PowerEdge 1950 will not power on
We put a Dell PowerEdge 1950 into storage for 13 months and when I tried to use it, it would not power on. The 'AC available' LEDs on both PSUs are lit, a LED on the motherboard is lit and there are 2 LEDs on the DRAC card - one is lit and the other is blinking. Comparing these LEDs with a working 1950, these would appear to be normal.
If I take the cover off, the intrusion switch triggers the usual I1910 warning message on the LCD display and removing items such as the riser cards also provokes the correct error messages. The CMOS battery was dud and I have replaced this and I have also reseated all connectors, CPUs and memory DIMMs and reset the NVRAM using the jumper on the motherboard. The DRAC works as normal but if I attempt to power the system on using SMCLP via a ssh connection with the 'start /system1' command, it says the system has successfully been powered on when in fact it hasn't
If it press and hold the front ID button for a few seconds, it comes up with the error code 'FB'. What does this mean? An identical server of the same age and also stored for 13 months powered up with no problems so I've really no idea what could be wrong - these servers are the ones with two 3.5" SAS disks and SAS RAID controller.
Any suggestions
Andy.
If I take the cover off, the intrusion switch triggers the usual I1910 warning message on the LCD display and removing items such as the riser cards also provokes the correct error messages. The CMOS battery was dud and I have replaced this and I have also reseated all connectors, CPUs and memory DIMMs and reset the NVRAM using the jumper on the motherboard. The DRAC works as normal but if I attempt to power the system on using SMCLP via a ssh connection with the 'start /system1' command, it says the system has successfully been powered on when in fact it hasn't
If it press and hold the front ID button for a few seconds, it comes up with the error code 'FB'. What does this mean? An identical server of the same age and also stored for 13 months powered up with no problems so I've really no idea what could be wrong - these servers are the ones with two 3.5" SAS disks and SAS RAID controller.
Any suggestions
Andy.
PowerEdgeTech
This apparent no-video issue is MOST often resolved by clearing the NVRAM and/or reseating the processors, but if you have tried both of those, I'm not sure what else it could be other than a bad board, although it's unusual when in storage for parts to actually go bad. Was it stored in a "friendly" environment, free from temperature extremes, radiation, and humidity? Did you try with just a single processor installed, swapping them to make sure both are good? Have you tried removing the DRAC altogether? You might also call Dell for additional ideas, although I think we've covered just about everything ... I don't know how things work in your part of the world, but in the States/Canada, support is always free, regardless of warranty status.
Can you retry reseating the SAS power cable to the backplane? According to Dell's documentation that's a possible cause for an FB error.
ASKER
The server was stored in a dry attic in a colleague's house, wrapped in bubble-wrap. I have not tried it with just one CPU fitted, or with the CPUs swapped so I will definitely try this when I'm on-site tomorrow. I did try removing the DRAC but it made no difference and when I refitted it and logged in again through the web interface, the text in the tabs along the top had all changed to 'undefined' - they had lost lost their original legends such as 'log', 'configuration', etc so I guess the DRAC has somehow been corrupted now and needs a firmware reload.
Also, I tried reseating the SAS backplane cable at both ends and also removed & refitted the disks. But I have not attempted to remove & refit the SAS controller card - it simply won't budge from the sideplane connector and I don't want to break anything. Could this be where the problem is, although I doubt a missing SAS controller would stop the BIOS starting up?
Thanks for your suggestions, I'll try removing one CPU & swapping them tomorrow, as well has having another go at taking the SAS card out.
Andy
Also, I tried reseating the SAS backplane cable at both ends and also removed & refitted the disks. But I have not attempted to remove & refit the SAS controller card - it simply won't budge from the sideplane connector and I don't want to break anything. Could this be where the problem is, although I doubt a missing SAS controller would stop the BIOS starting up?
Thanks for your suggestions, I'll try removing one CPU & swapping them tomorrow, as well has having another go at taking the SAS card out.
Andy
ASKER
Problem fixed!!
I came into my office this morning to find the room hot and the server roaring away on my desk. It had somehow powered itself on and the LCD error code was E1211, meaning check the RAID battery. I turned the screen on and found the server was stuck in the BIOS with an error message saying the battery was faulty or missing and giving me the option to use the controller in an insecure mode with no battery backup. Pressing F1 booted it normally and the battery error message has not returned since.
I guess leaving the power feeds connected to the server has allowed the battery to charge but what I did not know was that the RAID controller battery has to be functioning before the server will power on.This is the opposite behaviour to HP SmartArray cards and I can find nothing in the Dell docs about this either.
The only remaining problem is the DRAC web interface - everything works through the SMCLP console interface but now all the tabs in the web interface read 'undefined'. Looks like there is a firmware problem on the DRAC but at the server itself is working now.
Many thanks for all the suggestions,
Andy
I came into my office this morning to find the room hot and the server roaring away on my desk. It had somehow powered itself on and the LCD error code was E1211, meaning check the RAID battery. I turned the screen on and found the server was stuck in the BIOS with an error message saying the battery was faulty or missing and giving me the option to use the controller in an insecure mode with no battery backup. Pressing F1 booted it normally and the battery error message has not returned since.
I guess leaving the power feeds connected to the server has allowed the battery to charge but what I did not know was that the RAID controller battery has to be functioning before the server will power on.This is the opposite behaviour to HP SmartArray cards and I can find nothing in the Dell docs about this either.
The only remaining problem is the DRAC web interface - everything works through the SMCLP console interface but now all the tabs in the web interface read 'undefined'. Looks like there is a firmware problem on the DRAC but at the server itself is working now.
Many thanks for all the suggestions,
Andy
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Yes, removing & then replacing the DRAC card has fixed the corrupted web display, thank you.
This server is now fully operational and OMSA & DRAC both report the status of the ROMB (RAID battery) as being good.
Thanks everyone for their help,
Andy
This server is now fully operational and OMSA & DRAC both report the status of the ROMB (RAID battery) as being good.
Thanks everyone for their help,
Andy
ASKER
Although the server apparently fixed itself overnight when left with power applied for 2 days, this tip regarding getting the DRAC web interface working properly was a separate issue which worked for me.
Please do add this question to the Knowledge Base and Dell ought to do the same with their's.
Please do add this question to the Knowledge Base and Dell ought to do the same with their's.