[BlueOnyx:07566] Re: Can someone offer any insight on what is causing this crash on a DL360 G4?

[Darrell D. Mobley]

> -----Original Message-----
> From: blueonyx-bounces at mail.blueonyx.it [mailto:blueonyx-
> bounces at mail.blueonyx.it] On Behalf Of Support
> Sent: Thursday, May 12, 2011 10:43 AM
> To: 'BlueOnyx General Mailing List'
> Subject: [BlueOnyx:07283] Can someone offer any insight on what is causing
> this crash on a DL360 G4?
> 
> Hello,
> 
> I have a DL360 G4 with two drives in Raid 1 and I am experiencing crashes
> that look like this http://i.imgur.com/ooaS5.jpg and wondered if someone
> could point me in the right direction to resolve the issue?

Going back in time here.  A number of people have been having problems with
Compaq and HP DL-series servers with Smart Array controllers locking up
under heavy disk load with the newer kernels released with RHEL5.6/CentOS5.6
and the BQ/BX derivatives.  It was discussed in length on various Red Hat
Bugzilla reports about the issue and several people were able to reproduce
the problem and come up with a solution: upgrading the Smart Array
controller firmware to the latest versions.

I have successfully completed updating my DL380 G3 firmware over iLO using
the Firmware Maintenance CD 8.60 located at
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lan
g=en&cc=us&swLang=8&mode=2&taskId=135&swItem=MTX-124beffbed3842cea0b9b3f962.
I used this version for two reasons: 1) it said it was for a DL380 G3 of my
model number and 2) because it had Smart Array 5i firmware 2.76 on it.

I downloaded the file and extracted the iso and performed a md5 checksum on
it. When I was satisfied I had an intact file, I connected to a local DL380
G3 via iLO and launched an iLO virtual media window, selected it to set up a
virtual CD using my local ISO image.  I also opened an iLO remote console to
perform the updated.

I rebooted the server, and it came up under the Firmware Maintenance CD so I
went through the steps, choosing to install all recommended updates.  It
took about 25 minutes on the local network and I watched it on the local
monitor.  Because one of the updates was an iLO firmware update, I lost the
remote console at the end, but not to fear as I could see that the updates
were finished and waiting to reboot on the local monitor.  So I closed and
reloaded the iLO browser session, everything came up and I rebooted.

All was fine with that, so I repeated the same operation on a server two
time zones away this afternoon.  Both have 2.76 on them now, and I then
changed to the latest kernel 2.6.18-238.12.1.el5PAE.  We'll see if this
works tonight, as it used to puke when RaqBackup and other early AM scripts
were busily at work.

So this can be done remotely, just make sure you checksum your download, as
my first attempt failed the checksum test.