ProLiant Servers (ML,DL,SL)
1752794 Members
5904 Online
108789 Solutions
New Discussion юеВ

Re: Proliant DL380 G5 POST fail after processor upgrade

 
CambridgeMart
Occasional Contributor

Proliant DL380 G5 POST fail after processor upgrade

I have a DL380 G5 which was equipped with XEON E5160 CPUs; I've replaced the CPUs with X5460s (the mainboard is 436526-001). It booted initially and loaded Windows, but after a short time it shut down with all 8 DIMM LEDs amber and a long (5s on, 1s off) beep from the speaker. Now, when I power it up, it doesn't appear to POST and just lights all the DIMM LEDs and does the long beep. I've tried swapping CPUs, running with 1 CPU only (in the RH socket) and swapped the VRMs around, but no change. I did try looking at the ILO diagnostics, but didn't really see anything usful there.

I should add that this machine has been running fine up until the processor swap.

What further diagnostics should I try?

4 REPLIES 4
parnassus
Honored Contributor

Re: Proliant DL380 G5 POST fail after processor upgrade

Have you considered to check the BIOS System ROM version?

  1. With the BIOS System ROM version 2008.08.03 (A) - August, 26th 2008 - release was added support for the latest Dual-Core Intel Xeon 5200-series and Quad-Core Intel Xeon 5400-series processors...so you should first be sure that you have an updated BIOS System ROM at least (actually the latst is 2015.08.16 (A) - October, 1st 2015 - for the HP ProLiant DL380 G5).
  2. On HP ProLiant DL380 G5 QuickSpecs there are a sibilling statement about future supported CPUs: "Future upgradeable to Xeon┬о Series 5400/5300 Quad-Core processors (Please contact your regional sales support to determine quad core upgradeability. Some of the earlier shipping models may not support this upgrade). 5100 Dual-Core models are not upgradeable to the 5200 Dual-Core processors.".
  3. I'm not sure but also the DIMMs frequency should be taken into consideration: are them clearly PC2-5300 Fully Buffered DIMMs (DDR2-667) ECC capable of 667 Mhz in system with a FSB of 1333 MHz as per QuickSpecs? note that "Memory DIMMs in systems with the Front Side Bus (FSB) running at 1333 MHz will run at 667 MHz and in systems with a FSB running at 1066 MHz or 667 MHz the memory DIMMs will run at 533 MHz." the Intel Xeon 5110 and 5120 have a FSB of 1066 MHz instead the Intel Xeon X5400 has a FSB of 1333 MHz so DIMMs should be able to reach 667 MHz.

Probably the best thing you can do AFTER checking your BIOS System ROM version is to check the motherboard acoustic and LEDs signalling patterns.


I'm not an HPE Employee
Kudos and Accepted Solution banner
CambridgeMart
Occasional Contributor

Re: Proliant DL380 G5 POST fail after processor upgrade

Thank you. The BIOS version is 2015.08.16 

The mainboard is the later type that supports the quad core processors, up to the X5470. The DIMMS fitted are a mixture of 2GB, 667MHz, PC2-5300F-5, DDR2, dual-rank x4, 1.80V, registered, fully-buffered with ECC, dual in-line memory and 4GB, 667MHz, PC2-5300F-5, DDR2, dual-rank x4, 1.80V, registered, fully-buffered with ECC, dual in-line memory.

The machine did boot into Windows directly after the upgrade, but failed after 10 or 15 minutes, so I don't believe it's a compatibility problem.

The audible error is a long beep, repeated. The LED indications are that all 8 DIMM LEDs are amber, along with a red health status LED.

parnassus
Honored Contributor

Re: Proliant DL380 G5 POST fail after processor upgrade

I haven't checked documentation...but...couldn't be a heating dissipation problem? QC Xeon dissipates a lot more than a DC Xeon...was the new QC Xeon X5460 installed properly?


I'm not an HPE Employee
Kudos and Accepted Solution banner
CambridgeMart
Occasional Contributor

Re: Proliant DL380 G5 POST fail after processor upgrade

I took the machine back to a minimum configuration of 1 processor (one of the X5460s), 2 DIMMS and no expansion cards; in this state, it booted correctly. I gradually built it up through 4, 6 then 8 DIMMS with no problems, then added back the expansion cards, and finally added the 2nd CPU, Once everything was reinstated, it still booted and has now been running for several hours.

I can only guess that something wasn't correctly seated previously and removing then refitting everything has fixed it.