日本熟妇hd丰满老熟妇,中文字幕一区二区三区在线不卡 ,亚洲成片在线观看,免费女同在线一区二区

alicloud_hbr_backup_jobs

更新時(shí)間:

This data source provides the Hbr Backup Jobs of the current Alibaba Cloud user.

-> NOTE: Available in v1.138.0+.

Example Usage

Basic Usage

data "alicloud_hbr_ecs_backup_plans" "default" {
  name_regex = "plan-name"
}

data "alicloud_hbr_backup_jobs" "default" {
  source_type = "ECS_FILE"
  filter {
    key      = "VaultId"
    operator = "IN"
    values   = [data.alicloud_hbr_ecs_backup_plans.default.plans.0.vault_id]
  }
  filter {
    key      = "InstanceId"
    operator = "IN"
    values   = [data.alicloud_hbr_ecs_backup_plans.default.plans.0.instance_id]
  }
  filter {
    key      = "CompleteTime"
    operator = "BETWEEN"
    values   = ["2021-08-23T14:17:15CST", "2021-08-24T14:17:15CST"]
  }
}

data "alicloud_hbr_backup_jobs" "example" {
  source_type = "ECS_FILE"
  status      = "COMPLETE"
  filter {
    key      = "VaultId"
    operator = "IN"
    values   = [data.alicloud_hbr_ecs_backup_plans.default.plans.0.vault_id]
  }
  filter {
    key      = "InstanceId"
    operator = "IN"
    values   = [data.alicloud_hbr_ecs_backup_plans.default.plans.0.instance_id]
  }
  filter {
    key      = "CompleteTime"
    operator = "LESS_THAN"
    values   = ["2021-10-20T20:20:20CST"]
  }
}

output "alicloud_hbr_backup_jobs_default_1" {
  value = data.alicloud_hbr_backup_jobs.default.jobs.0.id
}

output "alicloud_hbr_backup_jobs_example_1" {
  value = data.alicloud_hbr_backup_jobs.example.jobs.0.id
}

Argument Reference

The following arguments are supported:

  • ids - (Optional, ForceNew, Computed) A list of Backup Job IDs.
  • source_type - (Required, ForceNew) The type of data source. Valid values: ECS_FILE, NAS, OSS, OTS, UDM_ECS, UDM_ECS_DISK.
  • status - (Optional, ForceNew) The status of backup job. Valid values: COMPLETE, PARTIAL_COMPLETE, FAILED, UNAVAILABLE.
  • sort_direction - (Optional, ForceNew) The sort direction, sort results by ascending or descending order based on the value jobs id. Valid values: ASCEND, DESCEND.
  • output_file - (Optional) File name where to save data source results (after running terraform plan).

Block filter

More complex filters can be expressed using one or more filter sub-blocks, which take the following arguments:

  • key - (Required) The key of the field to filter. Valid values: PlanId, VaultId, InstanceId, Bucket, FileSystemId, CompleteTime.
  • operator - (Required) The operator of the field to filter. Valid values: EQUAL, NOT_EQUAL, GREATER_THAN, GREATER_THAN_OR_EQUAL, LESS_THAN, LESS_THAN_OR_EQUAL, BETWEEN, IN.
  • values - (Required) Set of values that are accepted for the given field.

-> NOTE: Numeric types such as CompleteTime do not support IN operations for the time being.

Argument Reference

The following attributes are exported in addition to the arguments listed above:

  • jobs - A list of Hbr Backup Jobs. Each element contains the following attributes:
    • id - The ID of the backup job.
    • backup_job_id - The ID of the backup job.
    • source_type - The type of data source. Valid Values: ECS_FILE, OSS, NAS, UDM_DISK.
    • back_job_name - The name of backup job.
    • plan_id - The ID of a backup plan.
    • vault_id - The ID of backup vault.
    • actual_bytes - The actual data volume of the backup task (After deduplication) . Unit byte.
    • actual_items - The actual number of items in the backup task. (Currently only file backup is available).
    • backup_type - Backup type. Valid values: COMPLETE(full backup).
    • bucket - The name of target OSS bucket.
    • bytes_done - The amount of backup data (Incremental). Unit byte.
    • bytes_total - The total amount of data sources. Unit byte.
    • create_time - The creation time of backup job. UNIX time seconds.
    • start_time - The scheduled backup start time. UNIX time seconds.
    • complete_time - The completion time of backup job. UNIX time seconds.
    • updated_time - The update time of backup job. UNIX time seconds.
    • file_system_id - The ID of destination file system.
    • nas_create_time - File system creation time. UNIX time in seconds.
    • instance_id - The ID of target ECS instance.
    • items_done - The number of items restore job recovered.
    • items_total - The total number of items restore job recovered.
    • paths - List of backup path. e.g. ["/home", "/var"].
    • prefix - The prefix of Oss bucket files.
    • include - Include path. String of Json list. Up to 255 characters. e.g. "[\"/var\"]"
    • exclude - Exclude path. String of Json list. Up to 255 characters. e.g. "[\"/home/work\"]"
    • status - The status of restore job. Valid values: COMPLETE , PARTIAL_COMPLETE, FAILED.
    • progress - Backup progress. The value is 100%*100.
    • error_message - Error message.
    • cross_account_type - (Available in v1.190.0+) The type of the cross account backup. It is valid only when source_type is ECS_FILE, NAS, OSS or OTS.
    • cross_account_user_id - (Available in v1.190.0+) The original account ID of the cross account backup managed by the current account. It is valid only when source_type is ECS_FILE, NAS, OSS or OTS.
    • cross_account_role_name - (Available in v1.190.0+) The role name created in the original account RAM backup by the cross account managed by the current account. It is valid only when source_type is ECS_FILE, NAS, OSS or OTS.