topic Re: CIM7 Usability in Server Management (Insight Manager 7)

CIM7 Usability

Brent Ingler — Tue, 08 Oct 2002 17:17:39 GMT

In CIM32, you were able to get notifications when a system was no longer accessible, and then another when it was. Does anyone know of a way to do this EXACT SAME THING in CIM7? If so, how? Thanks.

Re: CIM7 Usability

J._1 — Tue, 15 Oct 2002 18:33:43 GMT

First you need to define an event query that lists out the proper events. The specific one's you're looking for are the "Device Status Change" events - (value Critical=not accessible; Normal=available)

Then you need to create a notification task, associate it with this query, with it executed upon any new matches to the query defined above.

J

Re: CIM7 Usability

Robert Huber — Mon, 21 Oct 2002 19:22:38 GMT

What type of query will produce the "Device Status Change" that you wrote about. I have looked at the queries and I am not sure where to begin. Your assistance is greatly appreciated.

Re: CIM7 Usability

Brent Ingler — Mon, 21 Oct 2002 19:30:42 GMT

Under the queries, then Device, there's the device by status queries. Those will give you the device status changed notification, I believe.

Here's the problem, though, that no one seems to understand (here or at the old Compaq board). The resulting message from this query and notification gives ZERO indication as to whether a system is critical because it's inaccessible, or because it's actually having a major hardware failure. CIM32 does this, and it astounds me that Compaq/HP would be so short-sighted as to not include this in CIM7.

Additionally, and this is still a problem no matter how many people ignore it, the RIBLOE cards do not report complete deatails in CIM7 either. For example, when the external power is disconnected, CIM32 actually shows that. CIM7 doesn't, nor does it have the apparent capacity to do so.

Re: CIM7 Usability

J._1 — Tue, 22 Oct 2002 11:32:24 GMT

First, for Robert:
You'll need to define an "Event Query" using an "event type" condition.

Second, for Brent:
Yes, the event descriptions in CIM could use some work. The Critical event without any real detail ("Device Status Change") is only triggered when the device is unreachable or comes back online. If a particular HW item was to blame for it, you have to hope that the associated Trap made it out before the system completely died - and this is usually the case in my experience, as most of our configurations are fault tolerant. Either way, the device status change is sufficient, IMHO.

As for the RIB support, I think I've seen traps when the power's been disconnected. Maybe that was with the older CIM XE...