Repeating action not acting as expected

Usage and Config
1

I’m just starting to learn how Bigfix/IEM works. just to test the capability and how things work, I wrote a quick action script and having it act like a policy. The action script is simple:

dos echo %DATE%_%TIME% LoopTest3 >> c:\test.txt

…then set the Reapply this action > whenever it becomes relevant again. and no limit.

I found another post that talked about having to remove it being relevant, then making it relevant again. so for my testing, here’s the relevance I set.

(version of client >= "6.0.0.0") AND ((exists true whose (if true then (exists (computer name) whose (it as string as lowercase = "MYNAME" as lowercase)) else false)) AND (exists true whose (if true then ((not exists last active time of it or (now - last active time of it) > (5 *minute)) of action) else false)))

Issue is, I’m not getting it to run every 5min.

Fri 03/06/2015_11:01:38.85 yay
Fri 03/06/2015_11:20:58.72 yay
Fri 03/06/2015_12:53:06.51 two
Fri 03/06/2015_13:22:09.12 two
Fri 03/06/2015_13:25:27.23 LoopTest
Fri 03/06/2015_13:59:30.46 LoopTest
Fri 03/06/2015_14:24:41.17 LoopTest2
Fri 03/06/2015_14:31:23.19 LoopTest2
Fri 03/06/2015_15:11:45.64 LoopTest2
Fri 03/06/2015_15:46:57.62 LoopTest2
Fri 03/06/2015_15:53:01.50 LoopTest2
Fri 03/06/2015_16:31:17.03 LoopTest2
Fri 03/06/2015_16:37:35.95 LoopTest2
Fri 03/06/2015_16:47:46.89 LoopTest2
Fri 03/06/2015_16:53:37.25 LoopTest2

…in fact, I can only really get it to run when I right click in the console on the PC and send refresh. Any idea what I’m doing wrong?

2

You aren’t doing anything wrong. This is probably dependent on your content loop. The action when it completes goes back into the background loop to see if it comes relevant again. It just may be taking that long to get around the loop

Try a property/analysis of

average duration of evaluationcycle of client

and it will tell you how long your client is taking on average to get around the loop

1 Like
3

Well, it was kind of funny. that query returned 9 min, but I was still getting an hour before each line. Our IEM admin told me to crank up the CPU (10%) on the agent and I got it down to 6-7min. The usage of something like this would have been for a post imaging task so cranking up the agent CPU would be acceptable.

Still it was strange the difference in the analysis and reality.

4

The evaluationcycle results do have to be taken with some knowledge. If you are constantly doing actions on an endpoint it will appear shorter as a “cycle” in those cases will only include actions. When an action begins processing we start paying attention to only the actions so some fixlets will not get re-evaluated in that cycle as we only will go through the sites that can contain actions ( actionsite, mailboxsite, opsites )

The “average” is an average over the last 10 cycles so it generally is showing a good result.