Research on TFs dual-machine hot backup mechanism

Research on TFs dual-machine hot backup mechanism

1. Introduction to heatbeat dual-machine Hot Backup principles

Hot Standby is a high-reliability application solution that combines software and hardware. The dual-machine hot standby system consists of two independent servers and corresponding control software. The two servers act as a single system in the network and as an independent network IP address, and control and management in a single system mode. The system mirrors the data and running status (including hard disk data and memory data) of the two servers, enabling Hot Backup between the two servers and seamless switching, this provides stable and reliable services to external users and achieves high availability solutions for dual-host systems.

Each server is assigned a fixed IP address, and an independent IP address is also specified as the cluster IP address. This IP address is an interface provided by the dual-host system for external services. One of the two servers becomes the master node (primary server), and the other becomes the slave node (standby server). The heartbeat signal between the master node and the slave node remains at a fixed interval, the heartbeat signal is used to implement periodic handshakes between two peer systems, so as to monitor the running status of each other and detect faults on the host and network.

The dual-machine hot backup system uses the memory image technology to mirror the data and running status (including hard disk data and memory data) of the two servers, you can also use a third-party disk array cabinet in a dual-host system.

The dual-Host Control Software is Linux-haheartbeat. In addition to installing the operating system and related application services, you also need to install Heartbeat Control Software. By configuring the heartbeat software, you can specify the master node and slave node for the two servers, specify the interval of Heartbeat signals, and specify the cluster IP address, subnet mask, broadcast address, and authentication method, you can also set the service started by heartbeat, which is finally provided by the dual-host system through the cluster IP.

The dual-Machine backup system can be configured in two ways: "Two-machine system based on system switching" and "two-Machine System Based on System Image". The two have the following differences:
(1) The dual-host system based on system switching only mirrors the hard disk data of the two servers. If the master node fails, the system-level switching will be performed from the slave node. This method cannot achieve seamless switching, but implements cold backup.
(2) The dual-host system based on the system image mirrors both the hard disk data and memory data of the two servers. To ensure a full memory image, the two servers need to maintain a continuous, two-way Image Process in real time through a dedicated connection network segment. This process ensures data synchronization between the two servers. Once a fault occurs, no matter whether one server is invalid or the data and applications on the server fail, the system will immediately switch to another server, in this way, you can continue to use the services provided by the dual-host system without interruption. This method can achieve seamless switching and achieve hot backup.

Ii. TFs local environment application
1. in the current situation, TFs also achieves Hot Backup through HA's heartbeat disaster tolerance. It also requires two machines, namely (nameserver) master and slave, to bind an external VIP, configure ha. CONF file.

2. in the local TFS, only one namaserver (192.168.5.21) and one dataserver (192.168.5.20) are used. nameserver does not adopt the master-slave replication function. Through comprehensive comparative analysis, the two-node Hot Backup mechanism is used, we recommend that you use a system image instead of a system switchover system. However, on machines 21, the hard disk data is not affected (the hanging points are different), but the data in the system memory is not clean. Because 21 machines have memcache, SOLR, TFS and other services, the backup and disaster tolerance data is not pure under the dual-machine hot backup mechanism based on system images rather than system switching.