Monitoring processes
The PATROL for Linux monitors and manages processes through the NUK_Process applications.
With these application classes, you can
- Designate specific processes for monitoring
- Review the overall status of processes on the system such as
- Total number of processes
- Total number of zombie processes
- List of all zombie processes
- List of all processes consuming excessive CPU
- Total number of processes waiting for resources
- Total number of user (nonroot) processes
- Average number of processes per user
- You can use the NUK_Process application class to monitor
- A single process on a single computer
- A local process across multiple computers within a process host group (this feature is helpful when you are monitoring processes in a clustered environment)
- Several running instances of a process on a single computer
- Several running instances of a process across multiple computers within a process host group (this feature is helpful when you are monitoring processes in a clustered environment)
- You can manage a process by way of its NUK_Process instance to
- Restart a process that has stopped
- Stop a process
- Alarm when an unauthorized User ID owns a process
- Alarm when the number of processes running in a host group drops below a specified minimum number
- Alarm when the number of processes running in a host group exceeds a specified maximum number
- Alarm when a process does not have a parent process ID (PPID) of 1 (a PPID of 1 indicates that a process is owned by init)
The following sections appear in this topic:
Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*