oracle Clusterware stack consists of two separate components: the Cluster Ready Service (CRS) daemon (crsd) and Oracle High Availability Service daemon (ohasd) and other components.
List This section describes the composition of the CRS, the list includes service on the process or windows on Linux and UNIX.
CRS: The main program for managing a cluster of high availability operations. crs daemon (crsd ocr based on configuration information stored in each resource to manage cluster resources. This includes starting, stopping, monitoring, and failover operation .crsd resource status change is that the process produces speaking event. After installing Oracle RAC, the crsd process speaks monitoring Oracle instance, listener, etc. and automatically restart them when a failure occurs..
CSS: cluster synchronization service, through which nodes control the cluster members and inform members when nodes join or leave the cluster to manage the cluster configuration. If you use a certified third-party clustering software, the process will css docking node membership information to manage your cluster software. the cssdagent process monitoring cluster and provides I / O shield. the service used to be by Oracle Process Monitor Daemon (oprocd) provides, cssdagent failure may lead to an Oracle cluster node is restarted.
Oracle ASM: provides disk management to Oracle Clusterware and Oracle Database.
CTSS:. cluster time synchronization service provides time management cluster for Oracle Clusterware
EVM: event management released Oracle Clus daemon events terware created.
ONS: Oracle for the Notification Service Publish and Subscribe service fast application notification.
Oraagent: Oracle Agent Extension clustering software to support Oracle-specific requirements and complex resources when FAN events. this process will run the server script mark, the process is referred to as RACG in Oracle Clusterware 11.1 in.
Orarootagent: Oracle Root Agent a special oraagent process, can help crsd manage root resource owners, such as network and grid virtual IP of the cluster synchronization service (css) Event Management (EVM) and ONS components with the same cluster database environment other node other components of the cluster communication layer on. These components are the primary communication link between Oracle databases, applications, and Oracle Clusterware. In addition, these background processes to monitor and manage database operations.
Oracle High Availability Services Architecture
This section describes the process of composing Oracle High Availability Service.
Ologgerd: Cluster recorder accepts service from all nodes in the cluster information, and stored in CHM-based repository database, the service only runs on a cluster of two nodes.
osysymond: System Monitor Service to send data to the monitoring service and the recording cluster operating system service metrics. This service runs in each nodes.
GPNPD: Provides access to plug and play configuration files, configuration files and coordination between the cluster nodes updated to ensure that all nodes have the latest profile.
GIPC: one Aspect process, which enables the use of redundant interconnection.
GNS: an external processing request sent by the DNS server, the name of the cluster definition resolution.
Summary：Fragmented learning may not work for you.The next blog introduces OCR,Voting Disk,Split Brain,heartbeat in RAC.