Surveillance Monitor
To monitor the health of PTV xServer services you need monitoring software. Many tools are available for that purpose.
A suitable choice for a free monitoring software is Nagios.
Please refer to the Administrator's Guide to learn what you can monitor.
System Redundancy
If availability is a concern, you need to provide redundant systems, preferably on multiple levels from networking to servers to processes:
-
Start at least one spare worker process per service, so that incoming requests can be processed while a crashed process is being restarted.
-
Host at least one spare server behind a load balancing server that will take care of fail-over behavior in case of server hardware failures.
-
Databases that are used as central repository for job processing have to be replicated with spare database systems.
-
Host at least one spare cluster that you can use for integration testing of new versions before going live with upgrades (staging).
-
For maximum availability, as well deploy extra networking equipment (switches, routers, firewalls, proxies, adapters) and emergency power supplies.