D-Link Forums
The Graveyard - Products No Longer Supported => D-Link Storage => DNS-321 => Topic started by: freekfly on October 20, 2009, 05:31:03 PM
-
Hi there,
I have a DNS-321 and am running Firmware 1.03 Hardware Rev: A1 (Lucky me).
It has been running a while now with a Hitachi 1 TB, 7200 rpm, 32MB Cache drive. Loads of data on there and tons of configuration (Users / folders, etc...).
I just purchased a Western Digital 1 TB, 7200 rpm, 32MB Cache drive to create a RAID 1 mirror. I shutdown the device, I add the new drive, I boot up. When I log in I immediately get a message that says:
*****
HARD DRIVE CONFIGURATION :
Your new second hard drive will now be formatted. The existing hard drive will not lose any data.
Vendor Model Serial Number Size
WDC WD10EADS-00M2B0 WD-WCAV51311315 1000 GB
Reconfigure to RAID 1
*****
This is perfect because it is exactly what I want to do. I check off the box that reads Reconfigure to RAID 1 and hit Next. The following windows prompts me to either Enable or Disable Auto Rebuild (I've tried both options) and when I hit Next, I brings me to a window where it starts formatting. It makes it to 5% and pauses for a few seconds before flashing to 100% and then giving the following Error message:
Hard Drive(s) Formatting Failure(Error Code:112).
The only option here is to Restart.
Any ideas? When I first put the new drive in it had 2 formatted partitions with data, I then blew both of those away and left unpartitioned space. That failed so I tried to just create a partition but leave it unformatted. That failed as well, all resulting with the same Error code.
Calling support resulted in me wasting about an hour. I figure this forum will give me better luck.
Also worthy of note, I was running Firmware 1.03b12 prior to upgrading to 1.03. I realize these were basically the same thing, but figured why not.
Thanks,
-Brian
-
I thought I read some where that you "should" make sure that if you dare to use RAID that at a minimum the two drives should be by the same vendor and same size. I know that sounds iffy, and I cannot recall where I read it, but that is all I can think of. DLINK support should be able to tell you exactly what 112 means and possibly how to overcome it.
-
1.03b12 amd 1.03 are indeed the same code, at least according to information posted by D-Link recently.
-
If you're running FFP, rename fun_plug to something like fun_plug.bak. I had the same issue, and that's what was causing it.
If that fixes it, let me know how long it takes your RAID 1 to sync after adding the second drive. I started with 1 WD10EADS with about 400 gigs full, and added a WD10EAVS. It's been rebuilding the array for about 2 weeks with no end in sight...
-
Not sure what FFP is or fun_plug. I don't have any additional apps installed on this DNS, just the basics that came with it. And I've rebuilt MANY RAID arrays in my time and none have ever taken any where close to that long. Usually a couple of hours max. I have a funny feeling that it might be telling you it's rebuilding....but it's either done or it's stuck in a loop.
As for the Same brand disks. I agree that it's better to have the same brand but it theoretically shouldn't make a difference. However it is why I made sure that I had the same speeds / cache on the disks. But it really shouldn't matter.
As for the Error 112....I made two calls to support, the second was escalated and none of them were helpful. I could start *****ing about the support, but I think everyone's called India in their time. They simply just don't have the knowledge / training for this particular product / issue and don't see why it's a big deal to reset the box from scratch. So I have sent an e-mail question with as much detail as I could. We shall see.
Thanks!
-
Odd...I sent an e-mail to DLink support and I have heard back from them yet. It's been a week. Any techs out there have any input on error 112?
Thanks!
-Brian
-
112 [format_ide] invalid file system.
I read your initial post but i am slightly confused. First you said it was a New Drive. Then you said it already had 2 partitions that you blew off of it.
Is it a "New Drive" to the DNS? Or a new drive out of the anti-static bag "new"?
What did you use to take the partitions off of the drive? The error you are receiving leads me to believe there is something on that drive that the DNS-321 is unable to remove/format.
-
Good point. It is a new drive out of the static bag, however I created 2 partitions on it on XP to move some data around. A few hours later I tried to add it to the DNS. The state that it's in now I deleted the partitions all together so it shows in XP as unallocated space. Should I use a disk utility to scrub it some more? Any suggestions? I'd rather avoid reverting my firmware and jumping through those hoops.
Thanks!
-
Run the disk manufacturer's diagnostic and do the full write to zeros diagnostic. It'll be "really" clean then. :)
-
Just another thought here. When you set up the drive in XP, what did you format it as, NTFS or FAT32? It shouldn't make a difference if the 321 is trying to reformat it and then set it up as a RAID 1 config, but I've heard of stranger things happening to prevent a drive from formatting properly. I wonder if it's possible for you to take the new drive and format it externally from the 321, with the same format that the first drive is using, EXT2 or 3? Then reinstall the drive and have it try to create the RAID 1 config that you were trying to do in the first place.
Just a thought, can't hurt at this point in time. Otherwise you might be stuck reverting your firmware to 1.01 to do the job, that worked for another forum member before. As long as you haven't formatted your original drive as EXT3 (firmware 1.01 doesn't deal with EXT3 AFAIK), then that should work, as a last resort. :(
Ryder
-
So I just pulled down the WD disk utility and did a full write to All Zero's...took about 2 1/2 hours. Same result. As for what I formatted it as before it was NTFS even thought I know it uses another file system.
I guess I'm back at square one. I also saw the other post about reverting firmware...might have to go that route.
Thanks for the suggestions.
-
At least you have proven it's not the fact that there's something on the disk that is causing this issue. :)
-
Brian,
There is some information missing here so I will go through some stuff. If you've already done this please say so. And it looks like you are only having a problem with formatting the hard drive. If you don't have a backup of your data I wouldn't place both drives into the NAS together otherwise you're taking a risk.
1) You never said you pulled the original NAS drive out so, Remove your good drive from the NAS and set it to the side.
2) Install your "new" drive into the NAS left drive bay and try to format it. If it works go to step 4.
3) If it fails install the drive into the right drive bay and try to format it. If it fails then buy another drive or use it elsewhere if you can.
4) Assuming the drive formatted without errors you will want to install your original drive only and boot up the NAS. Make sure the drive is recognized without errors.
5) Turn off your NAS and install your "new" formatted drive (risk here).
6) Turn on your NAS and see if you can RAID 1. What you don't want to do is copy backwards.
If everything worked they you have a working RAID 1 NAS.
If the drive fails to format, there could be a problem with the drive. You should run the WD tests on it to verify the drive doesn't have bad sectors before returning it for a replacement. I just wouldn't be suprised if it's bad sectors. I tried to look up that error code under Linux but was not happy with the google results.
-Joe
-
Have you tried formatting the drive as a single volume to see if you get the same error?
-
Have you tried formatting the drive as a single volume to see if you get the same error?
Didn't I just say that?
-
Joe....a little fearful of doing that you're saying, but then again, I guess I can simply back up my config and the drive with data will be out of it. I might try that.
ECF.....Yes, I tried formatting to a single partition in XP as well....I even formatted once as NTSF and then also just blew away the FS and left the un-formatted partition there. And I tried just the unallocated space .....AND the write to 0's, which is where it's at now.
I'll try again Monday...leaving town for the weekend. Thanks!
-
Brian, I guess where you might be left at for options is to revert the firmware to 1.01 and see if that helps, it seemed to work well for some other members of this forum.
But before you do that, the only other thing I would ask is, do you have enough storage space on other drives to empty the original drive and let the 321 format both and turn them into RAID drives? I know it's asking a lot, not many have an extra 1 TB of storage space lying around. I do, so that's why I ask. If you can back up the data and the config file then you should be able to try to get the 321 to format both drives and RAID them, and if that fails, swap them into the opposite slots and try again. The reason I suggest this is that, when I built my 321 RAID 1, I used 2 empty drives and all the 321 did was tell me that it was going to format the left drive and that all data would be lost, and did I want to continue. And since that day my 321 has been working like a charm, no problems, well...except for the auto-power up issue. ;) ;)
If that all fails, then I'd be reverting the firmware to 1.01 and giving that a whirl. At that point, what have you got to lose, with everything backed up and safe. :D
Good luck dude, I sure hope you solve this majorly perplexing puzzle!
Ryder
-
Well, I did it.
Last night I backed up the config. Reverted to Firmware 1.02. Rebooted.
When I did this it completely blew away the config. I shutdown the DNS, added the new drive and turned it on. It then prompted me just like it had done in 1.03 that there is a new drive and I can format it and add it to create a RAID 1. I selected this option and it formatted the drive perfectly and requested a reboot.
When it came back up and I navigated to the RAID section, it said it was busy synchronizing and that took about 3 hours. Once that was complete and it said it had a healthy RAID 1, I upgraded the firmware to 1.03 and restored my config. All of my data was there and the config was fine.
Thanks for all of the input!
-
JoeSchmuck
Are you running the 1TB Hitachi model hard drives?
-
My first drive that had been running for a while was the Hitachi HDS721010KLA330, a 1 TB, 7200 RPM 32 MB Cache hard drive. I then added a second hard drive that was Western Digital: WD10EADS-00M2B0, a 1 TB, 7200 RPM 32 MB Cache Hard drive. I was attempting to format the WD and create a RAID 1 array with that drive.
-
JoeSchmuck
Are you running the 1TB Hitachi model hard drives?
Nope. Running a new Samsung HD103SI in the right drive bay and an old Seagate ST31000340AS. I plan to replace the Seagate drive because it's older and get really hot, of course it's a 7200 RPM drive.
Why do you ask? Did I say something odd? Or is it because freekfly has Hitachi drives? Yea, thought firmware was fixed for the Hitachi drives. Is that what your leaning towards?
-Joe
-
Just seeing if it was specific to a certain drive make and models.
What drive bay did you have your original drive in before adding the second new drive?
-
Just seeing if it was specific to a certain drive make and models.
What drive bay did you have your original drive in before adding the second new drive?
I started with the left drive bay first.
-Joe
-
If you have not downgraded graded the firmware yet could you possibly try your first drive in the right bay, restart the unit then power it down again and insert the new drive into the left bay and boot the unit and see if you receive the same error message or if it allows you to format.
-
I had this same issue, Downgrading to 1.01 solved the formatting errors. Then it was as simple as returning to 1.03.
I had a 500gb WD Caviar drive in the right bay, and I got the error when adding a terabyte WD drive to the left bay.
-
If you're experience this problem, try this before resorting to the firmware revert...
I had the same problem. I ran a single 1TB WD drive for several months. I then purchased a brand new OEM identical 1TB WD drive. It went straight from the static bag to my DNS-321 (firmware 1.3). When powered up the 321 gave the Error Code:112 the OP described.
I was able to get the RAID1 setup without the firmware workaround. Just click "Skip" to ignore the format menu after login. Then go to Tools->RAID and setup the configuration. Setup went without a hitch. For some reason the auto-run format failed but the regular admin menu worked fine.
-
If you're experience this problem, try this before resorting to the firmware revert...
Snip snip... I was able to get the RAID1 setup without the firmware workaround. Just click "Skip" to ignore the format menu after login. Then go to Tools->RAID and setup the configuration. Setup went without a hitch. For some reason the auto-run format failed but the regular admin menu worked fine.
Now that's a great contribution. I'm sure this tid-bit of info should be placed in the FAQ/Sticky Topic somewhere (Hint Hint).
-
Now that's a great contribution. I'm sure this tid-bit of info should be placed in the FAQ/Sticky Topic somewhere (Hint Hint).
!!!Attention!!!
I just tried that and encountered another much more terrible software bug: when your first harddisk was installed in the right bay both disks will be deleted when switching to RAID1! The worst thing is that the webinterface says that no data would be lost. Luckily I made a backup before because I read about this problem in this forum.
Now I am nervous if the RAID really saves my data. I will do a couple of tests before I trust the NAS. Has anyone tried simulating a disk failure yet?
-
!!!Attention!!!
I just tried that and encountered another much more terrible software bug: when your first harddisk was installed in the right bay both disks will be deleted when switching to RAID1! The worst thing is that the webinterface says that no data would be lost. Luckily I made a backup before because I read about this problem in this forum.
Now I am nervous if the RAID really saves my data. I will do a couple of tests before I trust the NAS. Has anyone tried simulating a disk failure yet?
Well, I can't say I "simulated" one, but I did have a disc fail on me. I checked which disc it was, then took out the bad one and replaced it with a good one. I had it set to auto-rebuild and that's exactly what it did. Rebuild the RAID setup without a hitch, all data intact and even available during the rebuild. Worked for me.
Ryder
-
I've never heard of that before, not saying it didn't happen. So, just to be clear, you have a single drive in the right drive bay and used it for some period of time and then you backed it up before installing a second drive of the same capacity as the first drive into the left drive bay. Then you followed the prompts to create a RAID 1 and you original data was deleted.
So, my questions are:
1. What are the models of the two drives (please list them as first/second)
2. What firmware is the NAS running (did you update to 1.03)
3. If my scenario above is not accurate, please describe in detail the events so we all can better understand what occured.
And as for your final comment, of course we all have tried a disk failure and mine works fine, even to the point of installing a blank disk and it re-established the RAID 1 just fine. I think everyone does that because we need to prove this in our minds. Remember, do not pull the drive while power is applied, this NAS is not hot swap capable and you're likely to cause some damage.
-Joe
-
With my post I just wanted to mention that there is a problem when adding a second drive to the NAS. The problem has already been discussed in detail somewhere else in this forum, so initially I just wanted to mention the fact without giving details.
But for preventing someone from losing his data or for for the D-Link team to fix their firmware I am happy to provide details:
In my case I did exactly what Joe described: I used the D-Link 321 firmware 1.03 for some time with a single drive in JBOD mode. The first harddisk is the green WD 1.5TB and was installed in the right bay. Then I purchased a week later the exact same harddisk again and wanted to operate the NAS in RAID1 mode.
The NAS detected the second disk and asked if it should format it and switch to RAID1. Although I knew that this wouldn't work because of another bug in firmware 1.03 I tried it - and it failed. So next time I logged in I skipped this step and went to the tools menu because that was reported to work OK somewhere in this forum.
The instructions in the NAS webinterface say that only the newly added harddisk would be formatted. But in fact both disks were blank after the operation - like a freshly setup RAID1 system. A pretty terrible bug for my taste.
-
Ran into this issue when adding a second drive to DNS-321 and was able to clear it without changing firmware (I'm using ver 1.03).
Existing drive is Seagate 1 TB in left bay facing the unit front (Samba as Volume_1). Adding new WD Caviar Green 1.5 TB to right bay. Seagate is to be the "data" drive, and Caviar will be the "backup" to be removed and stored off site. I do not want to RAID; just wanted additional storage to appear as another share volume. Unwrapped the Caviar and put it in, got this 112 error many times despite reboot.
Scanned Caviar using WD Lifeguard diagnostic s/w and drive was fine (of course). Then got the idea to power down NAS, eject Seagate from left bay. Insert Caviar in right bay, power on. Logged into web interface, and got options to format to ext2, ext3. Selected the default ext2 format, and formatting took about 25 minutes and completed successfully. Now the Caviar was assigned Volume_1.
Powered down, inserted original Seagate to left bay, powered up. Two drives were now available, no data lost on the Seagate. So that bypasses the original error of this thread.
But Seagate was Volume_2. All my shares are setup for that drive to be Volume_1. Swapped bays, and Seagate was still Volume_2. Couldn't figure out how to be certain the Seagate would always be Volume_1, but found workaround.
Used this: http://blog.focusedonlight.com/archive/2009/january/howtorenamevolume
Created new folder "Data" in the Seagate root. Moved all files on Seagate into "Data" folder. Added the "Data" folder as a Samba share. Then on Caviar, created new folder "Backup" and added it as a Samba share. Deleted "Volume_1" and "Volume_2" samba shares in the "Network Access" menu. (no worries, easy to add back on this same menu. No data loss since this is just the Samba share config)
I'm mapping to "Data" share now on all machines, so doesn't matter which drives gets Volume_1 or Volume_2 label. "Backup" share disappears from the list when Caviar is not present. But when Caviar is reinserted and NAS rebooted, "Backup" share reappears. Perfect!
Reminder: can't hot swap...
Some folks hate replies to "stale threads" and this thread may be stale. But this thread was the first hit in Google, so I figure this is the best place to post my findings in order to benefit others.
-
I have the DNS-321 with 2 new 1.5 gig Seagate ST31500341AS. When I tried to format I got a format error code 101. From browsing this thread I downgraded to the 1.02 firmware, formated raid 1, upgraded back to the 1.03 and all is good. Thank you for the advice.