iLo Version 2 on HP DL360 G5

[Mark Gilchrist]

I've recently upgraded the firmware for my iLo 2. I had been having problems remoting into Servers from the iLOs; and found updating from 1.60 to 1.82 fixed the issue. I assumed since, version 1.60 was deployed when IE 6 (or lower) was standard, the later generation browsers (i.e. IE7 and IE8) included new Active-X components. Therefore, updating the firmware would alleviate the problem ... and it did until I started getting System restarts. The only Windows log entry for my Windows 2003 Servers - immediately before the reboot - "NetFN 0x4, command 0x2D timed out" and "Failed GET SENSOR READING, sensor 11". Both sources are from HPiLO2; and I've included a Pic of the result about 12 minutes after receiving the ILO entrys. See pics attached. This sequense has occured on 3 Servers so far. I'm thinking it's a coincidense but, I'd like to see if anyone can give any insight. I thought as long as the firmware was for a RiLO2, I could keep upgrading and it would not effect the Operating System side of the Server. I'm still of that belief but, my clients are insisting there is a correlation. Please let me know if you can add anything to bolster my argument, or show me how I've errored. Thanks

[Rajeshwari, Hiresave]

What version of the iLO and health drivers do you have ?
If you think it is old, please try updating it to the latest by installing the latest ProLiant support pack - http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=us&prodTypeId=15351&prodSeriesId=1121486&prodNameId=3288144&swEnvOID=1005&swLang=8&mode=2&taskId=135&swItem=MTX-866de15a91a04c84b109f844cc

[Mark Gilchrist]

Hi Rajeshwari ... You are correct in assuming the "health pack" is old. These are turn-key Servers delivered from the vendor. To update health, requires going through a pain-staking effort to convince the Developers these drivers do not cause problems for their application. Since the RiLOE firmware had always been isolated from the OS Side of the box, I felt we need not slow down projects to deploy/test health - while maintaining an emergency "back-door" into the System. I forgot to note that in my initial request for help. We have a very short time-frame to deliver these App Servers for the client. Of course, App folks do not understand the necessity of maintaining the integrity of the drivers - thus, our expedient thought of only updating what they do not see/use (RiLOE). Does this mean there is a correlation between these mystery reboots and old "health drivers"? More information: Here at AT&T, there is an internal group dedicated to such things are updating health/RiLOE on Servers. Unfortunately, these "turn-key" Systems are not a part of this internal groups agenda since, they do not match the corporate standard OS build. Thus, this internal group would not support them. That is where my team came in to help. We are not usually OS Support folks but, filled-in to support this client. With this disclosure, it may be understood we do not normally perform this function - therefore, I am admittedly our ignorance with such tasks. Once I get confirmation you (and possible others) agree these reboots could be caused by updating only the RiLOE firmware, I will setup a test environment and load all health drivers, with an updated version of the RiLOE firmware as described previously.

[Rajeshwari, Hiresave]

I do not know if the reboots are actually caused by this incompatibility of drivers.
I just thought that having the drivers uptodate would rule out any such possibility.

I will try to check up with the iLO2 teams here and in case I get any good information, will get back to you.

[Rajeshwari, Hiresave · Accepted solution]

Thanks for the quick response from the iLO2 team.
The issue seems to be due to incompatible drivers.
Please refer http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&objectID=c01802766&prodTypeId=12169&prodSeriesId=3239482
for more details.

[Mark Gilchrist]

Thanks again Rajeshwari! That is exactly the confirmation I was looking for. However, I am not in a position to upgrade drivers and ROM at this time. Would you/RiLOE 2 team suggest downgrading the RiLOE firmware? It had been at revision level 1.60 without these reboot problems. Knowing we probably can't go back to that revision without re-experiencing problems remoting into these Servers, I'm thinking we could down-grade to (perhaps) 1.61. In the meantime, I can put in a Change Management order, install the latest ROM and Management drivers on a Test box; and get the Development Team to bless going forward. Thanks for your time.

[Mark Gilchrist]

The issue seems to be due to incompatible drivers.
Please refer
http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&
;objectID=c01802766&prodTypeId=12169&prodSeriesId=3239482
for more details.