insight manager 5.5 sending too many user threshold falling alarms

[Iain Whyte]

I have insight manager 5.5 set up to alarm and send emails on disk and cpu utilization. The past couple of weeks it's been intermittently sending bunches of falling alarms. For no apparent reason it will send about 30 of them all at once - all for the same server. The messages all have the same time stamp but are not all caused by the same threshold ... I'll get messages for each of the three thresholds I have set up (2 for each processor and 1 for disk space). I should mention that I have updated the management agents on the servers to version 6.0 ...

[Miriam Haber]

Is there a "Set Disk Threshold" task running on a schedule that is generating these alerts?

Also just curious... why using CIM 5.5? Is there a reason you have not upgraded to CIM 7 (though you have upgraded the agents)?

[Iain Whyte]

Doesn't appear to be any tasks set that might cause this ... cim 5.5 doesn't have that ability and doesn't appear to be tasks set up on my cim 7 server ... I've upgraded the agents on our win2k dc's only and have started "playing" with cim 7 with regards to those machines ... however, I've got a bunch of other machines (about 70 servers) that we want to monitor etc ... most of these machines are older and have insight manager agents ranging from version 4.5 and up ... I don't have approval to update the management agents on these machines and am unlikely to get that type of approval ... There's an old software developers saying that fits in here ... God created the world in 7 days but he didn't have an installed base to deal with ...

[Iain Whyte]

Also, I've been playing with disk threshold alarms on cim 7 and I get a rising alarm when I create a big file but I don't seem to get a falling alarm when I delete it ... Any idea why that would be?

[Scott A. Boyd]

I get the same thing. Both rising and falling. I had never received them before I installed an instance of CIM7. Now I get them in in CIM32 as well. In addition, I have a couple of servers out there that give me literally hundreds of pages regarding services stopping and starting. When I check those server's event logs the services never stopped. Very frustrating. I had to stop paging for those servers because of the shear number of pages I would receive.

[Ted Jerzewski]

We have a similar type problem. Mainly we get bogus threshold alerts. I have not seen any evidence, in NT event log, that the alerts are coming from the server(s) in question. Do you definitely see the bogus alerts as a buildup of old alerts coming from server? Protocol analyzer indicates that the alert is coming from server(s), but I see nothing in NT event log suggesting alerts came from server(s). We have 250 NT 4.0 servers with CIM 5.2 and 5.2 Compaq Insight agents.

[Ted Jerzewski]

We are in switched network environment. Are all of you in switched network environment?

[stironzo]

Do you have a computer or a washing machine? Hey what are you waiting for to thrash it? Come on!