Proliant DL 380 G3 reports "mayor" error in Insight Manager

[Roy Harris_1]

The following problem occurred after the Firmware upgrade of the RILOE II board.
After the installation of Firmware 1.12, the following error was reported in the event log of Windows 2000.

Compaq Remote Insight Agent: The Compaq Remote Insight Board has detected a controller interface error.
[SNMP TRAP: 9006 in CPQSM2.MIB]

This error has been noted in the log every 15 to 30 minutes, and is still active until now.

Remote connecting to the RILOE card using the web interface is working perfectly, the iLO log shows no error of any kind, the systems status is okay.

When we check up with insight manager 7 the program reports a mayor error concerning the remote insight auxillery power, it is connected but the status is "Failed".

i am a little bit of a loss here, because the server cannot just be rebooted. i did try a reset of the Riloe board but that did not work.

I hope that this question is complete enough for a sollution.

[JohnWRuffo]

Have you tried to re-flash the RILOE II ?

[Jon Ward]

Consider upgrading to the latest Proliant Support Pack as well. It is very possible that the new firmware has rendered the old drivers obsolete.

Also check the System Management Homepage through http://localhost:2301 for any problems with the power supply, fans, etc. on the server itself, which may affect the auxiliary power source for the Remote Insight Lights-Out II.

[Roy Harris_1]

Thank you for the respons on my question, i checked the System management Homepage of the server and i saw something that i did not see before, the Remote Insight Board accorrding too the SMH never got the Firmware upgrade, the ROM version is still standing on 1.01 the interface status is: Not responding, i seemed to have missed this with Insight Manager.

However after flashing the ROM for a second time is still have the same respons, the status is Failed, and the ROM is still at 1.01. The strange thing is that when i check trough a Web interface with the RILOE II card it reports nothing and the Firmware is at 1.12.

There is a similar server with the same config and the same card, this one does have the 1.12 firmware installed and is working properly, they both have the same Compaq Agents Running, version 5.50.0.0

Is maybe the RILOE II card Broken ??

[Prashant (I am Back)]

Hi,

Possible solution: Again update firmware on RILO and run SCU then check it.

Regards,
Prashant S.

[Prashant (I am Back)]

Hi,

Please do try this

1. Download the Remote Insight Lights-Out Edition Diags Mode Recovery Diskette (SP19719) and create the bootable diskette.
2. Boot the server with from the diskette. Notice the beeps at the end.
1 long - RIB is in Factory Diagnostics Mode
2 long - RIB is in not in Factory Diagnostics Mode
3 long - RIB was reset to normal mode
3. Boot into F8 safe mode.
Go to Control Panel, Compaq Management Agents for Servers and disable the Remote Insight Information Agent (move it to Inactive Agents).

Or Install the latest mangment agents.

Regards,
Prashant S.

[JohnWRuffo]

Roy:

Make sure you are running the latest version of the RILOE II board Driver (version 3.2.1.0 dated June 18, 2004) before updating the bios:
http://h18023.www1.hp.com/support/files/server/us/download/20225.html

After that, maybe you can step directly up to bios version 1.13 instead of 1.12 and then 1.13

Here is the latest PSP if you want to use it instead:
http://h18000.www1.hp.com/support/files/server/us/download/19613.html

I hope this helps! -john

[Remon van Gijn]

Maybe an other angel, is the RILO II not having problems with the onboard iLO of the 380G3?

[Roy Harris_1]

It seems that i have misplaced the name, it is a G2 Server not a G3.

So it is in fact a RiLOE II card.
Thank you all for the response, i have updated the Compaq agents to the new version, the problem is now solved. i was away for a couple of days and it that time the server received a reboot, voila all the problems dissappeared.

CIM detected nothing wrong with the server and the logs stopped generating errors. A simple reboot was all that was needed in the end.