S8700 Server Complex - Avaya Application Solutions Deployment Manual

Reliability and Recovery

S8700 Server Complex

The high availability philosophy is scrupulously implemented the Avaya S8700 server complex:
Linux is the operating system (OS) for many reasons:
- Provides access to full source for quicker bug-fix turnaround
- Allows easy system customization for high availability enhancements
- Exhibits fewer known security flaws than other OSs, and allows customization for added
security features
Two servers with a memory shadowing link allow:
- One processor to take over when the other fails.
- Simple duplication of all other server components (for example, modem, disk, or
memory) to eliminate a single point of failure.
- Upgrades with insignificant time out of service.
High availability enhancements:
- Software sanity is continuously evaluated. Any insanity (due to unexpected conditions) is
detected, and the offending software is forced to go through escalating levels of recovery
until finally the entire system can switch over to a standby processor.
- Disks are partitioned to keep most of the variable information away from the invariant,
and to allow for automatic recovery if newly loaded software fails.
- All event logs are proactively scanned for potential service-affecting items. If found,
alarms are generated. If necessary, a service dispatch is launched.
- Applications running on the OS are thoroughly pr-tested to assure proper performance.
This OS is closed to any applications other than those provided by the manufacturer to
avoid interference of operation. Alarms can be generated if any untested software is
loaded on the system.
- Customers and technicians (in most cases) can access the operating system shell
directly, providing protection from inadvertent adverse alterations to the system.
- Enhanced tools allow secure, remote, nonstop system debugging and unattended data
collection.
234 Avaya Application Solutions IP Telephony Deployment Guide