Factory Recoveries are much easier to perform then Server Recoveries thing is as you know you need to copy your data back to the server. This is the case for all WHSv1 server system be it from a manufacturer like HP or one you may have built yourself.

One thing WHS may opt to do is run a CHKDSK on your drives and you know how long those can take. Your server does have a lot to do with the updates, with rebuilding the storage pool and without a debug board connected you can't exactly tell if the server is performing a CHKDSK or if its gotten hung on a process and requires a reboot. These factors are what makes using a Headless Server frustrating at times to work with and why its so important to have utilities installed to monitor various server conditions.

WHS is also very sensitive to drive issues and with hard drives we can monitor their physical condition but there are times when a hard drive will fail without prior warning and this is the most frustrating thing about hard drives. I have had new hard drives fail or act up within the first month of operation and yet I have hard drives that are old which never seen to have issues. This is where a docking station comes in very handy to allow you to check your hard drives and to allow you to read your data off the drives.

One thing you may want to try is to boot the server with only the system drive installed to see if it will boot up and yes you will see tons of warnings and alerts about missing files, corruptions and services not running just ignore them. Then shut the server down and insert one drive and reboot to see if the server will boot with that drive if it does not then shut it down and back out that drive and insert the next one and reboot. What your looking for is a hard drive which causes the server not to boot and it can be one or more but typically it will be only one drive causing the issue. Take that drive and connect it to a client and run a CHKDSK on it and then try rebooting the server with that drive installed. This is always a PITA but its the quickest way I know of to rule out a hard drive that's causing an issue with WHS Booting. The worse thing is when the server boots and then reboots because of a hard drive.

Anyway its not all about HP but knowing the sensitivities of the Operating System and once you get past this you will be much better equipped to addressing server issues when they occur. Learning by fire is never fun for sure.

_________________
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

Hello again chaps.

All was going swimmingly.... until...

So, factory reset to new WD Black drive. Successful. Left me with task of hooking up my 4 original hard drives to a client pc and copying over the files from the hidden ("DE?") folder. All going well, checking server after each disk. Three drives in and set up client to copy over 5Gb HD movie files. Got home to red light flashing on server and client stopped half way through transfer saying lost connection to server.

Paperclip in the hole - think 2 blue flashes followed by 3 red flashes.

Tried to do system restore, but having problems in this process now - "Failed Step 4 of 4."

Starting to wonder whether anything wrong with the network card/hardware. And to be honest, growing soooo tired of HP Mediasmart.

All server files appear to be on one pool drive installed (2Gb WD Green). I'm wondering whether to at least get all the files off the 4 'old' drives onto one disk. I could do this on a client - hooking up the pool drive on a client and copying over the 'old' drives to this (by hooking them up to the client too). This would leave me free to mess about with the system installation on the system drive.

Q being this: if I do get all the files onto one HD, would there be any way to transfer them onto the server after this? Would it be possible via a usb dock to the server, or would I have to go the client>server route?

I'm really disappointed by the HP MediaSmart. Did lots of research before buying, and all good reviews before. But I guess this was before it stood the test of time... I know, I know this is old OS now, but still....

That error code is listed as BIOS Recovery Source not Present.

The Blinking Health Light being Red indicates a Power On Test or BIOS has failed.

What is not clear is what the server was doing prior to this condition state. You indicate you were able to perform a Factory Reset which puts the server back to the original factory conditions. After which there would of been a ton of HP and MS Updates to download install as well as your having to configure the server and the server would of had to go through a series of reboots. Then of course it appears you have added a hard drive and began to move your data to the server and then found it in the condition you report. Then for some reason you attempted another server recovery.

Is there anything I missed in this?

Have you also experienced any unknown reasons for the server shutting down?

When the server was functional did you happen to look at the Health Status of the server hardware, PSU Power Levels, Fan Speeds, and CPU Temp and if so did the server report any issues?

If the server can now be booted can you make a Remote Desktop Connection and review the Event Logs to see what issues the server may have logged to help identify potential issues with the server hardware.

I am more concerned with the condition of the server hardware and so if the server were here I would be looking closely at the hardware to make sure it was in good working order before moving forward otherwise you may encounter additional issues if there is a hardware issue left in an uncorrected state.

_________________
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.

I'd completed all the updates etc - the server was clean as a whistle and running 100%. Both new system and pool drive checked with smart status - all fine.

The crash happened one day whilst out at work - left data being transferred from a client to server. Producing error code stated. The server doesn't boot - cycles through flashing blue for about 2 minutes then a brief red flash + repeat (looks that it's continuously trying to reboot).

The reason I attempted a server recovery was because it wouldn't boot!! Of course, as it won't boot, I cannot RDP or console.

Oh, and checked the sys and pool drive on client machine for SMART status - all fine

Any ideas?

A couple of troubleshooting things worth trying:

(1) Back out all drives in the server and turn it on to see if the server will boot into its recovery mode then shut the server off.

If the server does not boot into Recovery mode and displays the same Blink Code listed then you have a serious hardware failure & there is no need to proceed with the other items below.

(2) With the server off seat only the system drive and turn the server on.

What we are looking for here is if the server can boot off the system drive. Of course the server will flag a bunch of errors about missing drives, file corruption, corrupt backup data base and so forth this is to be expected because of the missing storage pool drive(s). Shut the server down.

(3) Seat one Storage Pool Drive (I believe you only have one but if you have more then one we want to do this one at a time) & Turn the server on.

If the server boots and appears to be fine with exception to the expected errors then shut the server down and insert the next storage pool drive and reboot the system. Continue to do this until you have all the storage pool drives installed or until the problem you initially reported occurs again.

What I am looking for is a potential issue with the PSU which will often show up when the PSU is placed under a load that's why I wanted you to start with no drives connected to the server as this will keep the load on the PSU to a minimum but also to see if the server would at least boot into its recovery mode which would rule out a BIOS issue. Following that I want you to add a drive to start placing a load on the PSU to help determine if the PSU is the cause. Of course its easier to determine if you had a debug board attached to the server then you could see what exactly is going on and you could even enter the BIOS and check the hardware health status.

So lets see what you come up with.

_________________
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

How will I know whether "server will boot into recovery mode"?? What will happen - what am I looking for?

thanks

EDIT: Nemmind. Rebooted without any drives in server. Flashed purple for a bit then settled on solid red. Error code = 1 blue followed by 4 red (blinking). Looking here:

http://h10025.www1.hp.com/ewfrf/wc/docu ... ct=3969718

this shows "Recovery mode failed."

So... hardware fault??

After I had got it up and running fine, I upgrade the processor to:
Intel Core 2 Duo E8500
Researched around before this and believe that it's compatible and couldn't see any specific setup requirements.
As above - seemed to run fine for a while.

Any suggestions. If I had the money - I'd throw this horrible box out of the window, but it's looking around £400 for a new server

ok first your server will be in recovery mode when the health light blinks Violet. If you see this then we can rule out the BIOS issue.

The reason you saw the Recovery Failure is because the server did not connect to a client to begin the recovery process so that's ok because all we were doing was verifying if there was an issue with the BIOS.

So the next thing to do is to see if you can boot the server with just the system drive to see if it will boot. If you have a problem where the server boots and reboots then I would say your PSU needs to be replaced. If the servers health light blinks blue and does not finish booting then you will need to pull the System Drive and perform a CHKDSK on it to correct any issues with the file system. You may also want to look at the System Drive to make sure it has 2 partitions names SYS and DATA.

_________________
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

As above, comp, system disk does not boot.

I will run chkdsk on it....

[UPDATE] chkdsk run - no errors!

I'm reformatting the system disk now to try factory restore on freshly formatted disk.

That's unfortunate.

If your Factory Rest is successful I would be observing the servers health status looking for issues. I would of course configure the server, download and install all available HP & MS Updates, install storage pool drives and move data but would only do so when I could observe the server. I would be watching the health status screens of the server looking for issues.

What is not really known is how your server got to the point you found it so it should warrant quite a bit of observation for the new few days.

Do you happen to have a UPS connected to the server to protect it in the event of a power failure where you would have software installed to instruct the server to shut itself down properly and cleanly after a certain period of time. What this will do is prevent hard drive corruption issues when power is lost if the hard drives are being written to. I understand the CHKDSK passed but something is not right and I would still be concerned with the PSU, FANS and Temps both for the CPU and Hard Drives.

_________________
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

Did all this, comp. Closely monitored CPU due to upgrage - ran a little hot @59 - but think it was OK



Highly, highly unlikely I had a power cut + can't keep buying more and more things to get this thing working as it should!

Also watched fans and HD temps closely - nothing concerning. As I say, crash happened when at work + transferring large files from client to server.

Doing a full format on the system disk. Will try factory reset tomorrow. Now have all my files on one HD. To go back to an earlier Q - is there any way to get these onto the server without using a client machine? If I plug it in USB dock (plugged into server) + have clean pool drive in the server (added) could I RDP in and would I be able to move from 'independent' usb hard drive to the "Data" folder on server?

You can connect a drive to your server but do not add it as a backup drive just leave it connected. Make a Remote Desktop connection to the server and you should be able to access the contents of the drive just like on a client. From there you can copy the data from the drives to the appropriate server shares or other locations of your choice.

Once you have moved all your data then you can if you like add the drive to your storage pool but make certain you have not missed anything because if you do happen to add the drive to a storage pool it will get formatted. If you have an available bay in your server you can connect the drive to that bay rather then using a docking station.

I understand where your coming from in terms of spending funds on the server but things like a Docking Station and UPS can be used for other applications. I use the docking station to assist me with servicing computers as well as servers. I also use Port Multipliers now as docking stations but also have USB Devices to connect both 2.5" and 3.5" IDE and SATA devices to a client for hard drive service work but then again I work on servers and computers for others year round. The UPS is an essential protection device that I spend all day talking about and are like having an insurance policy to keeping your devices protected not just from power failures but from surges and sags and with AVR or Automatic Voltage Regulation they keep operating power supplied to your devices within an acceptable range.

Anyway its better to let someone know how to protect their devices so that they can make an educated decision. Keep in mind your time has a value too and so when you figure out how much time you have spent on this then you can better understand how things actually pay off.

_________________
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 comp.

Finally got my server/HD accepting a factory reset (again!) - had to do full format of the previous installation (took 12 hours!). Wow - the EX495 really is a pedantic and temperamental piece of kit isn't it!

So - gonna be all day installing updates and setting it all up again. Dunno how you guys stay sane working with this!!

On the UPS - seems like a good idea. Doubt a power outage or surge caused these difficulties, but good idea non-the less. However, being new to these I have a question. If I was out at work and there was a power outage - how would my server know to shut itself down whilst running on battery power? Would the UPS send the server a signal or something?

I like the look of these:

http://www.ebay.co.uk/itm/APC-BK350ei-s ... 5d45c6f72d

I notice that it has some connector ports on the back (data, "wall outlet") - would these communicate with the server?

Thanks for your ongoing help

How we stay sane is basically knowing what to look for and knowing how to troubleshoot issues when they arise.

As for your UPS being able to shut the server down, well its done with software and a communications cable between the server and the UPS. Not all APC units have this option especially on smaller units. Also on the older units they may have a DB9 connector and would require a special cable and serial connection to communicate properly with the UPS. Its important to make sure the one you purchase has a cable that will work well with your server many of the newer APC Cables look like a Network cable on one end and USB on the other.

APC's software is called Power Chute and they have a version compatible with WHSv1. Basically you would make and RDP connection to the server to install and then configure the software. When configuring the software you will see an option for how long you would like the UPS to maintain power to your devices in which case you best to figure out about 20% of battery capacity. If your UPS can power your server for 20 mins then you would set it for 4 minutes of run time before instructing the server to shut down if power is not restored within 4 minutes.

The batteries are typically good for 3-5 years.

While your working on your server try to keep a close eye on the servers health indicators looking for issues mostly when its under a load. I do not recommend at this time that you make any attempt to configure the media functions of the server because they have been known to cause issues depending on how you configure them and the last thing you need right now is to have the server go about collecting and converting videos. That can be configured later once you know your server is running well.

_________________
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

OK.. all installed and set up OK. Now transferring data via USB dock into the server onto a pooled drive. This feels better way to do it rather then over a network and seems to be going OK. In retrospect, I think all problems center around large files (and possibly streaming)? Got some blu-ray movies and had lots of problems with streaming and last crach few days ago was when transferring video folder.

Keeping close eye on stats. Weirdly, my new WD Black (non-AF) drive is now showing errors. You'll remember that did new install onto it a week ago>had the crash>full format on it last night + today successful installation. Some pics:





UPS: this does look good + looking into it. Was gonna use grid junction, but appears all it's forums and sites have disappeared. Managed to find version 1.6, but can't find a UPS compatibility list anywhere! Thanks for the info. Most of the ones I'm looking at (£30-40 mark) have a serial port on them - some have ethernet too. I'm wondering whether you know of compatible units? I've seen serial to usb converter leads. Here's the one I'm looking at at the moment:

http://www.ebay.co.uk/itm/APC-UPS-BackU ... 53fb1cdb9d

Any advice appreciated!

If that drive is new then send it back to WD because you should not be experiencing bad sectors so soon. If its an older drive I would recommend you get another drive especially if you see the number of bad sectors grow.

I am not familiar with the UPS's sold outside the USA with regards to the model numbers but the serial port is a hard one to deal with because it may require a special cable. Also the USB-Serial Cables are not always compatible with everything so you could end up hitting a brick wall with a UPS with a Serial Port or more specifically a DB9 Port. Look for a UPS that has a USB Cable and APC does sell them but its a funny looking cable that looks like a network connector on one end which connects to the UPS and USB on the other which connects to the computer or server.

_________________
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