Briefly, this error occurs when the location specified in the Elasticsearch configuration does not start with a repository path. This could be due to incorrect configuration or a typo in the path. To resolve this issue, you can check the Elasticsearch configuration file and ensure that the specified location starts with the correct repository path. If the path is correct, check for typos or syntax errors. If the problem persists, you may need to check the permissions of the directory to ensure Elasticsearch has the necessary access.

It’s important to ensure your path repository setting is configured correctly to ensure location identification. we recommend you run the Elasticsearch Error Check-Up to check your configuration and resolve issues such as the above.

This guide will explain how to easily resolve this issue. Beyond that, it’s also important to understand the issues related to the log, so below there is an overview on common issues and tips related to the Elasticsearch concepts: repositories and repository-azure.

Overview

The process of setting up Elasticsearch to take snapshots of your data begins with you configuring Elasticsearch so it knows the available repositories where it will be able to save your data. When it comes to shared file system repositories, this is done in two steps:

1. You first have to list all available data repositories in the configuration file on every node of the cluster
2. You can then register repositories in the cluster state, so they can be used when taking snapshots

The “Doesn’t match any of the locations specified by path.repo because this setting is empty” error usually takes place when you try to execute the second step prior to the first one.

What it means

This error probably means you didn’t configure the path.repo parameter in the config/elasticsearch.yml configuration file of at least one of the nodes of your cluster.

Why it occurs

In order to create a snapshot or to configure a snapshot policy to secure your Elasticsearch cluster, you first have to register a repository to which it will save your data. You do this by sending a PUT request to the _snapshot endpoint, as shown below:

PUT _snapshot/my_fs_backup
{
  "type": "fs",
  "settings": {
    "location": "/mount/backups/my_fs_backup_location"
  }
}

Elasticsearch’s snapshot solution supports several repository implementations, such as AWS S3, Google Cloud Storage, Microsoft Azure among others. You can also use your own data infrastructure and provide a shared filesystem repository (which is the case for the example above). 

In this case, prior to registering the repository in the cluster’s state, you must include a path.repo parameter in the config/elasticsearch.yml configuration file on every node in your cluster. The value of this parameter should contain a path to the shared file system, which should be accessible in every node.

If for some reason you fail to do that, you’ll then get the “Doesn’t match any of the locations specified by path.repo because this setting is empty” error message. In the example above the reference to the /mount/backups/my_fs_backup_location location will cause the error, if not configured in the configuration file of all the nodes. 

This is the error Elasticsearch will return:

{
  "error" : {
    "root_cause" : [
      {
        "type" : "repository_exception",
        "reason" : "[my_fs_backup] location [/mount/backups/my_fs_backup_location] doesn't match any of the locations specified by path.repo because this setting is empty"
      }
    ],
    "type" : "repository_exception",
    "reason" : "[my_fs_backup] failed to create repository",
    "caused_by" : {
      "type" : "repository_exception",
      "reason" : "[my_fs_backup] location [/mount/backups/my_fs_backup_location] doesn't match any of the locations specified by path.repo because this setting is empty"
    }
  },
  "status" : 500
}

How to resolve it

You can simply include the path.repo parameter with a proper value in the config/elasticsearch.yml of all of your nodes and then restart them.

cluster.name: acme
node.name: node01
network.host: [ _local_ ]
cluster.initial_master_nodes: node01
path.repo: ["/mount/backups/my_fs_backup_location"]