Unsolved
This post is more than 5 years old
11 Posts
0
6801
PowerEdge 840 Flashing Amber Light
I have an old PE 840 that I got from work. When I first received it, the power light was flashing amber. It was still under warranty, and a Dell tech came out to replace the motherboard and/or processor. He connected to the DRAC 4, and cleared the logs and the server fired right up. I went to upgrade the processor from a Xeon 3210 to a 3230, and now the dreaded flashing amber light is back. For whatever reason I cannot connect to the DRAC any longer. I have tried across the network, and with a cross-over adapter.
The fans all work, the DRAC and network ports both show activity. I have removed all cards and RAM DIMMs and still no POST. When connected to power, the light is initially blue, but quickly goes to the flashing amber. Any further suggestions?
DELL-Chris H
Moderator
Moderator
•
8.8K Posts
0
July 13th, 2016 07:00
Five4FistiCuffs,
What I would suggest starting with is dropping the serfver to a minimum to post configuration. That is removing everything from the server, but the following devices;
Once you have removed the devices then locate the NVRAM CLR jumper on the motherboard. (Shown below) With the jumper we will clear the NVRam. The NVRAM will be cleared when the jumper is closed (use the jumper from the password bridge). To properly clear; AC power must be applied (not necessarily turned on) to the system for about 10 seconds while the jumper is closed.
Pins 1-3 shorted clears NVRAM
After that then power the server and let me know what the 4 diagnostic lights are showing, as well as if there is anything on the screen.
Let me know what you see.
five4fisticuffs
11 Posts
0
July 15th, 2016 11:00
I was able to clear the NVRam. The four diagnostic codes are a green C and amber A, B and D. This is the same behavior with both the original CPU and the upgrade CPU. All of the pins in the socket look okay.
DELL-Chris H
Moderator
Moderator
•
8.8K Posts
0
July 18th, 2016 13:00
5,
The led code of Green c, with the others remaining amber, indicates an error of "CPU configuration activity is in progress or a CPU failure was detected". With the error occurring on both CPU's, I am curious if the processors slot is the issue itself. Just to clarify, can' you power down and insert the original CPU, then before powering up you need to redo the NVRam procedure. Did you do that before, or did you do that only with the replacement CPU? Give that a try, as well as if you have a known good processor we can test with it. If all that fails then likely the issue will require replacing the motherboard.
Let me know what you see.
five4fisticuffs
11 Posts
1
August 11th, 2016 09:00
I had to replace the motherboard, but the server is up and running again.