How to install PRM (Process Resource Monitor)

What is PRM (Process Resource Monitor) and how to install it?

  • In a nutshell:

PRM monitors the table of active cases. If any of these processes exceeds the defined resource (memory, CPU or number of processes), an email is sent to the administrator and optionally the process is stopped. “Killer.”

The settings in the prm file "conf.prm" will determine some actions of the PRM. In some process rules folder are treated separately with individual settings for them.

The PRM (Process Resource Monitor) is without doubt very useful to prevent some types of abuse and mostly unnecessary burdens to the server.

Let’s see the installation:

SSH root, download the PRM:


Unzip the file:

tar xvfz prm-current.tar.gz

Go to the folder and install:

cd prm-0.5 /
. /

At this point the PRM is already installed, now let’s set it up:

Edit the configuration file:

pico / usr / local / prm / conf.prm

In option:

# Enable kernel logging [0 = disabled, 1 = enabled] USE_KLOG = "0"

Change to:

# Enable kernel logging [0 = disabled, 1 = enabled] USE_KLOG = "1"

In option:

# Enable user e-mail alerts [0 = disabled, 1 = enabled] USR_ALERT = "0"

Change to:

# Enable user e-mail alerts [0 = disabled, 1 = enabled] USR_ALERT = "1"

** The function USR_ALERT “1” will enable sending e-mail alerts to the administrator if any case exceed the feature set.


# Email address for alerts USR_ADDR = "root"

Change to:

# Email address for alerts USR_ADDR = ""

In most situations you will not need to make any other changes below will be just doing a job description (though some are quite obvious)

# Path to user e-mail message file USR_MSG = "$ INSPATH / usr.msg"

This is the template path to the “body of the e-mail alert will be sent to the administrator.

# Subject of e-mail alerts SUBJ = "Process status report from $ HOSTNAME"

This will be the subject of the e-mail alert sent to the administrator.
(The variable $ HOSTNAME will return the hostname of your dedicated server.)

# Check 5,10,15 minute load average. [1,2,3 respective of 5,10,15] LC = "1"

This will be the time that the PRM will check the number of active cases. Where:

1 = 5 minutes
2 = 10 minutes
3 = 15 minutes

# Min load level required to run (decimal values ??unsupported) MIN_LOAD = "1"

This option sets the minimum load for the PRM is enabled.

# Seconds to wait before rechecking the flaged pid (pid's Noted resource # 
intensive but not yet killed). WAIT = "12"

Seconds to wait for a new “rechecking” of pids of active processes.

# Counter limit That the process must reach prior to kill. the counter value 
# Increases the process is resource intensive on rechecks flaged. 

Limit that a process needs to run for the rules of the PRM are checked.

# Argument to pass onto kill commands KARGO = "9"

What the argument for the kill command (default and recommended 9).

# Max CPU usage readout for a process -% of all cpu resources (decimal 
values ??unsupported) MAXCPU = "40"

Maximum CPU (in percentage) to start reading the rules

Max # MEM usage readout for a process -% of total system memory 
(decimal values ??unsupported) MAXMEM = "20"

Maximum Memory (percentage) to start reading the rules

# Max processes for a Given command - this is not max user processes for the 
executable But rather MAXPS = "25"

Maximum of processes to start reading the rules.

Finally, save (ctrl + xy) and get off the peak.

Click here for 2u colocation in nashik india