 | |||||||||||
| |||||||||||
 |
|||||||||||
| |
| ||||||||||||||||
| ShaoLin HA Cluster 1.0 Installation Manual and Operations Guide | ||
|---|---|---|
| Prev | Chapter 2. Cluster Hardware Configurations and Installation | Next |
ShaoLin Ha Cluster supports the stonith I/O fencing facility. ShaoLin HA Cluster currently certified the WTI RPS-10 serial power switch. Other power switches are also supported by stonith, please refer to the Linux HA website for the latest information on the support.
ShaoLin HA Cluster supports the stonith I/O fencing facility. This requires power switches which enable a cluster system to power-cycle the other cluster system before the STANDBY NODE to failover the system. Power switches protect against data corruption if an unresponsive or hanging system becomes responsive after system failed over, and issue I/O to the shared disks. The ability to remotely disable a system ensures data integrity under any failure condition. It is recommended that production environments use power switches in the cluster configurations.
When a cluster configuration uses a physical power switch, each cluster system's power cable is connected to a power switch which this power switch is controlled by the other cluster either through serial port or network communications. This configuration enables the other cluster to physically control the power of the other cluster, power off the other active cluster before the standby cluster actually failover the system.
It is strongly recommended to use power switches in a cluster configuration. ShaoLin HA Cluster will still function properly without a power switch, only in certain cases which a power switch will be useful. Administrators who are aware of the risks may choose to set up a cluster without the use of power switches.
UPS (Uninterpretable Power Supply) is a power source device which contains backup batteries and electrical filters to ensure highly available and stable power source. It is possible to configure two UPS (one per server) for maximum reliability and manageability. Consider a single source of power supply for a cluster, powering both clusters and the shared disks, the UPS will be a single point of failure of the system. With individual UPS, UPS and power source can be removed in case of failure or maintenance. In addition, a single UPS may not have enough power to power all the cluster hardware to an adequate amount of time to enable a clean system shutdown. Therefore, it is highly recommended to use one UPS per server. If your server has redundant power supplies, you should connect each power cord to individual UPS. This can achieve a fully redundant power source configuration. See the following diagram.
If the shared storage subsystem has two power supplies with separate power cords, set up two UPS systems and connect one power switch (or one cluster system's power cord s no using power switches) and one of the storage subsystem's power cords to each UPS system. This will ensure redundant power supply configuration.
