Critical System Protection

Hi:

Can anyone point me to the direction to make an SCSP agent send a heartbeat status event as a log entry at the SCSP manager?
And the next question is, how to schedule it twice a day?
Either the heartbeat is pulled by the manager or triggered by the agent (by 12 hours cycle), it doesn't matter. Important thing is: it is configured from SCSP console (and not by batch jobs are scheduler at the client side).

The green status dot in the console is good enough to see that the agents are alive (for the last 5 minutes), but we want to forward events, including heartbeat, to a centralized event collector.

Thanks!

I have not attempt to do this but have wondered the same thing. I imagine it is possible to do this if you can monitor something that will consistantly generate a achange that can be detected and alerted on.  The  ...\IDS\log\SISIDSHearbeat.ini file may be a good candidate for this. The "ids.service.updatetime=" value in the file should always be updated every 5 min (or whatever the heartbeat is set to).  A filewatch rule that polls the file twice a day should do it.  Again, I have not done this just a theory.

You could also get fancy by working with on of the following Datetime paramters types:

datetimevalue:  Specify a date and time. The value is based on an agent’s local time or Coordinated Universal Time (UTC). For example, if the parameter value specifies 9/1/1960 5:00 AM and the local system's time zone is EST (UTC -0500), then the agent must add 5 hours to the parameter value, resulting in 9/1/1960 10:00 AM (UTC), so that the event actually happens at 5:00 AM local time.
In the Date box, adjust the date.
Select the Use Each Agent’s Local Time check box to use the agent’s local time. To use Coordinated Universal Time (UTC), clear the check box, and then select a UTC from the Time Zone combo box.  

datetimeduration:  Select the date and time duration by adjusting the values for Duration, Hours, Minutes, and Seconds.  

datetimerepeat:  Select the date and time repeat cycle (never, daily, weekly, monthly, yearly).

hi timl
that sounds a good idea that i want to explore. must find time to test it in virtual environment.
on my side, i tested monitoring for the 'system uptime' event at 12:00 noon time. it is generating the event all right but only once a day. your solution sounds more close to the thing i envisioned.