/Server OS/Browser): 2.19 / AWS / chrome/FF/terraform

Describe the issue: Datastream rolls over creating backing indices. Backing indices inherit the policy and seem to want to rollover themselves. They fail rollover with message that they can not copy alias. There is no alias and the policy has “copy_alias”: false. Policy never transition to the next state and does not delete itself. The main problems are:

  1. policy never transitions
  2. Opensearch is trying to non existing alias which .. it should not. The Datastream nor backing index has no alias

Configuration:

{
"id": "rollover_1D-delete_30d",
"seqNo": 51838,
"primaryTerm": 1,
"policy": {
"policy_id": "rollover_1D-delete_30d",
"description": "",
"last_updated_time": 1768472394381,
"schema_version": 21,
"error_notification": null,
"default_state": "inital",
"states": [
{
"name": "inital",
"actions": [
{
"retry": {
"count": 3,
"backoff": "exponential",
"delay": "1m"
},
"rollover": {
"min_index_age": "1d",
"copy_alias": false
}
}
],
"transitions": [
{
"state_name": "delete",
"conditions": {
"min_index_age": "30d"
}
}
]
},
{
"name": "delete",
"actions": [
{
"retry": {
"count": 3,
"backoff": "exponential",
"delay": "1m"
},
"delete": {}
}
],
"transitions": []
}
],
"ism_template": [
{
"index_patterns": [
"upc-ww-mg-s2sapiaccesslog-*",
"upc-ww-mg-profile-*",
"upc-ww-fb-service-logs-*"
],
"priority": 10,
"last_updated_time": 1768472394381
}
]
}
}

Relevant Logs or Screenshots:

Errors visible in kibana for affected indices.

{
    "cause": "Validation Failed: 1: Must specify at least one alias action;",
    "message": "Successfully rolled over but failed to copy alias from [index=.ds-s2sapiaccesslog-local-000018] to [index=.ds-s2sapiaccesslog-local-000019]"
}

GET /_cat/aliases

.opendistro-ism-managed-index-history-write .opendistro-ism-managed-index-history-2026.01.15-000042 - - - -
.kibana                                     .kibana_1                                               - - - -

Any advice will be appropriated. I am at my ends.
Atm I sometime go into kibana, highlight the problematic and old indices and re-run the policy with state “delete”.