topic Re: SIM 5.1 Alerting in Server Management - Systems Insight Manager

SIM 5.1 Alerting

Ron Tack — Tue, 16 Dec 2008 14:20:04 GMT

Not sure if you can answer my question or lead me in the right direction.

We are looking for ways to “reduce our after hours call outs for failed drives on raid sets with hot spares”.

Is it possible to have SIM 5.1 recognize when it’s alerting on a failed drive to verify there is a hot spare and alert differently than it would for failed drive without a hot spare. IE Critical verses Minor?

Re: SIM 5.1 Alerting

marsh_1 — Tue, 16 Dec 2008 16:32:09 GMT

Ron,

AFAIK this is not possible with HP SIM , it would have no visibility of the state of a raid /storage array hot spares.
the only event correlation it does is cancelling unreachable events for systems that become available again.
using scripting with SSSU with eva might be able to do something. you would probably be best to pump it into something more tailorable like ovo.

FWIW

Re: SIM 5.1 Alerting

Rob Buxton — Tue, 16 Dec 2008 23:32:01 GMT

I tend to agree with Mark on this one.
There is some ability of the agents to generate alerts based on pre-failure SMART information.
But the agents will not typically know if an array has a hot spare or not.

You'd need to send the events through to a script using the custom command event handler. You might be able to pick up any array rebuild messages and combine things before passing on the alert. But I don't think it would be very easy, or particularly reliable.
Or you could have the script reference a table of arrays that you would need to create. That table could contain say the arrays with hot disks. You'd e-mail based on the results.

I do something similar with Server alerting to our call management service. There's a list of servers to exclude. The script reads the list, if it matches the server then it doesn't update the call Management Service.