Hardware message monitoring

[mark.leimbach]

Hi, Is anyone using OPSBCPII or Address HWS to monitor hardware messages for outages? Specifically we had a service element call home to IBM and I would like to notify staff of such a problem. I looked through the doc but don't see any specific examples. Any help would be appreciated. 
Thanks,
Mark

------------------------------
Mark Leimbach
Information Tech Architect
Ericsson
------------------------------

[Marcel van Ek]

Hi Mark, yes we do.

Assuming you've got the OPSBCPii etc setup, you can add an )API HWSHWMSG rule.
Look for events in OPSLOG (BROWSEAPI parm needs to be set to YES) for events like these:
HWSHWMSG HARDWARE MESSAGE ISSUED CPC IBMxxxx.nnnnn Remote support call failed.
HWSHWMSG HARDWARE MESSAGE ISSUED CPC IBMxxxx.nnnnn Hardware problem. Problem # 310
etc.

I'm not aware of any IBM doc wrt all possible events to look for.

Depending on your HMC setup you'll receive events from one (local) or more (remote) CPC's
If you've setup OPSBCPII on multiple lpars you'll need to handle the fact that you'll get duplicate alerts from those multiple lpars.
(setup some sort of master/slave order for BCPii connected lpars to decide which is leading)

regards,

Marcel van Ek

------------------------------
Automated Operations
Atos
Netherlands

[mark.leimbach]

Thanks Marcel. I have OPSBCPII set up and will try the suggested API rule. Is there any way to test an API rule?

Thanks,
Mark


Mark Leimbach
Solution Architect
Information Technology
[e-mail]<mailto:[e-mail]>

[Hennie Hermans]

Hi Mark,
With OPSVIEW 2.1 (Editors, AOF Edit) you can also test API rules... 
Enable the rule in Test and run the test... 

Best regards
Hennie Hermans

------------------------------
Principal Support Engineer
Broadcom Software
------------------------------

[Marcel van Ek]

Mark, you should also have parm APIACTIVE set to ON to detect API events.

------------------------------
Automated Operations
Atos
Netherlands
------------------------------

[David Covemaeker]

Hi Mark,

We recently start monitoring the HW messages using the Broadcom sample APIHWSV in CCLXRULS.

The sample is covering all HW* API events, providing you with a well formatted multi-line WTO for each specific HW event type.

As we are only interested in a WTO for some HW event types and an alert only for event type HWSHWMSG, we added a mechanism to only enable the WTO and/or alerting for the HW event types of our choice.

Note that event type HWSHWMSG is covering many different types of (error) messages. We started with monitoring them all and keeping track of the different HWSHWMSG error messages to update the rule (enable / disable process on HWSHWMSG specific message text).
It's not ideal, but as Marcel, I was not able to find any IBM documentation on this.

Regards,
David