Alert Not Working

[Garrett Fletcher]

Hello,

I set up an alert, saved the alert file, and stopped and started the Monitor Daemon,  but it is not doing anything.

I want the alert to fire when the status of a outbound thread (LogicianDTS_lab_out) says there are more than 5 ‘Pending’. The action should shut down the inbound thread (med_lab_in).

Can someone please help me?

Garrett

Here are the specs for the alert:

Trigger

Alert Type: Outbound Queue Depth

Source: LogicianDTS_lab_out

Source Count: All

Comparing: > 5

Duration: Once

Alert Host:

Time Window: ******

Action

exec {hcicmd -p logician -c “med_lab_in pstop”}

Topic

[James Cobane]

Garrett,

Take a look at the hcimonitord.log and/or hcimonitord.err files (under $HCISITEDIR/exec/hcimonitord ).  You should be able to see in those files if the alert is triggering or if there is some error occurring.  Did you also specifiy a ‘notify’ action so you can see the window pop-up to show that the alert triggered?  Based on your definition, it looks like it should work with no problem.

Jim Cobane

Henry Ford Health

[Bob Richardson]

Greetings,

We discovered that you need to add the ampersand (&) at the end of the shell command to run them in the background; otherwise, they may hang up the monitor daemon while waiting for the shell command to complete.

It is possible that this command is not getting any time from the system (OS level) as it is “buried” at the same run level as the monitor daemon and thus does not get “spawned” or “forked” (I think that is the term in Unix).

Just a note: we are a Unix shop (AIX) so I am not sure how you would specifiy a shelll command to run in the background on a Windows platform.

Just a suggestion as your alert appears correctly configured.

Good luck!

[Author]

Replies