I believe that a lot of us are in the same boat when it comes to VMware ESXi 4.1 and hardware polling. With the current SW version - you can't.
What I am looking for is a workaround using SW. We already have SW configured as our end-all monitoring solution, so I'd rather not re-invent the wheel and use other tools like HP SIM, etc unless we absolutely need to. We do have that running but it is not used for alerting at this time. We also employ vCenter which of course, does have its own alerting engine (I dislike it immensely).
So - what is everyone else doing with their hardware alerts for ESXi? I have looked into possibly having vCenter send traps into SW that I forward along to our ticketing system. That will at least give me some visibility/flexibility into what is sent for ticket creation. Using vCenter to send traps directly to the ticketing system is a bit ugly. There is really no flexibility in the alerts unless you have them dumped to a text file and parsed, which is just too messy for me.
Naturally since I posted this here I am looking to keep using SW Orion NPM for this as a workaround for direct MIB polling. Is there anyone out there doing this or is everyone using other systems?
Not exactly related, but we currently use the Perl Datastore monitor for one aspect of inner ESX monitoring, but also will be rolling out a Perl Hardware status script (both use the VMware API).
So anything SW can't do, you should be able to do via Perl API (if someone there knows Perl).
(linking it in via APM Script btw...)