On Jun 29, 2009, at 10:14 PM, NickH wrote: > Hi all, I've got a reasonably simple problem that I'm sure can be > solved with ZenOSS. > > We use some automation software (Automate 6) on a Windows host that > has a bad habit of becoming unresponsive, usually in the middle of > the night. Detecting this through WMI, etc hasn't proven fruitful. > > As a solution, I'm hoping to have the automation software run a very > simple task (eg send an SNMP trap to ZenOSS) on a regular basis, and > have ZenOSS alert if the watchdog isn't serviced for a period of time. > > Anyone know if this type of thing is easily possible?
As James mentioned, Zenoss does have a built-in heartbeat mechanism that it uses to alert when one of the daemons becomes unresponsive. It uses the same strategy you're looking for, and you can use it yourself. I'd suggest that you use the SNMP trap you mentioned, and create an event mapping for it within Zenoss to make use of the heartbeat system. Let's say you send in a trap with an OID of .1.3.6.1.4.1.654321.1. You would create an event mapping in the /Heartbeat event class called "Automate6Heartbeat" and set its event class key to that OID. Then within the transform of the event you'd enter the following. evt.component = "Automate 6" evt.timeout = 600 That's all there is to it. If 600 seconds go by without receiving this trap Zenoss will automatically create one of those "heartbeat failure" events with the device set to the device sending the traps, and the component set to "Automate 6". This heartbeat checking is performed within the zenactions daemon which cycles once per minute. You can look at the heartbeat table within the MySQL events database to see what heartbeat Zenoss is currently monitoring. _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
