The probes in this section monitor essential aspects of your Linux systems, from CPU usage to virtual memory. Apply them to mission-critical systems to obtain warnings prior to failure.
Unlike other probe groups, which may or may not require the Red Hat Network Monitoring Daemon, every Linux probe requires that the rhnmd daemon be running on the monitored system.
The Linux::CPU Usage probe monitors the CPU utilization on a system and collects the following metric:
CPU Percent Used — The five-second average of the percent of CPU usage at probe execution.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to run this probe.
The Linux::Disk IO Throughput probe monitors a given disk and collects the following metric:
Read Rate — The amount of data that is read in kilobytes per second.
Write Rate — The amount of data that is written in kilobytes per second.
To obtain the value for the required Disk number or disk name field, run iostat on the system to be monitored and see what name has been assigned to the disk you desire. The default value of 0 usually provides statistics from the first hard drive connected directly to the system.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe. Also, the Disk number or disk name parameter must match the format visible when the iostat command is run. If the format is not identical, the configured probe enters an UNKNOWN state.
| Field | Value |
|---|---|
| Disk number or disk name* | 0 |
| Timeout* | 15 |
| Critical Maximum KB read/second | |
| Warning Maximum KB read/second | |
| Warning Minimum KB read/second | |
| Critical Minimum KB read/second | |
| Critical Maximum KB written/second | |
| Warning Maximum KB written/second | |
| Warning Minimum KB written/second | |
| Critical Minimum KB written/second |
Table C-16. Linux::Disk IO Throughput settings
The Linux::Disk Usage probe monitors the disk space on a specific file system and collects the following metrics:
File System Used — The percentage of the file system currently in use.
Space Used — The amount of the file system in megabytes currently in use.
Space Available — The amount of the file system in megabytes currently available.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Inodes probe monitors the specified file system and collects the following metric:
Inodes — The percentage of inodes currently in use.
An inode is a data structure that holds information about files in a Linux file system. There is an inode for each file, and a file is uniquely identified by the file system on which it resides and its inode number on that system.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Interface Traffic probe measures the amount of traffic into and out of the specified interface (such as eth0) and collects the following metrics:
Input Rate — The traffic in bytes per second going into the specified interface.
Output Rate — The traffic in bytes per second going out of the specified interface.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
| Field | Value |
|---|---|
| Interface* | |
| Timeout* | 30 |
| Critical Maximum Input Rate | |
| Warning Maximum Input Rate | |
| Warning Minimum Input Rate | |
| Critical Minimum Input Rate | |
| Critical Maximum Output Rate | |
| Warning Maximum Output Rate | |
| Warning Minimum Output Rate | |
| Critical Minimum Output Rate |
Table C-19. Linux::Interface Traffic settings
The Linux::Load probe monitors the CPU of a system and collects the following metric:
Load — The average load on the system CPU over various periods.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Memory Usage probe monitors the memory on a system and collects the following metric:
RAM Free — The amount of free random access memory (RAM) in megabytes on a system.
You can also include the reclaimable memory in this metric by entering yes or no in the Include reclaimable memory field.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Process Counts by State probe identifies the number of processes in the following states:
Blocked — A process that has been switched to the waiting queue and whose state has been switched to waiting.
Defunct — A process that has terminated (either because it has been killed by a signal or because it has called exit()) and whose parent process has not yet received notification of its termination by executing some form of the wait() system call.
Stopped — A process that has been stopped before its execution could be completed.
Sleeping — A process that is in the Interruptible sleep state and that can later be reintroduced into memory, resuming execution where it left off.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
| Field | Value |
|---|---|
| Timeout* | 15 |
| Critical Maximum Blocked Processes | |
| Warning Maximum Blocked Processes | |
| Critical Maximum Defunct Processes | |
| Warning Maximum Defunct Processes | |
| Critical Maximum Stopped Processes | |
| Warning Maximum Stopped Processes | |
| Critical Maximum Sleeping Processes | |
| Warning Maximum Sleeping Processes | |
| Critical Maximum Child Processes | |
| Warning Maximum Child Processes |
Table C-22. Linux::Process Counts by State settings
The Linux::Process Count Total probe monitors a system and collects the following metric:
Process Count — The total number of processes currently running on the system.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Process Health probe monitors user-specified processes and collects the following metrics:
CPU Usage — The CPU usage rate for a given process in milliseconds per second. This metric reports the time column of ps output, which is the cumulative CPU time used by the process. This makes the metric independent of probe interval, allows sane thresholds to be set, and generates usable graphs (i.e. a sudden spike in CPU usage shows up as a spike in the graph).
Child Process Groups — The number of child processes spawned from the specified parent process. A child process inherits most of its attributes, such as open files, from its parent.
Threads — The number of running threads for a given process. A thread is the basic unit of CPU utilization, and consists of a program counter, a register set, and a stack space. A thread is also called a lightweight process.
Physical Memory Used — The amount of physical memory (or RAM) in kilobytes used by the specified process.
Virtual Memory Used — The amount of virtual memory in kilobytes used by the specified process, or the size of the process in real memory plus swap.
Specify the process by its command name or process ID. (PID). Entering a PID overrides the entry of a command name. If no command name or PID is entered, the error Command not found is displayed and the probe will be set to a CRITICAL state.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
| Field | Value |
|---|---|
| Command Name | |
| Process ID (PID) file | |
| Timeout* | 15 |
| Critical Maximum CPU Usage | |
| Warning Maximum CPU Usage | |
| Critical Maximum Child Process Groups | |
| Warning Maximum Child Process Groups | |
| Critical Maximum Threads | |
| Warning Maximum Threads | |
| Critical Maximum Physical Memory Used | |
| Warning Maximum Physical Memory Used | |
| Critical Maximum Virtual Memory Used | |
| Warning Maximum Virtual Memory Used |
Table C-24. Linux::Process Health settings
The Linux::Process Running probe verifies that the specified process is functioning properly. It counts either processes or process groups, depending on whether the Count process groups checkbox is selected.
By default, the checkbox is selected, thereby indicating that the probe should count the number of process group leaders independent of the number of children. This allows you, for example, to verify that two instances of the Apache HTTP Server are running regardless of the (dynamic) number of child processes. If it is not selected, the probe conducts a straightforward count of the number of processes (children and leaders) matching the specified process.
Specify the process by its command name or process ID. (PID). Entering a PID overrides the entry of a command name. If no command name or PID is entered, the error Command not found is displayed and the probe enters a CRITICAL state.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Swap Usage probe monitors the swap partitions running on a system and reports the following metric:
Swap Free — The percent of swap memory currently free.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::TCP Connections by State probe identifies the total number of TCP connections, as well as the quantity of each in the following states:
TIME_WAIT — The socket is waiting after close for remote shutdown transmission so it may handle packets still in the network.
CLOSE_WAIT — The remote side has been shut down and is now waiting for the socket to close.
FIN_WAIT — The socket is closed, and the connection is now shutting down.
ESTABLISHED — The socket has a connection established.
SYN_RCVD — The connection request has been received from the network.
This probe can be helpful in finding and isolating network traffic to specific IP addresses or examining network connections into the monitored system.
The filter parameters for the probe let you narrow the probe's scope. This probe uses the netstat -ant command to retrieve data. The Local IP address and Local port parameters use values in the Local Address column of the output; the Remote IP address and Remote port parameters use values in the Foreign Address column of the output for reporting.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
| Field | Value |
|---|---|
| Local IP address filter pattern list | |
| Local port number filter | |
| Remote IP address filter pattern list | |
| Remote port number filter | |
| Timeout* | 15 |
| Critical Maximum Total Connections | |
| Warning Maximum Total Connections | |
| Critical Maximum TIME_WAIT Connections | |
| Warning Maximum TIME_WAIT Connections | |
| Critical Maximum CLOSE_WAIT Connections | |
| Warning Maximum CLOSE_WAIT Connections | |
| Critical Maximum FIN_WAIT Connections | |
| Warning Maximum FIN_WAIT Connections | |
| Critical Maximum ESTABLISHED Connections | |
| Warning Maximum ESTABLISHED Connections | |
| Critical Maximum SYN_RCVD Connections | |
| Warning Maximum SYN_RCVD Connections |
Table C-27. Linux::TCP Connections by State settings
The Linux::Users probe monitors the users of a system and reports the following metric:
Users — The number of users currently logged in.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.
The Linux::Virtual Memory probe monitors the total system memory and collects the following metric:
Virtual Memory — The percent of total system memory - random access memory (RAM) plus swap - that is free.
Requirements — The Red Hat Network Monitoring Daemon (rhnmd) must be running on the monitored system to execute this probe.