-
What is RAC
The real-time application cluster is a cluster working mode of Oracle. Compared with the single-point Oracle instance mode, the advantages are as follows:
- Multi-node load balancing
- Provide high availability
- Improving transaction response time through parallel execution techniques
- Increase transactions and connections per second with horizontal scaling
- Save hardware cost
- good scalability
2. How to monitor RAC
JDBC SSH AHENT
3. RAC monitoring items
Scan information
An interface that facilitates the connection of the client. This is a unique name. When the client connects, it only needs to know this name and connect. Each SCAN VIP corresponds to a scan listener, and the service inside the cluster is in each scan listener. There are registrations on the Internet, the scan listener accepts the client's request, and foward to different Local listeners, or the local listener provides services to the client.
SACN status
node information
It is necessary to display the basic information of each instance node in the RAC cluster environment, including:
Node name, instance name, node type, node status, node VIP, Public IP, real management IP
OCR (Cluster Registration Service) status
OCR stores all configuration information related to the cluster and RAC database. And this configuration information is public and shared by multiple nodes. So OCR will be stored on the shared disk.
OCR status monitoring: version, total space size, used space, free space,
Voting Disk Status
Each node in the cluster periodically evaluates its own health status, and then puts the health status on the voting disk. Each node will check the health status of each other and update it on the voting disk. When a cluster node fails, voting arbitration will be performed through the voting disk.
Voting disk monitoring: location number, status, file common ID, file name, disk group
CRS (cluster ready service) status
CRS is generally used to build Oracle's parallel database, that is, RAC, but in addition to the interface with RAC, CRS also provides a set of high-availability application programming interfaces (APIs) to build high-availability clusters for general applications, namely Generally, we often refer to dual-machine hot backup, such as using CRS to implement MySQL dual-machine hot backup.
database instance
An instance is a collection of "memory" and "background processes". A database is the physical storage of data. Note in particular that one instance can be used for one database, and multiple instances can also be used for one database at the same time. The relationship between an instance and a database is a one-to-many relationship.
Monitor database instance content: database instance name, status, and node.
VIP status
VIP virtual network protocol. When RAC node 1 fails, the VIP of this node 1 drifts to another node 2 in the cluster. The public network card is configured with a VIP address. When a node in the cluster is down, the purpose is to enable the application to seamlessly and quickly use any node of the RAC database (except the downtime node), which improves the availability (HA) of the database. The normal operation of the business is guaranteed.
VIP monitoring items: VIP name, VIP, VIP status, node.
Cluster resource status
ASM information/IO
Through the above monitoring items, a RAC structure status diagram can be drawn:
- Monitoring effect display
