Insight manager 7 sp2

[Ben Butler]

Is there any way of writing customised pager alerts which give more detail of critical events? if so how do you do it?

[Sunil Jerath]

Hello Ben,
As of yet we do not have those capabilities but never say never. May be soon.

Regards,

[David Claypool]

(Reminder: there is a dedicated IM7 forum at http://forums1.itrc.hp.com/service/forums/categoryhome.do?categoryId=266)

I'm going to hazard a guess that the critical events don't have the detail you are looking for regardless of whether you view them directly, get them as an email, or send them via pager. This is typically the case when the results of a "Device Status Change" event is generated when a poll detects a change (like the device is inaccessible). The result of this poll is only that information; there is nothing responding at the other end to give more detail.

The detailed events come from an SNMP trap being generated with the detail being sent from an agent to IM7. Make sure you have trap destinations set and the proper community strings in all of your agents for proper information exchange.

[Ben Butler]

thanks for the IM7 pointer i will check it out.

unfortunatley SNMP and the traps are set correctly, the detail is still missing as you corrctly assumed.

Do you know whether detailed alerting will possibly be a bolt on feature for SIM?

[David Claypool]

Whether with IM7 or hpSIM the content originates with the management agent. Are you saying that the event as viewed in IM7 has the detail you are looking for and that is not present in the message received by your pager? Are you using an email gateway and the "SMS/Pager" format?

[Ben Butler]

sorry to be confusing, the level of detail is very poor a the CIM console, the pager messages are sent as SMTP standard mails, then we use an application to convert and send via SMS. The question i am really asking i suppose, is there a way to write custom alerts to allow certain events to have specific error, rather than the generic critical device inaccessable/accessable. which i understand is a result of device polling failure. for instance, the generic polling failure would say device polling failed or something which is descriptive like that.

[David Claypool]

Okay, just to be clear, there are 2 kinds of events. I have a system set up to email me with the contents of an event whenever one occurs that affects a specific system. When I pull a drive to simulate a failure, I'm going to get a pretty generic event that is the result of a status polling task:

--------------------------------
Event Name: Device Status Change
Event originator: kasparov
Event Severity: Minor
Event received: 11-Feb-2004, 14:37:13
Event description: The device status has changed due to one of the polling tasks, see below for additional details.
Assigned To:
Comment:


Source of current status change:HP SNMP Agent Polling Previous device status:Normal
-----------------------------------

This event contains everything that IM7 knows; basically it knows that the status went from normal to minor. Nothing else. In fact, the status polling event has an option called "Send Status Change Events" that by default is checked which means rather than just changing the color of the server, it also creates an event to document that the change occured. Many people have decided that the usefulness of this event isn't that great and so have unchecked that box.

Now when IM7 receives an SNMP trap from the device when the drive is pulled, here is the email I receive:

------------------------------
Event Name: Physical Drive Status Change Event originator: kasparov Event Severity: Critical Event received: 11-Feb-2004, 14:37:56

Event description:Physical Drive Status Change. This trap signifies that the agent has detected a change in the status of a Compaq Drive Array physical drive. The variable cpaDaPhyDrvStatus indicates the current physical drive status. User Action: If the physical drive status is failed(3) or predictiveFailure(4), replace the drive.
Assigned To:
Comment:

Physical Drive Status: ok
Controller Slot Number: 0
Port: 2
Drive Bay: 1
Drive Model: COMPAQ BF01864663
Drive Firmware Revision: 3B0B
Drive Serial Number: 3EV0WTSH000073155UKT Drive Failure Code: 0
--------------------------------------

This contains very specific information about the failure and where it occurred.

This example is an email example, but all of the same information is passed through environment variables for the application launch task, so many people use that with a script or other application to pass those parameters.

In short, the event contains all of the details that IM7 knows; if it's from a status poll, all IM7 knows is that the status changed from one state to another. There's no other detail available.

[Ben Butler]

Thankyou for clearing that up for me, It looks as though I am not getting SNMP traps sent to IM7, I will have a look in to it, thanks for pointing me in the right direction.