本页面介绍了使用 Linux 部署选项的 AlloyDB Omni 当前版本 (17.5.0)。选择其他部署选项。

Google 会使用 AI 技术将内容翻译成您偏好的语言。AI 翻译可能包含错误。

监控 AlloyDB Omni 数据库性能

选择文档版本：

本页面介绍了如何使用 PostgreSQL 可观测性脚本监控 AlloyDB Omni 数据库性能。

查看已连接进程和等待事件的状态

您可以通过查询 pg_stat_activity 视图来确定与 AlloyDB Omni 实例连接的任何进程以及在等待活动的任何后端。

SELECT
    pid,
    datname,
    age(backend_xid) AS age_in_xids,
    now() - xact_start AS xact_age,
    now() - query_start AS query_age,
    state,
    wait_event_type,
    wait_event,
    query_id,
    query
FROM
    pg_stat_activity
WHERE
    state != 'idle'
    AND pid <> pg_backend_pid()
ORDER BY
    4 DESC
LIMIT 10;

查看最大表

您可以通过查询 pg_stat_user_tables 视图来确定最大表的大小。

SELECT
    oid,
    oid::regclass table_name,
    pg_size_pretty(pg_relation_size(oid)),
    relpages,
    s.seq_scan,
    s.idx_scan
FROM
    pg_class,
    pg_stat_user_tables s
WHERE
    s.relid = oid
    AND oid > 16383
    AND relpages > 100
    AND relkind = 'r'
ORDER BY
    relpages DESC
LIMIT 20;

查看排名靠前的顺序扫描

您可以通过查询 pg_stat_user_tables 视图来查看排名靠前的顺序扫描。

SELECT
    relid,
    relname,
    seq_scan,
    pg_size_pretty(pg_relation_size(relid))
FROM
    pg_stat_user_tables
ORDER BY
    seq_scan DESC
LIMIT 15;

查看排名靠前的索引扫描

您可以通过查询 pg_stat_user_tables 视图来查看排名靠前的索引扫描。

SELECT
    relid,
    relid::regclass table_name,
    idx_scan,
    pg_size_pretty(pg_relation_size(relid))
FROM
    pg_stat_user_tables
WHERE
    idx_scan > 10
ORDER BY
    idx_scan DESC
LIMIT 15;

查看运行时间最长的事务

您可以通过查询 pg_stat_activity 视图并检查事务的存在时间来查看运行时间最长的事务。

SELECT
    pid,
    age(backend_xid) AS age_in_xids,
    now() - xact_start AS xact_age,
    now() - query_start AS query_age,
    state,
    query
FROM
    pg_stat_activity
WHERE
    state != 'idle'
ORDER BY
    2 DESC
LIMIT 10;

查看完全清理 (vacuum) 进度

您可以通过查询 pg_stat_progress_vacuum 视图并使用进程 ID 将其与 pg_stat_activity 视图联接来检查完全清理 (vacuum) 操作的进度。

SELECT
  p.pid,
  now() - a.xact_start AS duration,
  coalesce(wait_event_type ||'.'|| wait_event, 'f') AS waiting,
  CASE
    WHEN a.query ~*'^autovacuum.*to prevent wraparound' THEN 'wraparound'
    WHEN a.query ~*'^vacuum' THEN 'user'
  ELSE
    'regular'
  END AS mode,
  p.datname AS database,
  p.relid::regclass AS table,
  p.phase,
  pg_size_pretty(p.heap_blks_total * current_setting('block_size')::int) AS table_size,
  pg_size_pretty(pg_total_relation_size(relid)) AS total_size,
  pg_size_pretty(p.heap_blks_scanned * current_setting('block_size')::int) AS scanned,
  pg_size_pretty(p.heap_blks_vacuumed * current_setting('block_size')::int) AS vacuumed,
  round(100.0 * p.heap_blks_scanned / p.heap_blks_total, 1) AS scanned_pct,
  round(100.0 * p.heap_blks_vacuumed / p.heap_blks_total, 1) AS vacuumed_pct,
  p.index_vacuum_count,
  round(100.0 * p.dead_tuple_bytes / p.max_dead_tuple_bytes,1) AS dead_pct
FROM pg_stat_progress_vacuum p
JOIN pg_stat_activity a using (pid)
ORDER BY now() - a.xact_start DESC;

查看异步查询

如需查看异步运行的查询，您可以查询 pg_stat_activity 视图，并过滤不是主要进程的查询。

SELECT
    query,
    leader_pid,
    array_agg(pid) FILTER (WHERE leader_pid != pid) AS members
FROM
    pg_stat_activity
WHERE
    leader_pid IS NOT NULL
GROUP BY
    query,
    leader_pid;

查看屏蔽锁定 SQL

您可以通过查询 pg_locks 视图并将其与 pg_stat_activity 视图联接来查看被屏蔽的活动。

SELECT blocked_locks.pid     AS blocked_pid,
       blocked_activity.usename  AS blocked_user,
       blocking_locks.pid     AS blocking_pid,
       blocking_activity.usename AS blocking_user,
       blocked_activity.query    AS blocked_statement,
       blocked_activity.wait_event AS blocked_wait_event,
       blocking_activity.wait_event AS blocking_wait_event,
       blocking_activity.query   AS current_statement_in_blocking_process
 FROM  pg_catalog.pg_locks         blocked_locks
  JOIN pg_catalog.pg_stat_activity blocked_activity  ON blocked_activity.pid = blocked_locks.pid
  JOIN pg_catalog.pg_locks         blocking_locks
      ON blocking_locks.locktype = blocked_locks.locktype
      AND blocking_locks.database IS NOT DISTINCT FROM blocked_locks.database
      AND blocking_locks.relation IS NOT DISTINCT FROM blocked_locks.relation
      AND blocking_locks.page IS NOT DISTINCT FROM blocked_locks.page
      AND blocking_locks.tuple IS NOT DISTINCT FROM blocked_locks.tuple
      AND blocking_locks.virtualxid IS NOT DISTINCT FROM blocked_locks.virtualxid
      AND blocking_locks.transactionid IS NOT DISTINCT FROM blocked_locks.transactionid
      AND blocking_locks.classid IS NOT DISTINCT FROM blocked_locks.classid
      AND blocking_locks.objid IS NOT DISTINCT FROM blocked_locks.objid
      AND blocking_locks.objsubid IS NOT DISTINCT FROM blocked_locks.objsubid
      AND blocking_locks.pid != blocked_locks.pid
  JOIN pg_catalog.pg_stat_activity blocking_activity ON blocking_activity.pid = blocking_locks.pid
 WHERE NOT blocked_locks.granted;

确定 `work_mem` 和 `temp_buffers` 大小有效性

如需确定 work_mem 和 temp_buffers 是否正确设置了大小以符合您的需求，您可以查询 pg_stat_database 视图并检查 postgres.log 文件。使用 pg_stat_database 执行以下查询，如果在各次执行之间 temp_files 或 temp_bytes 发生了任何增长，则可能需要对 work_mem 或 temp_buffers 进行调优。

SELECT
    datname,
    temp_files,
    temp_bytes
FROM
    pg_stat_database;

运行此命令后，检查 postgres.log 文件，查看是否使用了临时文件：

LOG: [fd.c:1772] temporary file: path "base/pgsql_tmp/pgsql_tmp4640.1", size 139264

目标是尽可能减少临时文件的创建，而不是完全阻止创建临时文件。这是因为同时设置 work_mem 和 temp_buffers 是为了在主机上的可用内存与需要内存的连接数量之间取得平衡。正确设置这些参数需要了解每个工作负载。

使用 Query Insights 查看查询性能

您可以在 AlloyDB Omni Web 界面中使用 Query Insights 来识别慢查询和高磁盘 I/O 语句。Query Insights 需要在数据库中启用 pg_stat_statements 扩展程序。如需查看设置说明，请参阅安装 Web 界面。

如需在 Web 界面中查看查询效果指标，请完成以下步骤：

连接到 Web 界面，然后选择您的集群。
在导航菜单中，点击 Query Insights。
可选：使用数据库列表按特定数据库过滤语句，或选择所有数据库。
查看查询表，分析 SQL 执行统计信息。每行都会显示以下指标，并在行下方显示完整的 SQL 语句：
- 数据库名称：运行语句的逻辑数据库。
- 调用次数：语句执行的总次数。
- 行数：语句检索或影响的总行数。
- 执行时间：平均值：每次调用的平均执行时间。
- 执行时间：总计：所有调用的累计执行时间。
- 共享块命中次数：共享缓冲区缓存命中总数。
- 共享块读取次数：从磁盘读取的共享块总数。
- 查询类型：对查询性能影响进行分类，例如 Slow query 或 High disk IO。