Let me begin by screaming:  ARRRRGGGHHHH

 

Just moved to a new house.  Packaged up my sever VERY carefully and hooked it up at the new location.  Unraid booted up just fine but the system stated that Disk 8 of 11 was missing.  After performing a shutdown, I opened the case and checked all the connections of power and sata cables, and everything was intact.  I rebooted, during which there apparently was a power-surge (my stereo in the other room clicked off).

 

Now my system will no boot.

 

When I turn on the system, I see the P5B VM DO logo screen asking if I want to enter bios etc.  Then a black screen with the following message:

 

"J Micron Technology Corp.  PCI-E to SATA II/IDE Raid control Bios 1. 06. 63

Copyright 2005 J Micron Tech               www. Jmicron.com

        Detect drives done     no any drive found"

 

What does this mean?   ("no any drive found" isn't a typo by the way)

 

Here is my equipment:

Intel Celeron 440 Conroe-L 2.0GHz LGA 775 35W Single-Core

CORSAIR XMS2 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400)

Power and cooling 750 Watt PS

SUPERMICRO AOC-SAT2-MV8 64-bit PCI-X133MHz SATA Controller Card

P5B VM DO

 

- I'm thinking it must be the Motherboard? 

-I've tried unplugging the supermicro controller card, and makes no difference.  I've tried both my USB Unraid memory sticks to no avail? 

 

Hopefully my harddrives are not fried!  But I have no way to check at the moment that I can think of?  I can live with buying a new MB, but if all the drives failed . . . . Shudder

 

Thanks in advance for your help!!

 

 

 

Sound's like the BIOS got hosed.  Power problems can do that.

 

Boot to the BIOS ,and reset it to factory/safe defaults.

 

The change the appropriate values to what you need.

 

You may need to do the same on the J Micron controller, if it has a user-changable BIOS.

Well, you were right!!  A littel bios work and now the system boots just fine.  Unfortunately disc 8 still is showing up as missing, plus the parity says that its not valid - It definitely was valid recently before my move.  I haven't copied any data to drive 8 in a long time - I added disc 10 recently and the system has been filling up that drive.  The dirve 8 hasn't been showing errors or acting funny lately, so it seems odd that it would have failed after my move.

 

Obviously, I have checked and replaced the sata cable and exchanged the power cable multiple times and this does not affect anything.

 

I've added a screenshot of my configuration, as well as syslog. 

 

Anyone have any ideas why the drive shows as missing? 

 

Thanks for your help!!

 

Edit.  Unfortunately, it says "upload folder is full"  I'll try to link to it from another host. 

 

 

http://www.mediafire.com/?sharekey=a1d3e3cc969a519a6e7203eb87368129e04e75f6e8ebb871

 

Sorry for the outside links, but the lime-tech site doesn't seem to have any room??

Once a drive "disappears", it will never come back on its own.  It could be that you DID have a cabling issue and fixed the underlying cause.  Getting humpty dumpty put back together again is likely if you haven't hit the restore button or written directly to the disks while not in the array.

 

Before doing anything, make a copy of the "config" folder on your USB stick.  This will enable you to get back to the current configuration in case something gets hosed.  It is a smart idea to backup the config directory as soon as you have a problem (right before you have the problem is better, but that's a little hard to arrange )

 

Post a syslog.  (Follow the troubleshooting link in my sig for instructions).  Also post a screenshot of your "Main" page.

 

Make sure the array is stopped (hopefully the array has been stopped through this ordeal).  Go the the devices tab.  Go to an unused disk slot.  Drop down the list of available drives for that slot, and see if you see some drives there.  DO NOT ASSIGN THEM TO THE SLOT (if you do, immediately unassign it).  If you can take a screenshot with the dropdown list of available drives being displayed, please post that as well.  Otherwise, just note the number of drives and whatever info it tells you about them.

 

Be patient and chances of recovery are good.  Don't start the array and, whatever you do, DON'T PRESS THE RESTORE BUTTON.  If you have already, please let me know.

bjp999.  Thanks.  I posted a screenshot of the "main" page and a syslog above.  Are you able to see them?  I had to link to offsite hosts because the site is full?

 

I have backed up the config folder as you suggested.  The array is stopped.  Under the devices tab, when I go to an unused disk slot, there are NO availbe drives listed what-so-ever.  Is this bad? 

 

I definitely haven't pressed the Restore button 

bjp999.  Thanks.  I posted a screenshot of the "main" page and a syslog above.  Are you able to see them?  I had to link to offsite hosts because the site is full?

 

I have backed up the config folder as you suggested.  The array is stopped.  Under the devices tab, when I go to an unused disk slot, there are NO availbe drives listed what-so-ever.  Is this bad? 

 

I definitely haven't pressed the Restore button 

 

Okay, this is a good start.

 

It appears that unRAID thinks there is a problem with Parity (confirm the icon is red), and disk 8 is missing.

 

It appears that unRAID is finding the parity disk, but not finding md8 - a drive whose serial number ends in P73.

 

Your first task will be to try to get that missing drive to come online.  Try replacing the data cable, plug in a differeent power cable, connect the drive to a different port - do what you can to get that drive to come online.  When you do it will either show on the main page, or when you go to the devices page and look at available drives, it will show there.

 

I don't see anything alarming in the syslog, although will defer to RobJ who has a lot of experience reading these. 

 

If you can get the P73 driive to get recognized, that would be best.  Otherwise you will need to get another disk >=500G and <=1T.

 

I have tried 3 different SATA and Power cables and it hasn't come online.  I will try it on a different port on the Supermicro card. 

 

Let's say that I can never get the drive online and it is officially "dead"

 

Do I have a chance that the parity drive could still rebuild that drive (eventhough it is red and says its unvalid) since I'm sure that I haven't written any data to disk 8 since my parity drive was last vaid? 

 

Thanks again.

I have tried 3 different SATA and Power cables and it hasn't come online.  I will try it on a different port on the Supermicro card. 

 

Let's say that I can never get the drive online and it is officially "dead"

 

Do I have a chance that the parity drive could still rebuild that drive (eventhough it is red and says its unvalid) since I'm sure that I haven't written any data to disk 8 since my parity drive was last vaid?   

 

Thanks again.

 

Yes, a very good chance IMO.

 

There is still the lingering question as to why the parity drive dropped.  Do you know?

 

I would run a smartctl report on that drive and see if anything unusual pops up.

 

If the disk in drive 8 really is dead dead, and parity is also dead or dieing, that would be bad for recoverability.  But if either or both are healthy, recovery is likely.

I'm pretty sure I know why the parity drive dropped off.  Right before I moved I had shut down the machine and when I restarted I dumped some data onto Disc 10.  I pretty sure that I interupted a parity check at that time is all.  The parity drive had been working just fine prior to that. 

 

When I get home, I'll check another port, and if that fails, I have a back-up 500gb drive at home I can substitute.  I  just need to know how to rebuild a new drive if the parity drive is red. 

 

I can help with that.  Just see if you can either get the old drive online, or the new one recognized.

 

UPDATE:  Look at this post and the remainder of the thread to see how another user in a similar situation recovered from this.  He was the guiney pig.  You would use invalidslot of 8 instead of 3.  But if you can get the current drive in slot 8 to become responsive, you could just follow the trust my parity procedure and not have to rebuild the data drive.

 

 

I located a copy of one of your older syslogs, to compare with, didn't see anything unusual except the loss of drive 'P73', replacement of the parity drive, and gigabit networking now.  P73 used to be connected to ata10, the 4th connector on the 8-port SATA controller.  Since it is not seeing anything there at all, I would move it to either one of the other empty ports on that card, or better yet, to one of the 2 onboard JMB ports, which are 2 of your fastest ports.  I suppose they could be eSATA ports though?  Would need an adapter to use them.

 

As Brian said, if the drive re-appears on any of the dropdown's on the Devices tab, then you can use the Trust My Array procedure to quickly return to all green.  If you cannot get that drive to work, then the 'Brian variation' of the Trust My Array procedure should rebuild your Disk 8 on that other 500GB drive.

One problem, you are still using v4.2.1, and the invalid slot command parameter only appeared around v4.3.2, so I don't think the Trust My Array procedure is available to you, unless you upgrade to v4.3.3 or v4.4.2.  It should be an easy upgrade, just replace bzroot and bzimage (and memtest for v4.4.2), and reboot.

Well, Things have taken a turn for the worse.

 

After being unable to get Disc 8 to be recognized after switching ports on my card, I inserted my spare 500gb drive into an empty slot.  I then booted the system and sure enough it was recognized under devices, and will allow me to assign the drive if I had chosen. 

 

The problem is now Disc 5 shows as missing!!!! :'(

 

In otherwords both the Disc 5 and 8 are now showing up as missing and remain so.  When I plugged disc 5 into an empty slot on the card, it still is not available to be assigned under the "device" tab.

 

My drives are dropping off like flies!!! What could be happening? I'm now quite paranoid obviously. 

 

Is it possible it is all a configuration problem with the UnRAID software, dropping those discs and not allowing me re-assign them under the device tab? 

 

Should I cut my losses and try to do a fresh configuration with my spare UnRAID usb card?  Would the drives likely be recognized then? 

 

I'm truly at a loss, but am so grateful for your help.  Thanks again

Don't panic.

 

It could be bad cables or it could be that the controller card is going bad.  I would experiment plugging disk 5 and the old disk 8 into motherboard ports and see if they get recognized.

 

You have your backup config folder which may be needed to recover from these configuration changes.

 

BTW, are you seeing evidence of thesee disks NOT being recognized during the boot process.  If so, you might use that to confirm before allowing unRAID to boot. (But with the config backup, this really doesn't matter).

 

I had a similar problem not long ago. My problems turned out to be a bad cable AND a loose port on the motherboard that needed a locking SATA cable to function reliably.  (Two problems at once are always fun).  After several unsuccessful attempts to get things working, I took the system apart, disconnected ALL the drives, and hooked each drive to the motherboard with a known good cable to convince myself that the drives were good.  I then went on from there.

 

Another USB stick is not the answer.

 

Be patient and don't panic, recovery is still very likely!

Is it possible it is all a configuration problem with the UnRAID software, dropping those discs and not allowing me re-assign them under the device tab?

 

No, it is the Linux kernel that locates, identifies, and configures the drives.  All unRAID can use is what the kernel says is there.  So far, there is no evidence of any problems with the USB flash drive or the unRAID software.

 

Please make another syslog available.  From the last one, I think Disk 5 is sdc?  If so, it was connected to the third SATA port on the motherboard, so that *is* strange.  What port did you connect the new Disk 8 to?  Do drives show up when connected to the JMB ports?

Disc 5 was DEFINITELY on the Motherboard, not the Supermicro SATA card.  I tried two different cables and a different power connector.  Both times it was not identifiable on the "device" tab.

 

the "New" disc 8 was plugged into an open slot on the Supermicro card.  It IS identifiable on the "device" tab.  But I didn't assign it. 

 

I will definitely post a new syslog when I get home from work.  I'm on 24 hour call tonight, so its possible it won't be until saturday. 

 

Correct me if I'm wrong, but if 2 drives are missing, one from the SATA card, and one from the Motherboard ports, isn't that really strange? 

 

 

 

Ok here's my syslog. 

 

http://www.mediafire.com/?sharekey=a1d3e3cc969a519a6e7203eb87368129e04e75f6e8ebb871

 

I'll be working on some of the questions you have asked, i'll post answers as I learn

 

Thanks

 

 

RobJ is much better at reading these syslogs, so I will not get in the middle of that.

 

My question is ...

 

If you go into BIOS at bootup, can you see disk5?

 

Ok new update.  I narrowed down the Disc 5 issue to an anboard SATA port on the motherboard.  I couldn't get it to work in either two sata ports that I checked on the MB, the interesting thing is a second sata port seemed to fail while I was switching drives around to try combos.

 

I am also now very convinced that Disc 8 the original disc that failed cannot be discovered.  I have tried multiple configurations of "good" sata cables, "good" ports (on both the MB and the IDE card), and "good" power supply connections.  None have allowed disc 8 "FP73" to be discovered or assigned.

 

So now I have a spare 500gb drive that can be assigned,

 

Should I proceed to assign that to disc 8, and see if I can rebuild it from the parity drive?  I understand I would need to update the UnRaid version somehow?

 

here is my most recent syslog:

 

http://www.mediafire.com/?sharekey=a1d3e3cc969a519a6e7203eb87368129e04e75f6e8ebb871

 

 

Ok, I've sucessfully updated to UnRaid 4.4

 

So now only Disc 8 is missing, but the parity drive says its unvaild. 

 

I believe I am ready to assign my spare 500gb drive to "disc 8" and try to rebuild it from the "Unvalid" parity drive - I'm not sure how to do this. 

 

Does this seem like the best next step?

Any help through this last step would be greatly appreciated!  Thanks

I think you are ready for Brian's earlier advice, found here.  Just make a copy of the flash drive first, or at least super.dat and disk.cfg from the config folder of the flash.  Follow the instructions very carefully.

 

Since you are assigning the new drive as Disk 8, you will use this version of the invalidslot command (instead of 99):

mdcmd set invalidslot 8

 

If in doubt at any time, show us a pic of the screen, and relate what you have done so far, and someone should be able to advise.  Be aware that this IS a very new procedure though.

Thanks Rob, and Brian.

 

I'm going to give this a try.  I've got nothing to loose, and even if it doesn't work, I should just be back where I started. 

 

Wish me luck, and I'll be sure to let you know how it goes!

 

Cheers!

Good luck. 

 

Just make sure you follow the instructions carefully.

 

No typos allowed on the set invalidslot command!!!!!!!

No typos, got it! 

 

Well, data rebuild for disc 8 is in progress.  . . . We'll know in the morning I guess if it was successful.

 

Either way, you guys have been great.  I'm truly grateful. 

 

 

Pics speak louder than words!!!    A choir of angels sounded when I got up this AM

 

 

Thanks to all who helped!

 

Weird though that the port on disc 5 on the MB failed during this whole fiasco?  Can't explain that . .

 

I haven't had time to check everything on the disc before work today, but those that I did worked fine