Debugging rollover failure on AWS cluster

Plugins Index Management
1

I’m experimenting with rollover on an AWS Elasticsearch cluster. I’m trying to set up a basic policy following the ISM documentation, but the rollover always fails:

{
    "cause": "[1d660923bd2c124ece7cbd0beec04560][10.212.17.7:9300][indices:admin/rollover]",
    "message": "Failed to rollover index [index=index_1]"
}

This message is not very verbose, and I don’t know how to get any more information about what’s going wrong. Also, it takes very long time to set a policy up, and for the rollover to be triggered, so it’s hard to test different configurations.

The index looks like this:

{
    "index_1": {
        "aliases": {
            "some_alias": {
                "is_write_index": true
            }
        },
        "mappings": {
           ...
        },
        "settings": {
            "index": {
                "opendistro": {
                    "index_state_management": {
                        "policy_id": "policy_1",
                        "rollover_alias": "some_alias"
                    }
                },
                "number_of_shards": "5",
                "provided_name": "index_1",
                "creation_date": "1602750088061",
                "number_of_replicas": "1",
                "uuid": "2rlLeXZgS2qd8v90_jD87w",
                "version": {
                    "created": "7070099"
                }
            }
        }
    }
}

The policy looks like this:

{
    "_id": "policy_1",
    "_version": 1,
    "_seq_no": 8,
    "_primary_term": 1,
    "policy": {
        "policy_id": "policy_1",
        "description": "ingesting logs",
        "last_updated_time": 1602754970147,
        "schema_version": 1,
        "error_notification": null,
        "default_state": "ingest",
        "states": [
            {
                "name": "ingest",
                "actions": [
                    {
                        "rollover": {
                            "min_doc_count": 5
                        }
                    }
                ],
                "transitions": [
                    {
                        "state_name": "search"
                    }
                ]
            },
            {
                "name": "search",
                "actions": [],
                "transitions": [
                    {
                        "state_name": "delete",
                        "conditions": {
                            "min_index_age": "5m"
                        }
                    }
                ]
            },
            {
                "name": "delete",
                "actions": [
                    {
                        "delete": {}
                    }
                ],
                "transitions": []
            }
        ]
    }
}

I would appreciate anything pointing me in the right direction to make rollover work.

2

Hey @bencem,

Could you cut a support case on the managed service? We’ll be able to help you out on there as we’d need to look at the logs.

Thanks

3

Hi @dbbaughe / @bencem,

Would you be able to share how you resolved the issue if you did?

I am experiencing a similar issue and was curious if you had resolved it.

Thanks
-Chris

4

I had to remove aliases from the index template. When I did that, the policy started working.