Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

257381

July 23rd, 2011 02:00

Memory/Battery problem in POWEREDGE 2800

Hi

 

Our company DELL POWEREDGE 2800 server with windows server sbs 2003 suddeenly got rebooted yesterday and stuck in the POST giving the error"

 

Memory/Battery problems were detected.

The adapter has recovered, but cached data was lost.

Press any key to continue. "

 

I three times rebooted the server pressing any key to continue .All the three times os kept loading for long time and rebooted to the same error screen.

 

Atlast I switched off the server and checked for any loose contacts as per the dell suggestion. I removed the HDD 0, HHD 2, HDD3 in channel o one by one and replugged in and started the server. Same system occurred. Since HDD 1 is already failed in RAID 5 , and the place for that HDD is empty, I suspected may be another HDD in that row is failing and so I removed HDD 0 and rebooted the server. the same symptom ( 2003 getting loaded and stuck); switched off the server 'replugged HDD0 and removed HDD2 and restarted the server but the same symptom.

Replugged the HDD 2 and unplugged HDD3 after rebooting the server it is giving error loading operating system.

There are two more hdd's HDD 4 & HDD 5 in another channel 1

 

How to recover from this problem and is our server's  data secured?

 

Please help me in this regard since I am helpless and the server has to be brought im production tomorrow

 

Thanks & Regards

 

S.Swaminathan

9 Legend

 • 

16.3K Posts

July 24th, 2011 12:00

If your LCD panel is blue and you are getting the memory/battery error on every boot, then you need to replace the RAID memory (99% chance, with it rarely being a faulty battery or riser).  The OS will usually not boot in this scenario.

If your LCD panel is amber and scrolling a message about the ROMB battery voltage, then you need to replace the RAID battery (occasionally it is because of the riser card).  The OS will usually boot and run just fine in this scenario.

First, stop messing with the drives and replace the RAID memory (and battery and riser to be 100% sure).  Once it is replaced, because of all the drive shuffling you have done, you will likely need to follow some very careful steps to recover your data and bring your array back online.  When you are ready to do this (again, it is pointless before repairing the memory/battery message), call Dell tech support - it is always free on servers, regardless of warranty status.

July 24th, 2011 13:00

Hi Thank you very much for your clear and precise reply.Atlast through one of my friend ,as you said, I already brought the array online very carefully and booted the server.It gave the error "windows could not load because ntoskrnl.exe could not be found.Please reinstall the above file".

So I booted the server through UBCD WINDOWS LIVE BOOT CD which detected the raid array through plug and play and displayed all the four partitions with volume labels.In our company we have another separate sbs 2003 domain .So I copied the ntoskrnl.exe file in the flash memoryand plugged the usb drive in the server and tried to copy the file but got stuck .I rebooted the server and from that time onwards it is not showing the partitions even in the live boot cd.So we are having no other way except reloading the total server and restoring the full backup one month old.I enquired with local dell asp in Muscat but they are charging too much for just inspecting and moreover they are saying that poweredge parts are not available with them.

 

Thanks & Regards

 

S.Swaminathan

9 Legend

 • 

16.3K Posts

July 24th, 2011 13:00

If your shuffling has not lost the array, then just know this:  with the memory/battery message, you CANNOT repair the OS - in fact, there is no actual problem with the OS.  The hardware problem is causing the failure of the OS to boot.  Your only fix is to replace the memory.  Once you replace the RAID memory, then the OS will boot and function as it should.  If the memory is too expensive, then plan to move the data/services to another server from your backup (or turn off RAID and run off of single SCSI configurations.

5 Practitioner

 • 

274.2K Posts

July 25th, 2011 09:00

Sorry to hear about the issues the server is having. One thing I suggest doing, while you wait for parts to come in, is to disable to write cache on the RAID controller.

Here is some documentation on entering the PERC BIOS and setting it to Read Cache disabled.

<ADMIN NOTE: Broken link has been removed from this post by Dell>

Once this is done, it should make it easier to attempt an OS recover. Because the PERC battery and memory module are not being used.

As mentioned above, it is not the best idea to randomly pull drives out, start rebuilds, etc, while there is an issue with a RAID controller component.

Let us know how this works out.

Thanks

4 Operator

 • 

1.8K Posts

July 25th, 2011 09:00

"So I booted the server through UBCD WINDOWS LIVE BOOT CD which detected the raid array through plug and play and displayed all the four partitions with volume labels."

Agree with flash, since the volumes appeared, unlikely the setup is damaged (unless you continue to rearrange the disks, causing permanent damage). Replace the memory. Possible corruption in the interim my require you to do a repair install, but beats a complete reinstall.

61 Posts

July 26th, 2011 08:00

If you are having difficulties in finding the parts you can find them here. Good Luck! Very common issue.

No Events found!

Top