ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

DL 585 G5 - hard locks / crashes.

 
JimTaylor
Advisor

DL 585 G5 - hard locks / crashes.

Hi all.

We've got 4 identically specced DL 585 G5's - each has 4x AMD 8354's (2.2GHz QC) and 64GB RAM (32x 2GB DDR6-667 ECC), a dual port HP NIC, a quad port HP NIC and a pair of QLogic HBA's. They're all running VMware ESX 3.5.0 U1.

One has taken to recently hard locking - no life at the console (black screen), the HBA's are all lit up at the back, and if I pop the lid off the server the status code on the LCD display is "01". If I ILO in to the server, I see no video (displays No Video in the Integrated Remote Console) and stuff such as fan speed and processor temperature are shown as 'Not Available'. The only way out of it is to power off and power on again.

The problem is we can't consistently reproduce the fault. I was sure it was around doing any kind of I/O but yesterday spent a portion of the day repeatedly deleting and creating VM's on the internal storage and was not unable to make it crash. However earlier today we rebooted the server, and it locked up on scanning the available LUN's during the VMware boot up sequence.

Does anyone have any suggestions on where to go here? I'm reluctant to open a call with HP as the fault is not easily reproducable - and with nothing concrete to go on, I doubt they'll have any suggestions on what to do!

I'm certain it's not the software as we've not made any changes to VMware for a few months - and as NOTHING gets logged prior to the crashes happening in any of the log files. It just.. happens.

I'm leaning towards one of the HBA's being at fault..