Vertica

Supported OS Linux Windows

Integrationv3.4.1

概要

このチェックは、Datadog Agent を通じて Vertica を監視します。

セットアップ

インストール

Vertica チェックは Datadog Agent パッケージに含まれています。サーバーに追加でインストールする必要はありません。

コンフィギュレーション

vertica のパフォーマンスデータの収集を開始するには、Agent のコンフィギュレーションディレクトリのルートにある conf.d/ フォルダーの vertica.d/conf.yaml ファイルを編集します。使用可能なすべてのコンフィギュレーションオプションの詳細については、サンプル vertica.d/conf.yaml を参照してください。

SSL の有効化

Vertica インテグレーションは、SSL を使用した Vertica への接続をサポートします。これを有効にするには、conf.yamluse_tlstrue にします。

注: Vertica インテグレーションのバージョン 1.9.0 以前では、tls_verifytrue にします。レガシーのサポートとして、tls_verify が明示的に true に設定されている場合、use_tlstrue に設定されます。

Vertica の準備

Datadog Agent のデータベースユーザーを作成します。vsql から、スーパーユーザーとしてデータベースに接続します。次に、CREATE USER ステートメントを実行します。

CREATE USER datadog IDENTIFIED BY '<パスワード>';

モニタリングシステムテーブルにアクセスするには、データベースへの接続に使用するユーザーに SYSMONITOR ロールを付与する必要があります。

GRANT SYSMONITOR TO datadog WITH ADMIN OPTION;

現在のライセンス使用のメトリクスは最新の監査の値を使用するため、Datadog は監査をできるだけ頻繁にスケジュールすることをお勧めします。詳細については、Vertica 監査ライセンスガイドを参照してください。

Agent を再起動すると、Datadog への Vertica メトリクスの送信が開始されます。

ログの収集

Agent バージョン 6.0 以降で利用可能

  1. Datadog Agent で、ログの収集はデフォルトで無効になっています。以下のように、datadog.yaml ファイルでこれを有効にします。

    logs_enabled: true
    
  2. Vertica のログの収集を開始するには、次の構成ブロックを vertica.d/conf.yaml ファイルに追加します。

    logs:
      - source: vertica
        type: file
        path: "/<CATALOG_PATH>/<DATABASE_NAME>/<NODE_NAME>_catalog/vertica.log"
        service: vertica
    
  3. Agent を再起動します

検証

Agent の status サブコマンドを実行し、Checks セクションで vertica を探します。

収集データ

メトリクス

vertica.connection.active
(gauge)
The current number of connections to a node.
Shown as connection
vertica.connection.total
(count)
The total number of connections to a node.
Shown as connection
vertica.delete_vectors
(gauge)
The total number of delete vectors.
Shown as item
vertica.disk.used
(gauge)
The total disk space (compressed) used in bytes.
Shown as byte
vertica.disk.used.ros
(gauge)
[Only versions <= 10] The total ROS disk space (compressed) used in bytes.
Shown as byte
vertica.disk.used.wos
(gauge)
[Only versions <= 10] The total WOS disk space (compressed) used in bytes.
Shown as byte
vertica.epoch.ahm
(gauge)
The current epoch number.
Shown as time
vertica.epoch.current
(gauge)
The AHM epoch number.
Shown as time
vertica.epoch.last_good
(gauge)
The smallest (min) of all the checkpoint epochs on the cluster.
Shown as time
vertica.file.max
(gauge)
The maximum number of files that can be open at one time.
Shown as file
vertica.file.open
(gauge)
The number of open files.
Shown as file
vertica.ksafety.current
(gauge)
The number of node failures the cluster can tolerate before it shuts down automatically.
Shown as node
vertica.ksafety.intended
(gauge)
The designed or intended K-safety level.
Shown as node
vertica.license.expiration
(gauge)
The time until license expiration.
Shown as second
vertica.license.latest_audit
(gauge)
The time elapsed since the latest license audit was triggered.
Shown as second
vertica.license.size
(gauge)
The total size of the license in bytes.
Shown as byte
vertica.license.usable
(gauge)
The total usable size of the license in bytes.
Shown as byte
vertica.license.used
(gauge)
The total used size of the license in bytes.
Shown as byte
vertica.license.utilized
(gauge)
The utilization of the license as a percentage.
Shown as percent
vertica.memory.swap.total
(gauge)
The total amount of swap memory in bytes.
Shown as byte
vertica.memory.swap.usable
(gauge)
The total usable amount of swap memory in bytes.
Shown as byte
vertica.memory.swap.used
(gauge)
The total used amount of swap memory in bytes.
Shown as byte
vertica.memory.swap.utilized
(gauge)
The utilization of swap memory as a percentage.
Shown as percent
vertica.memory.total
(gauge)
The total amount of physical RAM in bytes.
Shown as byte
vertica.memory.usable
(gauge)
The total usable amount of physical RAM in bytes.
Shown as byte
vertica.memory.used
(gauge)
The total used amount of physical RAM in bytes.
Shown as byte
vertica.memory.utilized
(gauge)
The utilization of physical RAM as a percentage.
Shown as percent
vertica.node.allowed
(gauge)
The number of nodes permitted for use by the license.
Shown as node
vertica.node.available
(gauge)
The number of available nodes remaining based on the license limit.
Shown as node
vertica.node.delete_vectors
(gauge)
The number of delete vectors in nodes.
Shown as item
vertica.node.disk.used
(gauge)
The total disk space used by nodes in bytes.
Shown as byte
vertica.node.disk.used.ros
(gauge)
[Only versions <= 10] The ROS disk space used by nodes in bytes.
Shown as byte
vertica.node.disk.used.wos
(gauge)
[Only versions <= 10] The WOS disk space used by nodes in bytes.
Shown as byte
vertica.node.down
(gauge)
The number of nodes that are currently down.
Shown as node
vertica.node.resource_requests
(gauge)
The cumulative number of requests for threads, file handles, and memory (in kilobytes).
Shown as resource
vertica.node.ros.containers
(gauge)
The number of ROS containers in nodes.
Shown as resource
vertica.node.row.ros
(gauge)
[Only versions <= 10] The number of ROS rows in nodes.
Shown as row
vertica.node.row.total
(gauge)
The total number of rows in nodes.
Shown as row
vertica.node.row.wos
(gauge)
[Only versions <= 10] The number of WOS rows in nodes.
Shown as row
vertica.node.total
(gauge)
The number of nodes currently in use.
Shown as node
vertica.projection.delete_vectors
(gauge)
The number of delete vectors in projections.
Shown as item
vertica.projection.disk.used
(gauge)
The total disk space used by projections in bytes.
Shown as byte
vertica.projection.disk.used.ros
(gauge)
[Only versions <= 10] The ROS disk space used by projections in bytes.
Shown as byte
vertica.projection.disk.used.wos
(gauge)
[Only versions <= 10] The WOS disk space used by projections in bytes.
Shown as byte
vertica.projection.ros.containers
(gauge)
The number of ROS containers in projections.
Shown as resource
vertica.projection.row.ros
(gauge)
[Only versions <= 10] The number of ROS rows in projections.
Shown as row
vertica.projection.row.total
(gauge)
The total number of rows in projections.
Shown as row
vertica.projection.row.wos
(gauge)
[Only versions <= 10] The number of WOS rows in projections.
Shown as row
vertica.projection.total
(gauge)
The total number of projections.
Shown as table
vertica.projection.unsafe
(gauge)
The number of projections that are not up to date.
Shown as table
vertica.projection.unsafe_percent
(gauge)
The percentage of projections that are not up to date.
Shown as percent
vertica.projection.unsegmented
(gauge)
The number of projections that are not segmented.
Shown as table
vertica.projection.unsegmented_percent
(gauge)
The percentage of projections that are not segmented.
Shown as percent
vertica.processor.core
(gauge)
The number of processor cores in the system.
Shown as core
vertica.processor.cpu
(gauge)
The number of system processors.
Shown as cpu
vertica.query.active
(gauge)
The current number of queries running on a node.
Shown as query
vertica.query.total
(count)
The total number of queries executed on a node.
Shown as query
vertica.resource_pool.memory.borrowed
(gauge)
The amount of memory borrowed from the GENERAL pool by requests running against pools.
Shown as byte
vertica.resource_pool.memory.max
(gauge)
The value of the MAXMEMORYSIZE parameter as defined for pools.
Shown as byte
vertica.resource_pool.memory.used
(gauge)
The amount of memory acquired by requests running against pools.
Shown as byte
vertica.resource_pool.query.running
(gauge)
The number of queries currently executing in pools.
Shown as query
vertica.ros.containers
(gauge)
The total number of ROS containers.
Shown as resource
vertica.row.ros
(gauge)
[Only versions <= 10] The total number of ROS rows.
Shown as row
vertica.row.total
(gauge)
The total number of rows.
Shown as row
vertica.row.wos
(gauge)
[Only versions <= 10] The total number of WOS rows.
Shown as row
vertica.socket.open
(gauge)
The number of open sockets.
Shown as file
vertica.storage.latency
(gauge)
The latency of a storage location in seeks per second.
Shown as read
vertica.storage.size
(gauge)
The total size of a storage location in bytes.
Shown as byte
vertica.storage.speed
(gauge)
The time (in seconds) it takes to read 1 mebibyte from a storage location.
Shown as second
vertica.storage.throughput
(gauge)
The throughput of a storage location in mebibytes per second.
Shown as mebibyte
vertica.storage.usable
(gauge)
The total usable size of a storage location in bytes.
Shown as byte
vertica.storage.used
(gauge)
The total used size of a storage location in bytes.
Shown as byte
vertica.storage.utilized
(gauge)
The utilization of a storage location as a percentage.
Shown as percent
vertica.table.disk.used
(gauge)
The total disk space used by tables in bytes.
Shown as byte
vertica.table.disk.used.ros
(gauge)
[Only versions <= 10] The ROS disk space used by tables in bytes.
Shown as byte
vertica.table.disk.used.wos
(gauge)
[Only versions <= 10] The WOS disk space used by tables in bytes.
Shown as byte
vertica.table.ros.containers
(gauge)
The number of ROS containers in tables.
Shown as resource
vertica.table.row.ros
(gauge)
[Only versions <= 10] The number of ROS rows in tables.
Shown as row
vertica.table.row.total
(gauge)
The total number of rows in tables.
Shown as row
vertica.table.row.wos
(gauge)
[Only versions <= 10] The number of WOS rows in tables.
Shown as row
vertica.thread.active
(gauge)
The number of active threads.
Shown as thread
vertica.thread.max
(gauge)
The maximum number of threads that can coexist.
Shown as thread

イベント

Vertica には、イベントは含まれません。

サービスのチェック

vertica.can_connect
Returns OK if the Agent is able to connect to the monitored Vertica database, or CRITICAL otherwise.
Statuses: ok, critical

vertica.node_state
Returns OK for each node that is UP, WARNING for nodes that are on a possible path to UP, or CRITICAL otherwise.
Statuses: ok, warning, critical, unknown

トラブルシューティング

ご不明な点は、Datadog のサポートチームまでお問合せください。