HP SIM on ESX hosts, any way to filter alerts?

[JohnGibson]

Hi I have a number of ESX Hosts, I have the SIM agent installed.

I have a number of Major alerts, but for things that are not really major.

i.e. I have dual Port Gb NICs but dont have all ports populated (loop back connectors dont work) so they generate alerts.

I also have Dual Port HBAs, but only one port populateda t the moment, so they alert.

i dont really want to have to go into the BIOs and disable all of the unused ports, I may need them someday, and its a lot of work to reconfigure 120 DL585's.

So has anyone found a way to either stop the Host alerting for specific ports or get the SIM server to mask them?

i doubt its possible, but thought I would ask.

Cheers,

John

[VirtualNoitall]

Hello,

You need to jump on your SIM management server and edit a couple of things. You can go to options | events | snmp trap settings and/or edit your automatic event handling by creating a custom collection, put your ESX hosts in there, and disregard certain alerts for them. This will allow you to set it up once and manage it by managing membership in your collections and what alerts you care or don;t care about.

Hope that helps!

[JohnGibson]

Do you have any further details, I looked at the SIM server, and these options, and couldnt spot a way to mask the Health alerts.

John

[VirtualNoitall]

Hello,

1) You have the option to configure what you alert on here:

Options | Events | Automatic Event Handling

2) You can edit the severity of incoming alerts here:

Options | Events | SNMP Trap Settings

You can, at a high level, edit the severity of traps that you alert on here:

3)Options | Events | Event Filter Settings

I think what you are asking for is #2 above. It will take a little time to navigate through and find the alert but it is doable.

Hope that helps!

[vmproteau]

The HP SIM suggestions may work but, I don't want to eliminate all NIC or HBA events just the non-used interfaces.

John- I've had the same issues.

NICs: I see loop back cables aren't working for you but, in my case it works like a charm. Server showed healthy after a few minutes.

HBA: In my case I use 2 dual port HBAs in the server and only 1 port from each card is active. The only thing I found to eliminate the alert is a short fiber cable connected between the 2 unused ports. That may not be feasible for you.

[vmproteau]

By the way, I just installed an 4-port GB NIC in an HP DL380. 2 ports aren't currently being used and I just made 2 loop back cables that eliminated the alert. Server shows healthy.

Just a basic loop back

pin1-pin3

pin2-pin6

[VirtualNoitall]

"The HP SIM suggestions may work but, I don't want to eliminate all NIC or HBA events just the non-used interfaces. "

then filter them.

[NCCVM]

What version of SIM agents are you using? also what brand of NIC?

Also, are you able to alert on a per port basis? It sounds like you are.

Thanks