Summary

This chapter described Solaris processes and the various Solaris utilities available to monitor them. Using commands such as ps, prstat, pargs, sdtprocess, and the SMC Process Tool, you can view all the attributes associated with a process. In addition, we described foreground and background jobs.

The concept of sending signals to a process was described. A signal is a message sent to a process to interrupt it and cause a response or action. You also learned how to send signals to processes to cause a response such as terminating a process.

Setting process priorities was described. We also described the concept of projects and tasks along with administrative commands used to administer them. The various commands, such as nice and priocntl, that are used to set and change process priorities were described. In addition, you learned how to use the crontab and at facilities. You can use these facilities to submit batch jobs and schedule processes to run when the system is less busy, to reduce the demand on resources such as the CPU and disks.

The system administrator needs to be aware of the processes that belong to each application. As users report problems, the system administrator can quickly locate the processes being used and look for irregularities. By keeping a close watch on system messages and processes, you'll become familiar with what is normal and what is abnormal. Don't wait for problems to happenwatch system messages and processes daily. Create shell scripts to watch processes for you and to look for irregularities in the system log files. By taking a proactive approach to system administration, you'll find problems before they affect the users.

In Chapter 6, "Managing the LP Print Service," we'll explore another topic that you'll need to become acquainted withthe LP Print Service, the facility responsible for printing within the Solaris environment.