I've inherited a "working" Power Edge 2850 with dual Xeon processors that just went off warranty a few months ago. After I racked it and powered up, it gave the error above on the LCD front panel display. I opened the box, reseated the RAM, removed the heatsink from Processor 1 and unclamped and clamped it back in place (I did not remove the CPU chip entirely, however). Same problem. Otherwise, the machine seems to go through the boot process normally - fans on high speed then back off, hard drives power up and initialize, but no display of BIOS - just LCD error and blinking orange i light and green power button.
Assuming I can't simply repair this problem, I figure I have two options: replace the Xeon 80546K with a Dell refurb ($450) - or remove PROC1 and run as a single processor. Any recommendations?
I've tried a few more things. First, I'd read a lot about how you must reseat all possible connections and components if the machine has been physically moved. So, I reseated both CPUs, RAM, fans, hard drives, power supplies, and cables. Same problem. Then I swapped PROC 1 and PROC 2 between sockets, then it reported PROC 2 IERR. Wow, I thought I was getting somewhere! So I removed the CPU from socket 2 completely. But then the PROC 1 IERR came back. OK, I'm getting a bit desperate now! Tried removing the CMOS battery for a minute, then replacing the same battery. Same old PROC 1 IERR. Soooo, I don't think this is a processor problem at all.
The server is 3 years and 2 months old so it's just off warranty! I'm pretty sure that this server has been running in a data center its entire life. But, is it possible that the CMOS battery needs to be replaced? Any other ideas?
The fact that the processor error occurred even before the POST made this difficult to troubleshoot. I didn't know the BIOS or firmware level and I couldn't run any diagnostics. I'd just have to assume that the admin at the datacenter had kept them current - probably not a good assumption! But the machine had been functional literally a day before in their datacenter.
BUT... I have finally fixed this problem! I replaced the processor chips into their original slots. Then I reseated RAM, riser board, RAID key, RAID RAM, power supplies, fans, etc. But this time, using manuals, I discovered that the whole front drive assembly plugs into the motherboard, so I unplugged and reseated that assembly, as well as all the drives into the assembly (again). I also pulled the motherboard CMOS battery and checked its voltage - OK. I powered it up and like magic, it went thru POST and booted up to the Dell CD I'd inserted prior. Wow! Since I'd removed the CMOS battery, I'd lost all the BIOS settings including the RAID array configuration, but I eventually figured that all out.
So it's working now. But since the problem originally occurred in my cold (~60 F) server room, I'd pulled the server and moved it into my warmer office, wondering if it was a temperature problem. So now I have to shutdown the server and take it back into the server room. Normally, I'd let equipment acclimate to the ambient temperature, but in this case, I think I'll quickly rack it, connect it, and power up. Wish me luck!
I believe this error code is a "red herring" and is a result of another problem and not a CPU problem. You will find that if you remove the DRAC card the error will disappear, indicating a faulty DRAC card. If you don't need or use the DRAC card then it dosen't matter otherwise you will have to purchase another DRAC card.
Hope this helps,
We just had 2 2850's experience the same E07F0 Proc Error. We spent an hour or two will dell support stepping though troubleshooting steps. DRAC was never once mentioned as the possible issue to the error. We pulled both CPU's, swapped out system board still was getting the error. Out of desperation, I turned to the web again and found this post. We pulled the DRAC and voilà , server is back on-line...a true red-herrign if ever encountered one. Thanks for the post KERNELBYTE,,,
The puzzling question in my mind is why/how 2 of my servers suffered this issue almost simultaneously..
It's not really a 'red herring' ... the CPU error is not specific and can indicate an input problem with any device that communicates with it ... along with expansion cards, storage controllers, memory, etc., also checking the DRAC should be part of any routine troubleshooting for this type of issue, especially by Dell Support, who should have known better.