Thread: [Opsview-users] SNMP trap monitoring..
Brought to you by:
alexatopsview,
tonvoon
From: Rob E. <re...@ck...> - 2007-05-22 16:29:52
|
Hi all.. I have a Barix Barionet device that monitors 8 discrete i/o. It issues a trap for each input and can be polled for the current status. We are a radio station in southern Ontario and I would like to use this device to monitor onair audio silence sensors as well as fire alarm and generator running, etc. To do so, I would use a combination of passive and active snmp monitoring with this device. The initial incoming trap would trigger the Critical alarm and then a recurring active poll would take place verifying the current state of that input. The second trap that is sent when the input returns to normal would reset the alarm state to Normal and then the recurring active poll would continue to verify that the input is indeed off. Any suggestions on how to do this? Or is there a better way to implement this? Thanks Rob |
From: James P. <jam...@al...> - 2007-05-22 21:19:03
|
Hi Rob > I have a Barix Barionet device that monitors 8 discrete i/o. It > issues a trap for each input and can be polled for the current status. Sounds reasonable enough. > We are a radio station in southern Ontario and I would like to use > this device to monitor onair audio silence sensors as well as fire > alarm and generator running, etc. To do so, I would use a > combination of passive and active snmp monitoring with this device. > The initial incoming trap would trigger the Critical alarm and then > a recurring active poll would take place verifying the current > state of that input. The second trap that is sent when the input > returns to normal would reset the alarm state to Normal and then > the recurring active poll would continue to verify that the input > is indeed off. > > It will be necessary to configure two service checks for each input. One to handle trap processing via 'Advanced SNMPtrap' type service check. Second to poll OID of input status via 'SNMP Polling' type service check. It is possible to set dependencies between these service checks so you don't get duplicate alerts. I guess you want to use traps so you get instant notification if an input goes offline. What is the reason for also polling each input? --James |