一次WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK
alert報錯資訊:
>>> WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK! pid=57
System State dumped to trace file /opt/app/oracle/product/10.2.0/db_1/admin/ XXXXXX/bdump/XXXXXXX2_j001_16704.trc
報錯的是 j001程式,是一個後臺job程式。
當時執行的是 8938號job ,根據scheduler$_job資料字典確認,該 job用來收集資料庫統計資訊。
Job Slave State Object
Slave ID: 1, Job ID: 8938
----------------------------------------
SO: 0xcd78567b0, type: 4, owner: 0xcd4553530, flag: INIT/-/-/0x00
(session) sid: 2969 trans: (nil), creator: 0xcd4553530, flag: (48110041) USR/- BSY/-/-/-/-/-
DID: 0002-0039-00000C4A, short-term DID: 0002-0039-00000C4B
txn branch: (nil)
oct: 3, prv: 0, sql: 0xcde99cc10, psql: 0xcd8a1f068, user: 0/SYS
service name: SYS$USERS
O/S info: user: oracle, term: UNKNOWN, ospid: 16704, machine: XXXXXX2
program: oracle@XXXXXX2 (J001)
application name: DBMS_SCHEDULER, hash value=2478762354
action name: GATHER_STATS_JOB, hash value=930355498
job執行期間:
大量等待16 號rowcache(dc_histogram_defs)
申請3級(行級排它鎖 SX)
反覆超時(3 秒),最後報 “WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK”
[3 samples, 23:34:45 - 23:34:47]
waited for 'row cache lock', seq_num: 8470
p1: 'cache id'= 0x10 (16號)
p2: 'mode'=0x0
p3: 'request'= 0x3
time_waited: 2.930612 sec (sample interval: 2 sec)
[3 samples, 23:34:42 - 23:34:44]
waited for 'row cache lock', seq_num: 8467
p1: 'cache id'= 0x10
p2: 'mode'=0x0
p3: 'request'= 0x3
time_waited: 2.931427 sec (sample interval: 2 sec)
很可能是本節點或另一個節點長時間使用柱狀圖(如硬解析)或執行 DDL(清算期間需要對大量表進行併發truncate)導致。
由於業務高峰發生在晚間清算,因此建議將JOB改到白天或清算前執行。
由於業務高峰發生在晚間清算,因此建議將JOB改到白天或清算前執行。
來自 “ ITPUB部落格 ” ,連結:http://blog.itpub.net/26239116/viewspace-1076276/,如需轉載,請註明出處,否則將追究法律責任。
相關文章
- WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK!AIENQ
- 故障排除:"WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK! "AIENQ
- WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK!的分析AIENQ
- Metlink:Troubleshooting:WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK!AIENQ
- WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK! 與 dc_tablespcesAIENQ
- WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK-[ID 278316.1]AIENQ
- 翻譯metalink關於WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK文章AIENQ
- 通過system state dump分析WAITED TOO LONG FOR A ROW CACHE ENQUEUEAIENQ
- 一次Row Cache Lock問題處理過程
- 等待事件之Row Cache Lock事件
- ORA-00494 enqueue [CF] held for too long故障分析處理ENQ
- 【ASK_ORACLE】Row Cache Enqueue鎖之概念篇OracleENQ
- ORA-00494: enqueue [CF] held for too long (more than 900 seconds)ENQ
- hanganalyze解決row cache lock(ZT)
- [ORACLE 11G]ROW CACHE LOCK 等待Oracle
- 記一次row cache lock引起的效能問題分析處理
- ORA-00494: enqueue [CF] held for too long (more than 900 seconds) -RACENQ
- 轉)用hanganalyze解決row cache lock
- (轉)用hanganalyze解決row cache lock
- 轉貼_用hanganalyze解決row cache lock
- 用hanganalyze解決row cache lock(轉貼)
- oracle enqueue lockOracleENQ
- WAITEVENT: "row cache lock" Reference Note (文件 ID 34609.1)AI
- 由row cache lock等待事件引起的效能問題事件
- Resolving Issues Where 'Row Cache Lock' Waits are OccurringAI
- Rac 環境中分割槽表建立index hang(row cache lock)Index
- How to Match a Row Cache Object Child Latch to its Row CacheObject
- 一次library cache lock 問題分析
- hang了,嚴重的row cache lock 等待事件--就因大sql文字事件SQL
- enable table lock 的enqueue等待ENQ
- Metlink:How to Match a Row Cache Object Child Latch to its Row CacheObject
- sqlserver kill long time lockSQLServer
- 當刪除oracle資料庫user時發生row cache lock 等待事件Oracle資料庫事件
- oracle??邏輯DG同步卡住,session等待row cache lock的處理過程OracleSession
- 11G 修改使用者密碼導致的row cache lock密碼
- row cache objects latch研究Object
- enq: TX - row lock contentionENQ
- 【MySQL】select for update 的Row Lock 與Table LockMySql