host_configs

A YDB cluster consists of multiple nodes, and one or more typical server configurations are usually used for their deployment. To avoid repeating their description for each node, there is a host_configs section in the configuration file that lists the used configurations and assigned IDs.

Syntax

host_configs:
- host_config_id: 1
  drive:
  - path: <path_to_device>
    type: <type>
    disk_scope: <disk_scope>  # optional attribute
  - path: ...
- host_config_id: 2
  ...

The host_config_id attribute specifies a numeric configuration ID. The drive attribute contains a collection of descriptions of connected drives. Each description consists of two mandatory attributes:

• path: Path to the mounted block device, for example, /dev/disk/by-partlabel/ydb_disk_ssd_01
• type: Type of the device's physical media: ssd, nvme, or rot (rotational - HDD)

Additionally, an optional disk_scope attribute can be specified — a label for calculating fail domains for some reduced configurations, see Configuring disk_scope.

Examples

One configuration with ID 1 and one SSD disk accessible via /dev/disk/by-partlabel/ydb_disk_ssd_01:

host_configs:
- host_config_id: 1
  drive:
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_01
    type: SSD

Two configurations with IDs 1 (two SSD disks) and 2 (three SSD disks):

host_configs:
- host_config_id: 1
  drive:
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_01
    type: SSD
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_02
    type: SSD
- host_config_id: 2
  drive:
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_01
    type: SSD
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_02
    type: SSD
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_03
    type: SSD

Configuring disk_scope

disk_scope is an optional string attribute of a disk that defines a finer-grained failure zone within a single node. It is taken into account when calculating fail domains during the selection of disks for placing VDisks of storage groups.

A fail domain is determined by the physical location of a disk (data center, rack, server, physical device), so either no two VDisks of the same group can be placed on the same node (if the fail domain corresponds to a server or a server rack), or several VDisks of the same group can be placed on the same node on different disks (if the fail domain corresponds to an individual physical device). In the first case, it is impossible to create a configuration in block-4-2 mode with fewer than 8 storage nodes, and in the second case the configuration might not stand the failure of a single node. To limit the number of VDisks of the same group placed on a single node, physical devices can be labeled with the disk_scope attribute, and fail domain calculation can be enabled at the disk_scope level. This makes it possible to build some fault-tolerant configurations in installations that have fewer servers than the number of fail domains required by the chosen fault tolerance mode.

Example

A cluster of 4 servers with 4 disks on each server, in the block-4-2 fault tolerance mode:

host_configs:
- host_config_id: 1
  drive:
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_01
    type: SSD
    disk_scope: fail-domain-1
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_02
    type: SSD
    disk_scope: fail-domain-1
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_03
    type: SSD
    disk_scope: fail-domain-2
  - path: /dev/disk/by-partlabel/ydb_disk_ssd_04
    type: SSD
    disk_scope: fail-domain-2

To calculate fail domains at the disk_scope level for the static group and other storage groups, specify the corresponding fail domain type in the yaml config (top-level key):

fail_domain_type: disk_scope

And in the domain configuration:

domains:
- domain_name: <domain name>
  ...
  storage_pool_kinds:
  - kind: <type of physical devices used>
    fail_domain_type: disk_scope
    ...

With this configuration, two VDisks of each storage group will be placed on each server, and the maximum tolerated failure in such a configuration is the failure of a single server.

Kubernetes Features

The YDB Kubernetes operator mounts NBS disks for Storage nodes at the path /dev/kikimr_ssd_00. To use them, the following host_configs configuration must be specified:

host_configs:
- host_config_id: 1
  drive:
  - path: /dev/kikimr_ssd_00
    type: SSD

The example configuration files provided with the YDB Kubernetes operator contain this section, and it does not need to be changed.