Using the agentcmd utility

There are two main scenarios for using the agentcmd utility:

• With an agent on a system that has been added to up.time. In this case, the hostname is in the up.time DataStore.
• With an agent that has been deployed on a system that has not been added to up.time. If there is no agent on the system, or there are network problems, agentcmd returns the following error message:

  ERR Unable to contact Agent (10.1.2.34 on port 9998)

You can use agentcmd in two ways: from the command line or in an agent-side script.

Using agentcmd at the command line

The command line syntax of agentcmd is:

agentcmd <-/+s> <-p #> hostname command

Where:

• -s = do not use an SSL connection to the agent (default).
• +s = connect to the agent using SSL.

  If SSL is not enabled on the host, the following error message appears:

  ERR The host was not configured to use SSL

• -p # = connect to the agent on the specified port (default is 9998).
• hostname = the name or IP address of the host on which the agent is running.
• command = the agent command to run.

For example, to get basic system information from an agent running on the server Solaris1, type the following command:

agentcmd +s Solaris1 sysinfo

The agent will return output similar to the following:

SYSNAME=Solaris1
DOMAIN=myDomain.com
ARCH="SunOS Solaris1 5.9 Generic_118558-03 sun4u sparc SUNW,Ultra-Enterprise"
OSVER=5.9
NUMCPUS=4
MEMSIZE=1048576
PAGESIZE=8192
SWAPSIZE=1021624
GPGSLO=25
VXVM=""
SDS="YES"
HOSTID="808cf673"
CPU0=" 0 - - 168 - UltraSPARC - "
CPU1=" 1 - - 168 - UltraSPARC - "
CPU4=" 4 - - 168 - UltraSPARC - "
CPU5=" 5 - - 168 - UltraSPARC - "
NET0=lo0=127.0.0.1
NET1=hme0=10.0.0.010

Using agentcmd in an agent-side script

As with netcat, agentcmd is typically used in a script on the monitoring station to perform the following tasks:

• Contact an agent system.
• Attempt to run a predefined agent-side script.
• Return the results of an agent-side script.
• Validate the status of those results.
• Return the status to up.time.

The following example illustrates how to run agentcmd in an agent side script:

NOTE: The rexec in the command below does not indicate use of the rexec system utility. It is a keyword that is used to indicate to the agent that you are attempting to run a pre-defined command.

/usr/local/uptime4/scripts/agentcmd [-s/+s] -p [agent port] [agent hostname] rexec [password] [path]

If you use netcat in your scripts, see the following Knowledge Base articles for information on replacing netcat with agentcmd:

• Creating Custom Service Monitors in up.time
• Creating Custom Service Monitors with Retained Data Collection
• Creating Plug-in Service Monitors in up.time