Insight Manager 7 SP2 false CMX alerts

[ian green_2]

We are using Insight Manager 7 to monitor several business critical clusters and servers.

Recently we have been recieving a high number of false alerts from the cluster monitor as detailed below.

Event Name: Cluster Monitor Status Change - Node
Event originator: lds-1080
Event Severity: Critical
Event received: 07-Jan-2005, 01:03:31
Event description: Node
Assigned To:
Comment:


Event Details:

CMX TEXT: DISK CMX: at 07-Jan-2005, 01:03:31 GMT device status on node lds-1080 of cluster lds-5001 became critical

Upon investigation there is nothing at fault with the server.

HTTP and SNMP timeouts and retries have been extended with no success.

HTTP Timeout 30 Retries 5
SNMP Timeout 30 Retries 5

Polling
Cluster CMX polling 1 minute
CMX NODE DISK polling 5 minutes
CMX NODE CPU polling 5 minutes
CMX NODE ENV polling 5 minutes
Cluster Status Polling Task Every 10 min
SNMP Status Polling for Servers Every 5 min

Any help would be greatly appreciated

[Ian Miller.]

what software are you running on the VMS systems ?
(tcpip version, agent version, vms version)

[Jan van den Ende]

How did you get Ids-1080 to be accepted as nodename?
Or is it only a local name in the Inside Manager?
Also, the wording "several business critical clusters" raises the question why you have complicated things by not consolidating in ONE cluster?

In other words: Are you sure you are talking about a VMS cluster?

Proost.

Have one on me.

Jan