3D Streams Monitor help page

Streams monitor is an web application which provides complex online overview of the Oracle Streams enviroment that is used in the Distributed Deployment of Database(3D) project - part of LCG project.

Monitor pages

Overeview
Displays basic monitor status about monitored enviroment - number of databases, streams setups and streams processes currently active or disabled
Maps view
Provides streams connections topology of each experiment.
Databases view
Listing of all participating databases with its main metrics
Streams view
Dashboard view of active streams with LCRs flow and processes status
Graphs generator
Generates plots
Error view
Listing of all errors and warining in streans enviroments
Availability view
Displays percentag of database availability during last days
Reports repository view
Provides access to the reports repository directory

Monitored metrics

All dynamic stats are cumulative (except latency) - encountered since start of the instance or process. Additionaly metrics with "/s" shows current payload per secound

Instance stats
  • Id - identity number of instance - if database is non RAC it is always 1. Otherwise it is correspondent to inst_id displayed in GV$ views
  • Name - name of the instance - usually SID+inst_id
  • Stream pool size - current stream pool size
  • Stream pool free - amount of not used pool
  • Stream pool usage - percentage usage retio
  • Stream pool usage - percentage usage retio
  • Redo size - amount of redo data generated since start of the instance
  • Redo generated per sec - current growth of redo size
  • Bytes read per sec - current physical input from datafiles
  • Bytes read per sec - current physical output from datafiles

Capture process
  • Instance nr - number of an instance on which the process resides
  • Queue id - id number of destination queue where process ships captured LCRs
  • LCRs captured - number of logical change records that have been read by log miner from redo log
  • LCRs enqueued - number of logical change records that have been assigned by the capture rules to be replicated and inserted to the queue
  • Capture latency - time between creation of an entry in the redo log and being read by log miner
  • Enqueue latency - time between creation of an entry in the redo log and being enqueued by the capture process
  • Type - type of process:
    • Local - capture process uses local instance redo logs for capturing changes
    • Downstream (source) - capture process uses redo logs downloaded from the logs source database
  • State - current state of the capture process
  • Info - homemade diagnostic:
    • F - process is under flow control
    • B - process is a bottleneck
  • Error time - time of the last process error
  • Error message - oracle process error message
  • Last SCN enqueued - last enqueued system change number before process crash(appears only when process was aborted or disabled)

Propagation process
  • Source queue id - id number of source queue of LCRs to be propagated. When streams setup is consistent queue id should be the same for propagation and capture process.
  • LCRs propagated number of LCRs that have been propagated to the destination database
  • Bytes propagated - size of data propagated to the desination database
  • Destination queue - name of the process target queue at the destination database where propagated LCRs are enqueued
  • _Destination database - name of the replica(target database)
  • State - state of the process
  • Error time - time of the last process error
  • _Error message - oracle process error message

Apply process
  • Instance number - id number of the instance where process resides
  • Queue id - id number of the source queue from which process dequeues LCRs to be applied
  • LCRs dequeued - number of LCRs dequeued by the process
  • LCRs applied - number of LCRs applied to the database
  • Dequeue latency - time between creation(at the source) of last applied LCR and being dequeued.
  • Total replication latency - time between creation (at the source) of last apllied LCR and being applied.
  • Txns recived - number of transactions recived by the process
  • Txns assigned - number of transactions assigned to be applied by the process.
  • Txns applied - number of transaction applied by the process. Transaction is applied when all representing it LCRs are applied
  • State - process state
  • Info - "home made" diagnostic of the process work:
    • B - process is a bottleneck
  • Error time - time of the last process error
  • _Error message - oracle process error message
  • Last SCN applied - last applied system change number before process crash(appears only when process was aborted or disabled)
Queue status
  • Outstanding - current number of LCRs in the queue waiting for dequeueing
  • Cumulative messages - number of enqueued LCRs since queue creation
  • Cumulative spilled - number of LCRs spilled to the disk since queue creation. Spilling happens when there is no enough space in the queue buffers.


This topic: PSSGroup > PhysicsDatabasesSection > LCG3DWiki > ServiceStatus > StreamsMonitorHelp
Topic revision: r5 - 2007-07-10 - ZbigniewBaranowski
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback