elasticsearch-index-migrate
v0.8.3
Published
This software is a command line tool for version control and mapping changes of Elasticsearch mappings.
Downloads
616
Maintainers
Readme
elasticsearch-index-migrate
This software is a command line tool for version control and mapping changes of Elasticsearch mappings.
Elasticsearch versions 6.x and 7.x (up to 7.7.0) are supported.
You can see a sample of this tool here.
- Breaking change
- Requirements
- Usage
- Setting
- Migration script
- Query parameters
- Where to store migration scripts
- Commands
Breaking change
-i, --indexName Remove the Flags option in version 0.9 or later. Instead of -i, --indexName Flags option, you can pass the value as a command argument.
Requirements
- node.js (>=16.15.1)
- npm (>=6.14.7)
- elasticsearch 6.x and 7.x
This project has been tested with node.js(v16.15.1, v18.4.0), npm(6.14.7) and elasticsearch(6.8.12 and 7.7.1).
Usage
$ npm install -g elasticsearch-index-migrate
$ elasticsearch-index-migrate COMMAND
running command...
$ elasticsearch-index-migrate (-v|--version|version)
elasticsearch-index-migrate/0.8.3 darwin-x64 node-v16.15.1
$ elasticsearch-index-migrate --help [COMMAND]
USAGE
$ elasticsearch-index-migrate COMMAND
...
Setting
You need to configure the connection settings and migration settings for this software.
The settings can be written in JSON format or read from environment variables.
A JSON file is read by passing a path to the argument of the executable command.
If you set it in an environment variable, the system will read it automatically when you execute the command.
If you put a configuration file named config.json
in ~/.config/elasticsearch-index-migrate/
, it will be loaded automatically.
The environment variables take precedence over the settings. Then the -O
option takes precedence.
If you want to use a JSON file, please use the following.
{
"elasticsearch": {
"version": "7",
"connect": {
"host": "http://0.0.0.0:9202",
"sslCa": "",
"cloudId": "",
"username": "",
"password": ""
}
},
"migration": {
"locations": ["/migration"],
"baselineVersion": "v1.0.0"
}
}
- elasticsearch.version : Enter the version of Elasticsearch you are using. If you are using OpenSearch, please write 'opensearch'.
- elasticsearch.connect.host : Enter the Elasticsearch Host you are using.
- elasticsearch.connect.sslCa : Fill in the location of your Elasticsearch SSL certificate.
- elasticsearch.connect.cloudId : Enter your Elasticsearch Cloud ID.
- elasticsearch.connect.username : Enter your Elasticsearch username.
- elasticsearch.connect.password : Enter your Elasticsearch password.
- migration.locations : Enter the full path or relative path to the directory where you want to store your migration script
- migration.baselineVersion : Please fill in the base version of the migration script
To set the environment variables, set the following variables.
- ELASTICSEARCH_MIGRATION_LOCATIONS
- ELASTICSEARCH_MIGRATION_BASELINE_VERSION
- ELASTICSEARCH_VERSION
- ELASTICSEARCH_HOST
- ELASTICSEARCH_SSL
- ELASTICSEARCH_CLOUDID
- ELASTICSEARCH_USERNAME
- ELASTICSEARCH_PASSWORD
Migration script
Migration scripts are written in JSON format. File names should follow the format v{version}__{name}.json.
Currently the following migration types are supported:
- ADD_FIELD
- CREATE_INDEX
- DELETE_INDEX
- ALTER_SETTING
Migration script format
In the case of ADD_FIELD scripts, describe the mapping information that can be executed with the Put mapping API under the migrate_script property.
{
"type": "ADD_FIELD",
"description": "description",
"migrate_script": {
"properties": {
"user_id": {
"type": "keyword"
},
...
}
}
}
In the case of CREATE_INDEX scripts, write a Request body that can be executed with the Create index API under the migrate_script property.
{
"type": "CREATE_INDEX",
"description": "description",
"migrate_script": {
"settings": {
"index": {
"refresh_interval": "1s",
"number_of_shards": 1,
"number_of_replicas": 0
}
},
"mappings": {
"properties": {
"user_id": {
"type": "keyword"
},
...
}
}
}
}
For a DELETE_INDEX script, do the following.
{
"type": "DELETE_INDEX",
"description": "description"
}
In the case of ALTER_SETTING scripts, write a Request body that can be executed with the Update Indices Settings API under the migrate_script property.
{
"type": "ALTER_SETTING",
"description": "description",
"migrate_script": {
"index" : {
"number_of_replicas" : 1
}
}
}
Query parameters
Query parameters can be specified in JSON format in the migration script. You can use the query parameters according to the migration type. For more information, please check out Elasticsearch.
For Example:
{
"type": "CREATE_INDEX",
"description": "description",
"migrate_script": {
"settings": {
"index": {
"refresh_interval": "1s",
"number_of_shards": 1,
"number_of_replicas": 0
}
},
"mappings": {
"properties": {
"user_id": {
"type": "keyword"
},
...
}
}
},
"query_parameters": {
"include_type_name": true,
"timeout": 10
}
}
Where to store migration scripts
Save your migration script to the directory you set up in ELASTICSEARCH_MIGRATION_LOCATIONS or JSON's migration.locations. If the storage location is /elasticsearch/migration, store the script in a directory like the following. This example is for the index names {index_name} and {index_name}_v1, {index_name}_v2. If the index name has a version, it is necessary to separate the directories.For names in the form {index_name}-{version} or {index_name}_{version}, separate the directory from {index_name}/{version}/. In this case, please put the following script in {version}/.
elasticsearch/
┗ migration/
┣ indices/
│ ┣ {index_name}/
│ │ ┣ v1/
│ │ │ ┣ V1.0.0__init_mapping.json
│ │ │ ┗ V1.0.1__add_field.json
│ │ ┣ v2/
│ │ │ ┣ V2.0.0__init_mapping.json
│ │ │ ┗ V2.0.1__add_field.json
│ │ ┗ v3/
│ ┗ {index_name}/
│ ┗ V1.0.0__create_index.json
Commands
elasticsearch-index-migrate baseline [NAME]
elasticsearch-index-migrate clean [NAME]
elasticsearch-index-migrate help [COMMAND]
elasticsearch-index-migrate init
elasticsearch-index-migrate migrate [NAME]
elasticsearch-index-migrate plan [NAME]
elasticsearch-index-migrate recovery [NAME]
elasticsearch-index-migrate baseline [NAME]
Create a baseline in migration_history if you were running Elasticsearch before the tool was implemented.
USAGE
$ elasticsearch-index-migrate baseline [NAME]
ARGUMENTS
NAME migration index name.
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-D, --delimiter=delimiter The separator between index name and index version.
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-d, --description=description Description to be saved to history.
-h, --help show CLI help
-i, --indexName=indexName migration index name.
-n, --[no-]natural-name Set to true if the index name contains _ or -(Ex: my-index).
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
-v, --index-version=index-version index version. (Ex: For my-index_1970.01.01, the version is
1970.01.01. For my-index_v1, the version is v1.)
--[no-]init If the init command has not been executed in advance, the
migration will be performed after initialization has been
processed.
See code: src/commands/baseline.ts
elasticsearch-index-migrate clean [NAME]
Delete all history stored in the migration_history index
USAGE
$ elasticsearch-index-migrate clean [NAME]
ARGUMENTS
NAME migration index name.
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-D, --delimiter=delimiter The separator between index name and index version.
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-h, --help show CLI help
-i, --indexName=indexName migration index name.
-n, --[no-]natural-name Set to true if the index name contains _ or -(Ex: my-index).
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
-t, --target=(history|index|all) [default: history] Selecting what to delete
history : Delete the target index migration history from
migration_history
index : Delete the target index from elasticsearch
all : Delete both migration history and index
-v, --index-version=index-version index version. (Ex: For my-index_1970.01.01, the version is
1970.01.01. For my-index_v1, the version is v1.)
-y, --yes Always answer "yes" to any prompt that appears during processing
--[no-]init If the init command has not been executed in advance, the
migration will be performed after initialization has been
processed.
See code: src/commands/clean.ts
elasticsearch-index-migrate help [COMMAND]
display help for elasticsearch-index-migrate
USAGE
$ elasticsearch-index-migrate help [COMMAND]
ARGUMENTS
COMMAND command to show help for
OPTIONS
--all see all commands in CLI
See code: @oclif/plugin-help
elasticsearch-index-migrate init
Set up a migration environment.
USAGE
$ elasticsearch-index-migrate init
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-h, --help show CLI help
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
See code: src/commands/init.ts
elasticsearch-index-migrate migrate [NAME]
Migrate the index of Elasticsearch to the latest version based on the execution plan.
USAGE
$ elasticsearch-index-migrate migrate [NAME]
ARGUMENTS
NAME migration index name.
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-D, --delimiter=delimiter The separator between index name and index version.
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-h, --help show CLI help
-i, --indexName=indexName migration index name.
-n, --[no-]natural-name Set to true if the index name contains _ or -(Ex: my-index).
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
-v, --index-version=index-version index version. (Ex: For my-index_1970.01.01, the version is
1970.01.01. For my-index_v1, the version is v1.)
--[no-]init If the init command has not been executed in advance, the
migration will be performed after initialization has been
processed.
--showDiff Outputs the difference between before and after the migration at
the end.
See code: src/commands/migrate.ts
elasticsearch-index-migrate plan [NAME]
Outputs the migration execution plan.
USAGE
$ elasticsearch-index-migrate plan [NAME]
ARGUMENTS
NAME migration index name.
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-D, --delimiter=delimiter The separator between index name and index version.
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-h, --help show CLI help
-i, --indexName=indexName migration index name.
-n, --[no-]natural-name Set to true if the index name contains _ or -(Ex: my-index).
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
-v, --index-version=index-version index version. (Ex: For my-index_1970.01.01, the version is
1970.01.01. For my-index_v1, the version is v1.)
--[no-]init If the init command has not been executed in advance, the
migration will be performed after initialization has been
processed.
See code: src/commands/plan.ts
elasticsearch-index-migrate recovery [NAME]
Delete failed migration history.
USAGE
$ elasticsearch-index-migrate recovery [NAME]
ARGUMENTS
NAME migration index name.
OPTIONS
-B, --baseline_version=baseline_version Migrate from the baseline set in the
ELASTICSEARCH_MIGRATION_BASELINE_VERSION environment variable
-C, --elasticsearch_cloudid=elasticsearch_cloudid Connect to Elasticsearch with the value set in the
ELASTICSEARCH_CLOUDID environment variable
-D, --delimiter=delimiter The separator between index name and index version.
-H, --elasticsearch_host=elasticsearch_host Connect to Elasticsearch with the value set in the
ELASTICSEARCH_HOST environment variable
-I, --elasticsearch_ssl_insecure If true, allow insecure server connections when using SSL.
-L, --migration_locations=migration_locations Read the migration file from the directory set in the
$ELASTICSEARCH_MIGRATION_LOCATIONS environment variable
-O, --option_file=option_file Load migration setting file (.json) from file path (Environment
variables take precedence)
-P, --elasticsearch_password=elasticsearch_password Connect to Elasticsearch with the value set in the
ELASTICSEARCH_PASSWORD environment variable
-S, --elasticsearch_ssl=elasticsearch_ssl Connect to Elasticsearch with the value set in the
ELASTICSEARCH_SSL environment variable
-U, --elasticsearch_username=elasticsearch_username Connect to Elasticsearch with the value set in the
ELASTICSEARCH_USERNAME environment variable
-V, --elasticsearch_version=elasticsearch_version Run migration with Elasticsearch version set in
ELASTICSEARCH_VERSION environment variable
-h, --help show CLI help
-i, --indexName=indexName migration index name.
-n, --[no-]natural-name Set to true if the index name contains _ or -(Ex: my-index).
-r, --number-of-replicas=number-of-replicas [default: 2] number of replicas in migration history index.
-s, --number-of-shards=number-of-shards [default: 1] number of shards in migration history index.
-v, --index-version=index-version index version. (Ex: For my-index_1970.01.01, the version is
1970.01.01. For my-index_v1, the version is v1.)
--[no-]init If the init command has not been executed in advance, the
migration will be performed after initialization has been
processed.
See code: src/commands/recovery.ts