101
PostgreSQL 鎖等待監控 珍藏級SQL - 誰堵塞了誰
標簽
PostgreSQL , pg_locks , pg_stat_activity , 鎖監控 , 誰堵塞了誰
背景
在數據庫中，通過鎖以及多版本並發控製可以保護數據的一致性，例如A正在查詢數據，B就無法對A訪問的對象執行DDL。A正在更新某條記錄，B就不能刪除或更新這條記錄。
鎖是數據庫自動管理的，同時數據庫還提供了AD LOCK或者LOCK語法，允許用戶自己控製鎖。
例如AD lock的應用可以參考如下：
《PostgreSQL 使用advisory lock實現行級讀寫堵塞》
《PostgreSQL 無縫自增ID的實現 - by advisory lock》
《PostgreSQL 使用advisory lock或skip locked消除行鎖衝突, 提高幾十倍並發更新效率》
當然，如果應用程序邏輯設計不慎，就可能導致嚴重的鎖等待，或者死鎖的產生。
如果你發現SQL請求大多數時候處於等待鎖的狀態，那麼可能出現了業務邏輯的問題。
如何檢查或監控鎖等待呢？
PostgreSQL提供了兩個視圖
1. pg_locks展示鎖信息，每一個被鎖或者等待鎖的對象一條記錄。
2. pg_stat_activity，每個會話一條記錄，顯示會話狀態信息。
我們通過這兩個視圖可以查看鎖，鎖等待情況。同時可以了解發生鎖衝突的情況。
pg_stat_activity.query反映的是當前正在執行或請求的SQL，而同一個事務中以前已經執行的SQL不能在pg_stat_activity中顯示出來。
所以如果你發現兩個會話發生了衝突，但是他們的pg_stat_activity.query沒有衝突的話，那就有可能是他們之間的某個事務之前的SQL獲取的鎖與另一個事務當前請求的QUERY發生了鎖衝突。
如果追蹤詳細的鎖衝突信息：
1. 可以通過lock trace跟蹤鎖等待的詳細信息，
《PostgreSQL Developer Options (debug, trace, system table mod and so on...) 詳解》
2. 通過數據庫日誌（開啟lock_timeout, log_lockwait參數）(csvlog)跟蹤鎖等待信息，
3. 或者通過數據庫日誌（開啟log_statements='all'，SQL審計）追蹤事務中所有的SQL (csvlog)，分析事務之間的鎖衝突。
4. 通過SQL查看持鎖，等鎖的事務狀態。
鎖的釋放時機：
大多數鎖要等待事務結束後釋放，某些輕量級鎖（數據庫自動控製）是隨用隨釋放的。
查看當前事務鎖等待、持鎖信息的SQL
這條SQL非常有用，建議DBA珍藏。
with
t_wait as
(
select a.mode,a.locktype,a.database,a.relation,a.page,a.tuple,a.classid,a.granted,
a.objid,a.objsubid,a.pid,a.virtualtransaction,a.virtualxid,a.transactionid,a.fastpath,
b.state,b.query,b.xact_start,b.query_start,b.usename,b.datname,b.client_addr,b.client_port,b.application_name
from pg_locks a,pg_stat_activity b where a.pid=b.pid and not a.granted
),
t_run as
(
select a.mode,a.locktype,a.database,a.relation,a.page,a.tuple,a.classid,a.granted,
a.objid,a.objsubid,a.pid,a.virtualtransaction,a.virtualxid,a.transactionid,a.fastpath,
b.state,b.query,b.xact_start,b.query_start,b.usename,b.datname,b.client_addr,b.client_port,b.application_name
from pg_locks a,pg_stat_activity b where a.pid=b.pid and a.granted
),
t_overlap as
(
select r.* from t_wait w join t_run r on
(
r.locktype is not distinct from w.locktype and
r.database is not distinct from w.database and
r.relation is not distinct from w.relation and
r.page is not distinct from w.page and
r.tuple is not distinct from w.tuple and
r.virtualxid is not distinct from w.virtualxid and
r.transactionid is not distinct from w.transactionid and
r.classid is not distinct from w.classid and
r.objid is not distinct from w.objid and
r.objsubid is not distinct from w.objsubid and
r.pid <> w.pid
)
),
t_unionall as
(
select r.* from t_overlap r
union all
select w.* from t_wait w
)
select locktype,datname,relation::regclass,page,tuple,virtualxid,transactionid::text,classid::regclass,objid,objsubid,
string_agg(
'Pid: '||case when pid is null then 'NULL' else pid::text end||chr(10)||
'Lock_Granted: '||case when granted is null then 'NULL' else granted::text end||' , Mode: '||case when mode is null then 'NULL' else mode::text end||' , FastPath: '||case when fastpath is null then 'NULL' else fastpath::text end||' , VirtualTransaction: '||case when virtualtransaction is null then 'NULL' else virtualtransaction::text end||' , Session_State: '||case when state is null then 'NULL' else state::text end||chr(10)||
'Username: '||case when usename is null then 'NULL' else usename::text end||' , Database: '||case when datname is null then 'NULL' else datname::text end||' , Client_Addr: '||case when client_addr is null then 'NULL' else client_addr::text end||' , Client_Port: '||case when client_port is null then 'NULL' else client_port::text end||' , Application_Name: '||case when application_name is null then 'NULL' else application_name::text end||chr(10)||
'Xact_Start: '||case when xact_start is null then 'NULL' else xact_start::text end||' , Query_Start: '||case when query_start is null then 'NULL' else query_start::text end||' , Xact_Elapse: '||case when (now()-xact_start) is null then 'NULL' else (now()-xact_start)::text end||' , Query_Elapse: '||case when (now()-query_start) is null then 'NULL' else (now()-query_start)::text end||chr(10)||
'Query: '||case when query is null then 'NULL' else query::text end,
chr(10)||'--------'||chr(10)
order by
( case mode
when 'INVALID' then 0
when 'AccessShareLock' then 1
when 'RowShareLock' then 2
when 'RowExclusiveLock' then 3
when 'ShareUpdateExclusiveLock' then 4
when 'ShareLock' then 5
when 'ShareRowExclusiveLock' then 6
when 'ExclusiveLock' then 7
when 'AccessExclusiveLock' then 8
else 0
end ) desc,
(case when granted then 0 else 1 end)
) as lock_conflict
from t_unionall
group by
locktype,datname,relation,page,tuple,virtualxid,transactionid::text,classid,objid,objsubid ;
如果覺得寫SQL麻煩，可以將它創建為視圖
create view v_locks_monitor as
with
t_wait as
(
select a.mode,a.locktype,a.database,a.relation,a.page,a.tuple,a.classid,a.granted,
a.objid,a.objsubid,a.pid,a.virtualtransaction,a.virtualxid,a.transactionid,a.fastpath,
b.state,b.query,b.xact_start,b.query_start,b.usename,b.datname,b.client_addr,b.client_port,b.application_name
from pg_locks a,pg_stat_activity b where a.pid=b.pid and not a.granted
),
t_run as
(
select a.mode,a.locktype,a.database,a.relation,a.page,a.tuple,a.classid,a.granted,
a.objid,a.objsubid,a.pid,a.virtualtransaction,a.virtualxid,a.transactionid,a.fastpath,
b.state,b.query,b.xact_start,b.query_start,b.usename,b.datname,b.client_addr,b.client_port,b.application_name
from pg_locks a,pg_stat_activity b where a.pid=b.pid and a.granted
),
t_overlap as
(
select r.* from t_wait w join t_run r on
(
r.locktype is not distinct from w.locktype and
r.database is not distinct from w.database and
r.relation is not distinct from w.relation and
r.page is not distinct from w.page and
r.tuple is not distinct from w.tuple and
r.virtualxid is not distinct from w.virtualxid and
r.transactionid is not distinct from w.transactionid and
r.classid is not distinct from w.classid and
r.objid is not distinct from w.objid and
r.objsubid is not distinct from w.objsubid and
r.pid <> w.pid
)
),
t_unionall as
(
select r.* from t_overlap r
union all
select w.* from t_wait w
)
select locktype,datname,relation::regclass,page,tuple,virtualxid,transactionid::text,classid::regclass,objid,objsubid,
string_agg(
'Pid: '||case when pid is null then 'NULL' else pid::text end||chr(10)||
'Lock_Granted: '||case when granted is null then 'NULL' else granted::text end||' , Mode: '||case when mode is null then 'NULL' else mode::text end||' , FastPath: '||case when fastpath is null then 'NULL' else fastpath::text end||' , VirtualTransaction: '||case when virtualtransaction is null then 'NULL' else virtualtransaction::text end||' , Session_State: '||case when state is null then 'NULL' else state::text end||chr(10)||
'Username: '||case when usename is null then 'NULL' else usename::text end||' , Database: '||case when datname is null then 'NULL' else datname::text end||' , Client_Addr: '||case when client_addr is null then 'NULL' else client_addr::text end||' , Client_Port: '||case when client_port is null then 'NULL' else client_port::text end||' , Application_Name: '||case when application_name is null then 'NULL' else application_name::text end||chr(10)||
'Xact_Start: '||case when xact_start is null then 'NULL' else xact_start::text end||' , Query_Start: '||case when query_start is null then 'NULL' else query_start::text end||' , Xact_Elapse: '||case when (now()-xact_start) is null then 'NULL' else (now()-xact_start)::text end||' , Query_Elapse: '||case when (now()-query_start) is null then 'NULL' else (now()-query_start)::text end||chr(10)||
'Query: '||case when query is null then 'NULL' else query::text end,
chr(10)||'--------'||chr(10)
order by
( case mode
when 'INVALID' then 0
when 'AccessShareLock' then 1
when 'RowShareLock' then 2
when 'RowExclusiveLock' then 3
when 'ShareUpdateExclusiveLock' then 4
when 'ShareLock' then 5
when 'ShareRowExclusiveLock' then 6
when 'ExclusiveLock' then 7
when 'AccessExclusiveLock' then 8
else 0
end ) desc,
(case when granted then 0 else 1 end)
) as lock_conflict
from t_unionall
group by
locktype,datname,relation,page,tuple,virtualxid,transactionid::text,classid,objid,objsubid ;
例子
postgres=# create table locktest(id int primary key, info text);
CREATE TABLE
postgres=# insert into locktest values (1,'a');
INSERT 0 1
會話A
postgres=# begin;
BEGIN
postgres=# update locktest set info='a' where id=1;
UPDATE 1
postgres=# select * from locktest ;
id | info
----+------
1 | a
(1 row)
會話B
postgres=# begin;
BEGIN
postgres=# select * from locktest ;
id | info
----+------
1 | a
(1 row)
會話C
postgres=# begin;
BEGIN
postgres=# insert into locktest values (2,'test');
INSERT 0 1
會話D
postgres=# begin;
BEGIN
postgres=# truncate locktest ;
waiting......
會話E
postgres=# select * from locktest ;
waiting......
會話F
postgres=# \x
Expanded display is on.
postgres=# select * from v_locks_monitor ;
-[ RECORD 1 ]-+------------------------------------------------------------------------------------------------------------------------------------------------------
locktype | relation
datname | postgres
relation | locktest
page |
tuple |
virtualxid |
transactionid |
classid |
objid |
objsubid |
string_agg | Pid: 23043 +
| Granted: false , Mode: AccessExclusiveLock , FastPath: false , VirtualTransaction: 4/1450064 , Session_State: active +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:43:43.735829+08 , Query_Start: 2017-05-21 21:43:50.965797+08 , Xact_Elapse: 00:01:11.919991 , Query_Elapse: 00:01:04.690023+
| Query: truncate locktest ; +
| -------- +
| Pid: 40698 +
| Granted: true , Mode: RowExclusiveLock , FastPath: false , VirtualTransaction: 6/1031925 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:43:15.173798+08 , Query_Start: 2017-05-21 21:43:24.338804+08 , Xact_Elapse: 00:01:40.482022 , Query_Elapse: 00:01:31.317016+
| Query: insert into locktest values (2,'test'); +
| -------- +
| Pid: 17515 +
| Granted: true , Mode: RowExclusiveLock , FastPath: false , VirtualTransaction: 3/5671759 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:42:19.199124+08 , Query_Start: 2017-05-21 21:42:47.820125+08 , Xact_Elapse: 00:02:36.456696 , Query_Elapse: 00:02:07.835695+
| Query: select * from locktest ; +
| -------- +
| Pid: 17515 +
| Granted: true , Mode: RowExclusiveLock , FastPath: false , VirtualTransaction: 3/5671759 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:42:19.199124+08 , Query_Start: 2017-05-21 21:42:47.820125+08 , Xact_Elapse: 00:02:36.456696 , Query_Elapse: 00:02:07.835695+
| Query: select * from locktest ; +
| -------- +
| Pid: 40698 +
| Granted: true , Mode: RowExclusiveLock , FastPath: false , VirtualTransaction: 6/1031925 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:43:15.173798+08 , Query_Start: 2017-05-21 21:43:24.338804+08 , Xact_Elapse: 00:01:40.482022 , Query_Elapse: 00:01:31.317016+
| Query: insert into locktest values (2,'test'); +
| -------- +
| Pid: 40199 +
| Granted: true , Mode: AccessShareLock , FastPath: false , VirtualTransaction: 5/1029276 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:43:01.745129+08 , Query_Start: 2017-05-21 21:43:05.928125+08 , Xact_Elapse: 00:01:53.910691 , Query_Elapse: 00:01:49.727695+
| Query: select * from locktest ; +
| -------- +
| Pid: 17515 +
| Granted: true , Mode: AccessShareLock , FastPath: false , VirtualTransaction: 3/5671759 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:42:19.199124+08 , Query_Start: 2017-05-21 21:42:47.820125+08 , Xact_Elapse: 00:02:36.456696 , Query_Elapse: 00:02:07.835695+
| Query: select * from locktest ; +
| -------- +
| Pid: 40199 +
| Granted: true , Mode: AccessShareLock , FastPath: false , VirtualTransaction: 5/1029276 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:43:01.745129+08 , Query_Start: 2017-05-21 21:43:05.928125+08 , Xact_Elapse: 00:01:53.910691 , Query_Elapse: 00:01:49.727695+
| Query: select * from locktest ; +
| -------- +
| Pid: 17515 +
| Granted: true , Mode: AccessShareLock , FastPath: false , VirtualTransaction: 3/5671759 , Session_State: idle in transaction +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:42:19.199124+08 , Query_Start: 2017-05-21 21:42:47.820125+08 , Xact_Elapse: 00:02:36.456696 , Query_Elapse: 00:02:07.835695+
| Query: select * from locktest ; +
| -------- +
| Pid: 24781 +
| Granted: false , Mode: AccessShareLock , FastPath: false , VirtualTransaction: 7/1025270 , Session_State: active +
| Username: postgres , Database: postgres , Client_Addr: NULL , Client_Port: -1 , Application_Name: psql +
| Xact_Start: 2017-05-21 21:44:20.725834+08 , Query_Start: 2017-05-21 21:44:20.725834+08 , Xact_Elapse: 00:00:34.929986 , Query_Elapse: 00:00:34.929986+
| Query: select * from locktest ;
處理方法
1. 前麵的鎖查詢SQL，已經清晰的顯示了每一個發生了鎖等待的對象，按鎖的大小排序，要快速解出這種狀態，terminate最大的鎖對應的PID即可。
postgres=# select pg_terminate_backend(23043);
-[ RECORD 1 ]--------+--
pg_terminate_backend | t
會話D
FATAL: terminating connection due to administrator command
server closed the connection unexpectedly
This probably means the server terminated abnormally
before or while processing the request.
The connection to the server was lost. Attempting reset: Succeeded.
幹掉23043後，大家都清淨了
postgres=# select * from v_locks_monitor ;
(0 rows)
參考
https://www.postgresql.org/docs/9.6/static/view-pg-locks.html
https://www.postgresql.org/docs/9.6/static/monitoring-stats.html
https://www.postgresql.org/docs/9.6/static/mvcc.html
《PostgreSQL Developer Options (debug, trace, system table mod and so on...) 詳解》
《PostgreSQL 使用advisory lock實現行級讀寫堵塞》
《PostgreSQL 無縫自增ID的實現 - by advisory lock》
《PostgreSQL 使用advisory lock或skip locked消除行鎖衝突, 提高幾十倍並發更新效率》
《Compare PostgreSQL and Oracle dead lock detect and transaction》
《PostgreSQL lock waiting order》
《PostgreSQL row lock and htup.t_infomask thinking》
最後更新：2017-05-21 23:01:18