Know more about RAC statistics and wait event

简介:
以下列出了RAC中的主要统计信息和等待事件:  

1. Statistics:

1.1 V$SYSSTAT, V$SESSTAT (join to V$STATNAME)

  • gc cr blocks served
  • gc cr block build time
  • gc cr block flush time
  • gc cr block send time
  • gc current blocks served
  • gc current block pin time
  • gc current block flush time
  • gc current block send time
  • gc cr blocks received
  • gc cr block receive time
  • gc current blocks received
  • gc current block receive time
  • gc blocks lost
  • gc claim blocks lost
  • gc blocks corrupt
  • global enqueue gets sync
  • global enqueue gets async
  • global enqueue get time
  • global enqueue releases
  • gcs messages sent
  • ges messages sent
  • global enqueue CPU used by this session
  • gc CPU used by this session
  • IPC CPU used by this session
  • global undo segment hints helped (10.2)
  • global undo segment hints were stale (10.2)

1.2 V$SYSMETRIC

  • GC CR Block Received Per Second
  • GC CR Block Received Per Txn
  • GC Current Block Received Per Second
  • GC Current Block Received Per Txn
  • Global Cache Average CR Get Time
  • Global Cache Average Current Get Time
  • Global Cache Blocks Corrupted
  • Global Cache Blocks Lost

1.3 V$SEGMENT_STATISTICS, V$SEGSTAT

  • gc buffer busy
  • gc cr blocks received
  • gc current blocks received

1.4 DBA_HIST_SEG_STAT

  • GC_CR_BLOCKS_SERVED_TOTAL
  • GC_CR_BLOCKS_SERVED_DELTA
  • GC_CU_BLOCKS_SERVED_TOTAL
  • GC_CU_BLOCKS_SERVED_DELTA

1.5 V$GES_STATISTICS

  • acks for commit broadcast(actual)
  • acks for commit broadcast(logical)
  • broadcast msgs on commit(actual)
  • broadcast msgs on commit(logical)
  • broadcast msgs on commit(wasted)
  • dynamically allocated gcs resources
  • dynamically allocated gcs shadows
  • false posts waiting for scn acks
  • flow control messages received
  • flow control messages sent
  • gcs assume cvt
  • gcs assume no cvt
  • gcs ast xid
  • gcs blocked converts
  • gcs blocked cr converts
  • gcs compatible basts
  • gcs compatible cr basts (global)
  • gcs compatible cr basts (local)
  • gcs cr basts to PIs
  • gcs cr serve without current lock
  • gcs dbwr flush pi msgs
  • gcs dbwr write request msgs
  • gcs error msgs
  • gcs forward cr to pinged instance
  • gcs immediate (compatible) converts
  • gcs immediate (null) converts
  • gcs immediate cr (compatible) converts
  • gcs immediate cr (null) converts
  • gcs indirect ast
  • gcs lms flush pi msgs
  • gcs lms write request msgs
  • gcs msgs process time(ms)
  • gcs msgs received
  • gcs out-of-order msgs
  • gcs pings refused
  • gcs queued converts
  • gcs recovery claim msgs
  • gcs refuse xid
  • gcs regular cr
  • gcs retry convert request
  • gcs side channel msgs actual
  • gcs side channel msgs logical
  • gcs undo cr
  • gcs write notification msgs
  • gcs writes refused
  • ges msgs process time(ms)
  • ges msgs received
  • global posts dropped
  • global posts queue time
  • global posts queued
  • global posts requested
  • global posts sent
  • implicit batch messages received
  • implicit batch messages sent
  • lmd msg send time(ms)
  • lms(s) msg send time(ms)
  • messages flow controlled
  • messages queue sent actual
  • messages queue sent logical
  • messages received actual
  • messages received logical
  • messages sent directly
  • messages sent indirectly
  • messages sent not implicit batched
  • messages sent pbatched
  • msgs causing lmd to send msgs
  • msgs causing lms(s) to send msgs
  • msgs received queue time (ms)
  • msgs received queued
  • msgs sent queue time (ms)
  • msgs sent queue time on ksxp (ms)
  • msgs sent queued
  • msgs sent queued on ksxp
  • process batch messages received
  • process batch messages sent

2. Wait Events:

在10g中RAC等待事件可以分为3类,下面列出了主要的等待事件包括一些undocumented wait event.

2.1. Real time only :

Those wait events are only defined while the process is waiting; after the wait is over, they are reclassified according to the outcome of the global cache operation. They should appear only on the following views: V$SESSION_WAIT, V$ACTIVE_SESSION_HISTORY, V$EVENT_HISTOGRAM.  
  • gc cr request
  • gc current request

2.2. Historical only:

These are events represent the outcome of a GC request (fixup events). They can appear in the following views: V$SESSION_EVENT, V$SYSTEM_EVENT, DBA_HIST_SYSTEM_EVENT, V$EVENTMETRIC.
  • gc cr block 2-way
  • gc cr block 3-way
  • gc cr block busy
  • gc cr block congested
  • gc cr block lost
  • gc cr block unknown
  • gc cr grant 2-way
  • gc cr grant busy
  • gc cr grant congested
  • gc cr grant unknown
  • gc current block 2-way
  • gc current block 3-way
  • gc current block busy
  • gc current block congested
  • gc current block lost
  • gc current block unknown
  • gc current grant 2-way
  • gc current grant busy
  • gc current grant congested
  • gc current grant unknown

2.3. Other events:

The remaining wait events may appear in any of the views listed before, namely: V$SESSION_WAIT, V$ACTIVE_SESSION_HISTORY, V$EVENT_HISTOGRAM, V$SESSION_EVENT, V$SYSTEM_EVENT, DBA_HIST_SYSTEM_EVENT, V$EVENTMETRIC.
  • LMON global data update
  • cr request retry
  • gc assume
  • gc block recovery request
  • gc buffer busy
  • gc claim
  • gc cr cancel
  • gc cr disk read
  • gc cr disk request
  • gc cr failure
  • gc cr multi block request
  • gc current cancel
  • gc current multi block request
  • gc current retry
  • gc current split
  • gc domain validation
  • gc freelist
  • gc prepare
  • gc quiesce wait
  • gc recovery free
  • gc recovery quiesce
  • gc remaster
  • gcs ddet enter server mode
  • gcs domain validation
  • gcs drm freeze begin
  • gcs drm freeze in enter server mode
  • gcs enter server mode
  • gcs log flush sync
  • gcs remastering wait for read latch
  • gcs remastering wait for write latch
  • gcs remote message
  • gcs resource directory to be unfrozen
  • gcs to be enabled
  • ges LMD suspend for testing event
  • ges LMD to inherit communication channels
  • ges LMD to shutdown
  • ges LMON for send queues
  • ges LMON to get to FTDONE
  • ges LMON to join CGS group
  • ges cached resource cleanup
  • ges cancel
  • ges cgs registration
  • ges enter server mode
  • ges generic event
  • ges global resource directory to be frozen
  • ges inquiry response
  • ges lmd and pmon to attach
  • ges lmd/lmses to freeze in rcfg - mrcvr
  • ges lmd/lmses to unfreeze in rcfg - mrcvr
  • ges master to get established for SCN op
  • ges performance test completion
  • ges pmon to exit
  • ges process with outstanding i/o
  • ges reconfiguration to start
  • ges remote message
  • ges resource cleanout during enqueue open
  • ges resource cleanout during enqueue open-cvt
  • ges resource directory to be unfrozen
  • ges retry query node
  • ges reusing os pid
  • ges user error
  • ges wait for lmon to be ready
  • ges1 LMON to wake up LMD - mrcvr
  • ges2 LMON to wake up LMD - mrcvr
  • ges2 LMON to wake up lms - mrcvr 2
  • ges2 LMON to wake up lms - mrcvr 3
  • ges2 proc latch in rm latch get 1
  • ges2 proc latch in rm latch get 2
  • global cache busy
  • global enqueue expand wait
  • latch: KCL gc element parent latch
  • latch: gcs resource hash
  • latch: ges resource hash list
本文转自maclean_007 51CTO博客,原文链接:http://blog.51cto.com/maclean/1278190
相关文章
|
Oracle 关系型数据库 物联网
|
4月前
|
运维 Oracle 前端开发
Oracle 11g RAC集群日常运维命令总结
Oracle 11g RAC集群日常运维命令总结
112 2
|
4月前
|
Oracle 关系型数据库
分布式锁设计问题之Oracle RAC保证多个节点写入内存Page的一致性如何解决
分布式锁设计问题之Oracle RAC保证多个节点写入内存Page的一致性如何解决
|
5月前
|
存储 负载均衡 Oracle
|
5月前
|
存储 Oracle 关系型数据库
|
7月前
|
存储 Oracle 关系型数据库
Oracle RAC:数据库集群的舞动乐章
【4月更文挑战第19天】Oracle RAC是Oracle提供的高可用性数据库解决方案,允许多个实例共享同一数据库,确保业务连续性和数据完整性。通过集群件和全局缓存服务实现服务器间的协调和通信。RAC提供高可用性,通过故障转移应对故障,同时提升性能,多个实例并行处理请求。作为数据管理员,理解RAC的架构和管理至关重要,以发挥其在数据管理中的最大价值。