Running a TCL script through EEM every 60 seconds

[bharath_chin]

I have a script written in the TCLSH of IOS, I can test this script from the shell using the following commands

LAB-RTR#tclsh
LAB-RTR(tcl)#source flash:test.tcl

 NO CALLS

Is there a way I can run this test.tcl script every 60 seconds on the router.

SCRIPT test.tcl

proc get_ints {} {
set syslog [open "syslog:" w+]
#puts "BEFORE\n"
set check ""
set int_out [ exec "show voice call status\n" ]
#puts "INTOUT $int_out\n"
set mylist [regexp -nocase -line -inline {(^([0-9]*|No) active calls? found)} $int_out]

#puts "MYLIST $mylist\n"

foreach int $mylist {
 #puts "INT $int\n"
 if {$int == "No"} {
  puts " NO CALLS\n";
  puts $syslog "NO CALLS"
 } elseif {[string is integer -strict $int]} {
  puts "$int CALLS\n";
  puts $syslog "$int CALLS"
 } else {
  #puts "GARBAGE\n";
 }
}
close $syslog
}

#puts ""

puts [eval get_ints]

[Joe Clarke · Accepted solution]

Convert this to an EEM Tcl policy.  Just add:

proc exec { cmd } {

    global errorInfo

    if { [catch {cli_open} result] } {

        error $result $errorInfo

    }

    array set cli $result

    set output [cli_exec $cli(fd) $cmd]

    catch {cli_close $cli(fd) $cli(tty_id)}

    return $output

}

Then register this as an EEM Tcl policy with this at the top:

::cisco::eem::event_register_timer watchdog time 60

namespace import ::cisco::eem::*

namespace import ::cisco::lib::*

It will then run every 60 seconds.

[bharath_chin]

Hey Joseph, Thanks a lot for pointing me in the right direction. I have my script running the way I want. I was able to generate a syslog using the script, now I need to generate a SNMP trap can you give me an example of how to do it.

 action_syslog priority emergencies msg "TOO MANY CALLS" -- This works

[Joe Clarke]

Something like this would work:

action_snmp_trap strdata "TOO MANY CALLS"

[bharath_chin]

Hey Joseph, Thanks for all your help. I have one last question, so now that I am running my script every 60 seconds if there is an issue I send multiple traps to our monitoring system. Is there any way you can keep track if the trap was already sent I will not send another trap again.

One way I can think of is using a loopback interface. But I would prefer not shutting down the loopback interface just to keep track of what my script is doing.

IF issue {
   IF Loop1 is up {
            don't send trap
    } ELSE {
            bring Loop1 up
            Send Trap
    }
 } ELSE (No issue) {
   shutdown Loop1
 }
 

[Joe Clarke]

EEM has the concept of contexts where you can store data between executions.  You save data using context_save and retrieve it using context_retrieve.  Note: context data is volatile, so when it is retrieved, the context is destroyed.  See http://www.cisco.com/c/en/us/td/docs/ios/netmgmt/configuration/guide/12_2sx/nm_12_2sx_book/nm_eem_policy_tcl.html#wp1210964 for more details.

[bharath_chin]

Thanks  lot Joseph.