yandex_yq_object_storage_binding (Resource)
Written by
Updated at September 11, 2025
Manages Object Storage binding in Yandex Query service. For more information, see the official documentation.
Example usage
//
// Create a new Object Storage binding.
//
resource "yandex_yq_object_storage_binding" "my_os_binding1" {
name = "tf-test-os-binding1"
description = "Binding has been created from Terraform"
connection_id = yandex_yq_object_storage_connection.my_os_connection.id
compression = "gzip"
format = "json_each_row"
path_pattern = "my_logs/"
column {
name = "ts"
type = "Timestamp"
}
column {
name = "message"
type = "Utf8"
}
}
//
// Create a new Object Storage binding with Hive partitioning.
//
resource "yandex_yq_object_storage_binding" "my_os_binding2" {
name = "tf-test-os-binding2"
description = "Binding has been created from Terraform"
connection_id = yandex_yq_object_storage_connection.my_os_connection.id
format = "csv_with_names"
path_pattern = "my_logs/"
format_setting = {
"file_pattern" = "*.csv"
}
column {
name = "year"
type = "Int32"
not_null = true
}
column {
name = "month"
type = "Int32"
not_null = true
}
column {
name = "day"
type = "Int32"
not_null = true
}
partitioned_by = ["year", "month", "day"]
column {
name = "ts"
type = "Timestamp"
}
column {
name = "message"
type = "Utf8"
}
}
//
// Create a new Object Storage binding with extended partitioning.
//
resource "yandex_yq_object_storage_binding" "my_os_binding3" {
name = "tf-test-os-binding3"
description = "Binding has been created from Terraform"
connection_id = yandex_yq_object_storage_connection.my_os_connection.id
compression = "gzip"
format = "json_each_row"
partitioned_by = [
"date",
"severity",
]
path_pattern = "/cold"
projection = {
"projection.date.format" = "/year=%Y/month=%m/day=%d"
"projection.date.interval" = "1"
"projection.date.max" = "NOW"
"projection.date.min" = "2022-12-01"
"projection.date.type" = "date"
"projection.date.unit" = "DAYS"
"projection.enabled" = "true"
"projection.severity.type" = "enum"
"projection.severity.values" = "error,info,fatal"
"storage.location.template" = "/$${date}/$${severity}"
}
column {
name = "timestamp"
not_null = false
type = "String"
}
column {
name = "message"
not_null = false
type = "String"
}
column {
name = "date"
not_null = true
type = "Date"
}
column {
name = "severity"
not_null = true
type = "String"
}
}
Schema
Required
connection_id(String) The connection identifier.format(String) The data format, e.g. csv_with_names, json_as_string, json_each_row, json_list, parquet, raw, tsv_with_names.name(String) The resource name.path_pattern(String) The path pattern within Object Storage's bucket.
Optional
column(Block List) (see below for nested schema)compression(String) The data compression algorithm, e.g. brotli, bzip2, gzip, lz4, xz, zstd.description(String) The resource description.format_setting(Map of String) Special format setting.partitioned_by(List of String) The list of partitioning column names.projection(Map of String) Projection rules.
Read-Only
id(String) The resource identifier.
Nested Schema for column
Required:
name(String) Column name.
Optional:
not_null(Boolean) A column cannot have the NULL data type. Default:false.type(String) Column data type. YQL data types are used.
Import
The resource can be imported by using their resource ID. For getting the resource ID you can use Yandex Cloud Web Console.
# terraform import yandex_yq_object_storage_binding.<resource Name> <resource Id>
terraform import yandex_yq_object_storage_binding.my_os_binding ...