Personally think that an OPS person should be the most expected day of work schedule:
40%, deal with urgent and important matters, 60%, carry out important non-urgent work. Urgent important is easy to understand, in fact, is to save the class work. Important not urgent work, is the most able to reflect the operation and maintenance of the value of the work.
Monitoring system, this is a big topic. In addition to passively monitoring the normal operation of all types of services, there are proactively developing systems to assist with system analysis, as well as a planned development tool-based system for the future of the entire system. Tools to improve the productivity of yourself and everyone in your team, especially those that can be quickly resolved.
Study, this is the most important. Operation and maintenance involves a very wide range of knowledge, continuous learning in order to successfully solve the problems in the work, while continuing to try, continue to experience only enough experience. In the event of God killing God, the Buddha killed Buddha!
One day, do important not urgent work, can make operation and maintenance work more efficient, the whole system more stable, the future development more predictable.
The experience of Linux operation and maintenance is small