New mechanism of rac2--11g Grid infrastructure

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

As you can see, the origin of the initial CRS (Cluster ready Services) noun is the reason for the cluster software in 10.1. Later went through the Clusterware, to the 11g, finally changed to GI.

11g New Process:

After installation, some background processes are generated to ensure that the cluster is working properly and that it can communicate with the outside. The requirements for some of these ordered Linux platforms need to be started with root user privileges. For example, changes to the network configuration require higher permissions. Other background processes will run with the permissions of the system user on which the grid software resides. The following table describes some of the main background processes

Background process	Description
Oracle High-Availability Service (Ohas)	Ohas is the first grid infrastructure component that opens after the server starts. It is configured to open with Init (1) and is responsible for generating the agent process.
Oracle Agent	The Grid infrastructure uses two Oracle agent processes. The first one, in summary, is responsible for opening up some resources that require access to OCR and voting files. It is created by Ohas. The second agent process is created by CRSD and is responsible for opening all resources that do not require root access. This process runs with the permissions of the user who owns the grid infrastructure, and is responsible for RACG the work done in RAC11.1.
Oracle Root Agent	Similar to the Oracle proxy process, there are two root agent processes created. The initial agent process was raised by Ohas, which provides initialization for resources in the Linux system that require higher privileges. The main background processes created are CSSD and CRSD. In turn, CRSD will trigger another root agent. This agent will open the resources that require root, primary and network related
Cluster Readiness Service Process (CRSD)	Background main process of cluster software, using Oracle Cluster registration information to manage resources in the cluster
Cluster Synchronization Service Process (CSSD)	Managing the configuration and node members of a cluster
Oracle Process Monitoring (OPROCD)	OPROCD is responsible for I/O isolation in version 11.1. It was introduced in the 10.2.0.4 patch set for Linux systems. Before this patch set, the kernel Hangcheck-timer modules to do similar tasks. Interestingly, OPROCD used to be used in non-Linux platforms. Grid infrastructure replaces the OPROCD process with the cssdagent process.
Event Manager (EVM)	EVM is responsible for publishing the events created by grid infrastructure
Cluster time Synchronization Service (CTSS)	The CTSS service is an option that provides time synchronization for a cluster over a network time-contract server, which is important for RAC. It can be run in two modes: watching or moving. When NTP is activated, it runs in observation mode, and if NTP is not started, it synchronizes all nodes ' time based on the master node.
Oracle Warning Service (ONS)	The main background process responsible for publishing events through the Rapid application framework.

In RAC11.2, the startup order of Grid infrastructure has changed significantly. Instead of opening the CRS, CSS, and evm,ohas processes directly through Inittab (5), the process is now primarily responsible for creating agent processes, monitoring the health of other nodes, and opening cluster resources. In a non-Oracle management process, NTP is a special role, and in each cluster it needs to provide clock synchronization, and Grid infrastructure is no exception.

The common CRS commands are as follows:

$ORA _crs_home/bin/crs_stat–t This command is primarily used to check the CRS resource status (can be performed with root,grid,oracle users, but the standard should be performed with a grid user)

Ora_crs_home/bin/crs_start–all This command is used to start all cluster services, but the prerequisite for booting is that the resource state cannot be unknown.

$ORA _crs_home/bin/crs_start Cluster service name
$ORA _crs_home/bin/crs_stop-all
$ORA _crs_home/bin/crs_stop Cluster service name (Cluster service name can be obtained by $ora_crs_home/bin/crs_stat)

A set of CRS commands above is used to initiate the shutdown of the CRS service, which is described below through a real-world environment, specifically explaining the use of these commands.

Examples are as follows:
Oracle RAC restarts the RAC after installation, some services do not start, and manual startup is useless.
The following is the system automatically start after the situation;

Some other common commands:
$ORA _crs_home/bin/crsctl Check CRS for background process status
$ORA _crs_home/bin/crsctl Start Resources starts the CRS resource
$ORA _crs_home/bin/crsctl start CRS starts CRS and requires super user operation.
$ORA _crs_home/bin/crsctl stop CRS shuts down CRS and requires superuser operation.
$ORA _crs_home/bin/crs_unregister to unregister CRS
_crs_home/bin/ocrcheck Checking OCR setup information with Ocrcheck $ORA
$ORA _crs_home/bin/crsctl query CSS votedisk Check the voting disk information

Example:
[Email protected] css]$ Ocrcheck
Status of Oracle Cluster Registry is as follows:
Version:2
Total Space (Kbytes): 104344
Used Space (Kbytes): 4568
Available Space (Kbytes): 99776
id:328160432
Device/file Name:/dev/raw/raw4
Device/file Integrity Check Succeeded
Device/file Name:/dev/raw/raw5
Device/file Integrity Check Succeeded

Cluster Registry Integrity Check succeeded

[Email protected] css]$ $ORA _crs_home/bin/ocrcheck
Status of Oracle Cluster Registry is as follows:
Version:2
Total Space (Kbytes): 104344
Used Space (Kbytes): 4568
Available Space (Kbytes): 99776
id:328160432
Device/file Name:/dev/raw/raw4
Device/file Integrity Check Succeeded
Device/file Name:/dev/raw/raw5
Device/file Integrity Check Succeeded

Cluster Registry Integrity Check succeeded

[Email protected] css]$ crsctl query CSS Votedisk
0.0/dev/raw/raw6
1.0/dev/raw/raw7
2.0/dev/raw/raw8

Located 3 Votedisk (s).
Attention:
The CRSCTL command can be either root or Oracle, if the root user is to specify an absolute path, but for Crsctl start CRS and CRSCTL stop CRS must be the root user to execute.

CRS Main Process

(1) CRSD
Responsible for managing HA operations
Manage CRS resources, such as LINSTENER,VIP,ONS,GSN, etc.
Managed, initiated by root user
(2) OCSSD
Manage the relationships of each node for inter-node communication
Run by Oracle User management
(3) Oprocd
Cluster process management-processmonitor for the cluster.
Run only in a cluster software state that is not using vendor
(4) EVMD
Event detection process, run by Oracle User management
(5) Main log location
$ORA _crs_home/log/Node host name/RACG
$ORA _crs_home/log/Node host name/CRSD
$ORA _crs_home/crs/init
$ORA _crs_home/css/log
$ORA _crs_home/css/init
$ORA _crs_home/evm/log
$ORA _crs_home/evm/init
$ORA _crs_home/srvm/log

Ext.: http://blog.csdn.net/cymm_liu/article/details/7899052

New mechanism of rac2--11g Grid infrastructure