I have had the EX485 up and running for about 3 years with 2.5 version. It has a 750GB WD Black Caviar drive for the system drive. A Hitachi 1TB drive in slot 4 and a 2TB Samsung Drive in Slot 3. About a month ago we had a series of Brown outs during a big storm. The next morning I noticed that I didn't have a connection to my WHS. The lights were blinking red and blue. I powered it down and the lights blinked blue for quite a while and then shutdown. It kept doing this for hours. So I pulled out 1 drive at a time and it did the same thing. I then took the system drive and placed into another computer and it seemed to work. I also tried the 2TB and 1TB and they were both seen by the system. I ran harddrive checks on the system drive and the 2TB and they passed. I didn't check the 1TB (until later).

I tried it with no drives installed and it would eventually show all lights on, no blinking. However a few times during bootup with no drives it would reboot a few times before stabilizing. Completely stumped I ordered a Debug cable.

Once I installed the debug cable I tried booting up with only the system drive installed and watched the screen. It would go through the bios screens and then go black for 10-15 seconds, eventually I would get a bar across the screen stating something like, windows is installing files, sorry I don't recall the exact wording but it looked like it was preparing to install or boot windows. But at about 75% of the bar the system would power off and reboot. It would not finish with the drive installed. About every 4 or 5 reboots it would not make it to the bar and state that windows is missing files to properly boot. Based on the warning it seems that the Windows install is corrupt.

So just to see what happens I installed the 2TB and 1TB data drives. With both installed it would reboot immediately after the bios screen. It did this every time. I pulled the 1TB drive out and then it was stepping past the bios screen to the corrupt windows or 75% of "windows install" before rebooting.

I then put the 1TB drive in the other computer and ran the Hitachi Harddrive Diag software and it immediately gave a S.M.A.R.T. failure warning recommending removing any valuable data and to discontinue use of the drive.

So 1TB dead after the Brown outs.

I then tried a SYSTEM RESTORE with the 750GB drive and the 2TB drive. It would not get past the 75% on the bar. I eventually left only the 2TB drive plugged in and let the SYSTEM RESTORE complete the 100% and boot to the point of getting the "Listener Created" point.

I then hot-plugged in the 750GB drive and started the SYSTEM RESTORE software. It found my HPSERVER and began the restore. This actually took 3 times doing this before it completed. However I was able to finish the restore but a bunch of warnings appeared on the EX485 screen. I was not able to access the server from the Console from my main computer. The only thing I could do at this time was reach the Shared Folders. Unfortunately it corrupted some of my data. It appeared to show the file but if I tried opening them it was blank or reported as invalid. It looked like I was getting ghost links to nothing.

At this point I gave up one the drives not wanting to lose anymore data.

So I pulled a 250GB drive out of my main computer to try a Factory Reset on only this drive. It too would not get past the initial windows bar giving a system type error. So I booted into Restore mode without a drive. I then hot plugged the drive. The Factory Reset completed 1st try and seemed to work. I let it run on this drive for about a week and it seemed to be stable. I then formatted the 750GB original system drive and added it as a pool drive. But with the 750GB drive now installed the system would reboot or lockup every couple of hours.
So I had another 200GB drive that I was not using, pulled out the 750GB drive. I installed the 200gb, added it to the pool. I then copied over 250GB of data. It was stable for about 2 days but then the system started rebooting again.

This time the screen was showing that the 200GB drive had been corrupted. It was running CHKDSK finding a ton of errors on that drive. I did get it to boot one time by skipping the chkdsk. I tried to remove the 200GB drive from the pool but it failed and the system shutdown again.

That is were I currently am stuck.

1TB drive Smart dead.
750GB drive passes WD diags but does not play nicely with EX485, even as a pool drive.
2TB drive passes diags and seems to not have an issue. It is the only drive that has my data though so I am not risking it in the server for now.
250GB drive worked alone for a week. It may still be working but the 200GB drive won't let it boot.
200GB drive worked for a few days but then corrupted.

So can bad components on the EX485 be corrupting these drives?
Is there anything else I can check?

Thanks,

Hi giga welcome to the forums,

WOW there is alot of info to digest there. The first order of business is to protect your data and its good to see you are doing just that. The thing about a power failure is that if data was being written to a hard drive it can become corupted and prevent the server from booting. This is why its important to protect a server or any computer with a Battery Backup set up with the necessary software to keep the device running for a period of time and if power is not restored to have the device shut down properly. Just want to point this out as most think they are used just to keep things running but after a period of time the batteries drain and power is lost just the same so software is needed to perform the necessary shut down protection.

What I would do is take a drive which the diagnostics say its in good working order and install it in a computer and delete the partitions on the drive. This will give your drive a clean slate to start with and then I would install it in the bottom bay, disconnect any USB or eSATA connected devices with exception to a keyboard or mouse which you may have connected since you have a debug board.

Turn the server on and it will boot into its recovery mode and then run the server recovery software and perform a FACTORY RECOVERY. I have had system drives that did not play well for a recovery until I removed the partitions on the drive. If all goes well then you can see what if any difference that made. One thing to note is if you repurpose a System Drive for a Storage Pool Drive its important to remove the two partitions SYS and DATA before repurposing the drive or it will corrupt your server causing you another headache and will require you to perform a FACTORY RECOVERY. I am uncertain if this is what caused your 2nd round of issues but its important to remove the partitions from any drive previously used as a SYSTEM DRIVE.

The other things to look when your server is up and running are the health indicators and are they working properly and most important are the fans working right because your EX485 is a passive cooling system its vital the fans function properly to keep the server cool or if it heats up it will just shut down for its own protection. When I am uncertain of the fans I always place a 4" fan in front of the server with the door open to blow air accross the motherboard and PSU.

Hopefully this make sense to you and helps you. I will try to read through your original message better tomorrow when I have more time to make certain I have not over looked anything.

_________________
Exploring the possibilities!

Migrated from WHS2011 to WS2012E: HIGHLANDER

~ Norco 4220 Enclosure
~ Gigabyte GA-990FXA-UD5
~ AMD Phenom II X4 995 3.2 GHz
~ 8 GB Corsair Vengeance DDR3 1600 (PC3 12800)
~ 3 Supermicro AOC-SASLP-MV8

Thanks for the reply and sorry for the amount of information in the first entry. I kept remembering as I typed.

I do understand the importance of a UPS and had one, but my UPS battery had gone out about a year ago and I never replaced it. So that is my fault, I am now motivated to replace the UPS though.

The 750GB drive that was initially a system drive, that I converted to a pool drive, was reformatted using the WD utility removing the partitions and starting fresh. Even after this procedure the server would reboot every couple of hours with the 750GB drive installed. The odd thing is that the 750gb drive passed diags so really should work.

I know the fans are are spinning at start up, however I don't know if they stop spinning after a matter of time. I will check that.

It could just be that I have a series of bad drives, and I definitely had the 1TB drive fail diags and has now been sent out for RMA.
It is just that with the struggles I am having I was thinking that the EX485 was causing the drives to fail.

Is there another test I can run on the WD750gb drive besides their software? I know at work running diags on a machine doesn't always find the suddle problems with equipment.

The 1TB drive is currently in RMA processing so I will have a new drive to play with in the next couple of weeks. But I will try to identify one of the other 2 small drives as healthy for testing purposes.

Thanks Again,
Steve

You could install the ADD-IN HomeServerSmart which will provide you with a wealth of information about your drives more so than the WD Lifeguard Diagnostics Utility. You can download HomeServerSmart at the link below:

http://www.dojonorthsoftware.net/Freebi ... SMART.aspx

After installing the ADD-IN you can slide in a drive and not add it to the storage pool and it will give you a ton of information about the current health state of your hard drives.

I will tell ya honestly that WHS can be sensitive to the condition of hard drives and when WHS acts up its no joy and with the costs of hard drives still high its difficult to justify replacing them.

_________________
Exploring the possibilities!

Migrated from WHS2011 to WS2012E: HIGHLANDER

~ Norco 4220 Enclosure
~ Gigabyte GA-990FXA-UD5
~ AMD Phenom II X4 995 3.2 GHz
~ 8 GB Corsair Vengeance DDR3 1600 (PC3 12800)
~ 3 Supermicro AOC-SASLP-MV8

I will give that Add-in a try. I rescanned the 750GB drive with chkdsk and Seatools and they didn't find a problem.
The Hitachi 1TB drive was confirmed as bad by Hitachi and they shipped out an RMA replacement on Friday, should get it this week.

But in the mean time I pulled out the 200GB drive that worked for 2 days as a pool drive and it would only be seen occasionally by the other computer. Once recognized I ran chkdsk and it found a bunch of errors. Even after chkdsk it was not always recognized, so it is now going in the garbage.

Unfortunately the bad drive corrupted the system install on the 250GB drive so I will need to run a Factory Reset. Once I realized that, I ran chkdsk and Seatools scans on the 250GB drive and it passed as healthy.

At this point I am not sure which drive I want to use as the System drive. I want to at least try out the Add-in to check the 750GB.

Does the add-in work on system drives?

Also I ran the Microsoft Memory Diagnostic from a bootable CD and the memory in the EX485 passed after running through Advanced Testing around 3 full passes. I have had PC's with bad memory that caused odd issues causing random lock ups and reboots, so I figured I'd give it a try. At least I can check that off the list of possible hardware faults.

I can not get the EX485 to do the Server Recovery. I was going to do the Factory Reset on the 250GB drive in order to try the Smart ADD-in on the drives. However it spontaneously reboots about 75% of the "Windows is loading files .." screen.

I first tried it with the drive in the bottom shelf so I removed all drives. No drives installed it still only gets to 75% and reboots. I have let it run for about an hour no change. The fans are spinning.
So I am now thinking that the EX485 is really bad and maybe these spontaneous reboots are corrupting the drives.

I am now Stuck.

It did get me thinking though. Can the SMI USB portion be corrupted? And is there a way to put the contents of the SMI onto a usb stick or bootable CD to try to boot into recovery mode? But if the SMI is corrupted how would I be able to get an uncorrupted version?

My 1TB Hitachi arrived back from RMA.

But I still can not get to the Listener step to re-install the software. It just makes it to about 75% of "Windows is loading files.." and reboots.

Is there another option to install the WHS software?

I have an external USB DVD drive. Is the recovery cd bootable and directly installable?

Any help is greatly appreciated.

I finally got a drive to work with assistance. So I let it run on this drive for about a month. It seemed stable so I tried to install a second drive. It immediately crashed the system and it rebooted. I plugging in the debug board and watched it. It would either crash and reboot during the boot process or get to the point that the server is starting up and as soon as it tries to use the second disk it crashes and reboots.

At this point it has to be a hardware issue with the EX485. It could be the power supply but since the output is different then standard ATX I have not easy way of testing it.

Would it be worth it to try a different power supply or just put that toward something like a N40L?

Super frustrated.

After converting an older 300W ATX power supply to the MSS pinout, I tried it on the ex485 and it works. I then added another drive to the pool and it worked without causing the server to auto-reboot. So far it has been running for about an hour and backing up my main computer.

I am so hoping that the power supply was my only problem. If it stays alive for a few days I will be picking up the correct size power supply, although this big power supply hanging out of the case is probably better for cooling, lol.

Update May 2nd 2012:

After running for a few weeks and adding a second pool drive, the system has been running perfectly off of the converted ATX power supply.
Will be ordering a proper fitting power supply shortly since I am confident that was my main problem.

I will say that I could have never even started to guess the issue without getting the debug cable.

Thanks for the help.